DesignCoder / eval /reports.html
xingxm's picture
Add full-run HTML report (4 models x 200 cases)
a9e2060 verified
Raw History Blame
232 kB
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8"/>
<meta name="viewport" content="width=device-width, initial-scale=1"/>
<title>DesignCoder bench-200</title>
<style>
:root {
--bg:#f6f7f9; --fg:#14181f; --mut:#5c6673; --line:#e2e6ec; --card:#fff;
--accent:#2563eb;
}
* { box-sizing:border-box; }
body { margin:0; background:var(--bg); color:var(--fg);
font:15px/1.6 -apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,"Helvetica Neue",Arial,sans-serif; }
.wrap { max-width:1180px; margin:0 auto; padding:40px 24px 80px; }
header h1 { margin:0 0 6px; font-size:29px; letter-spacing:-.02em; }
header p { margin:0; color:var(--mut); font-size:14px; }
section { margin-top:38px; background:var(--card); border:1px solid var(--line);
border-radius:12px; padding:22px 24px; }
h2 { margin:0 0 4px; font-size:19px; }
h3 { margin:0; font-size:15px; }
h4 { margin:18px 0 8px; font-size:14px; color:var(--mut); text-transform:uppercase;
letter-spacing:.06em; }
.lead { color:var(--mut); font-size:13.5px; margin:0 0 18px; }
.cards { display:grid; gap:14px; grid-template-columns:repeat(auto-fit,minmax(232px,1fr)); }
.card { border:1px solid var(--line); border-radius:10px; padding:16px; background:#fff; }
.card.top { border-color:var(--accent); box-shadow:0 0 0 3px rgba(37,99,235,.09); }
.ch { display:flex; align-items:center; justify-content:space-between; gap:8px; }
.crown { background:var(--accent); color:#fff; font-size:10.5px; font-weight:700;
padding:2px 7px; border-radius:20px; text-transform:uppercase; letter-spacing:.05em; }
.sub { color:var(--mut); font-size:12px; margin:3px 0 10px; }
.big { font-size:36px; font-weight:750; letter-spacing:-.03em; margin-bottom:8px; }
.kv { display:flex; justify-content:space-between; font-size:12.5px;
padding:3px 0; border-top:1px dashed var(--line); }
.kv span { color:var(--mut); }
table { border-collapse:collapse; width:100%; font-size:13px; }
th,td { padding:7px 9px; text-align:right; border-bottom:1px solid var(--line);
white-space:nowrap; }
th { background:#f0f2f5; font-weight:650; font-size:12px; text-align:right;
position:sticky; top:0; }
th:first-child, td:first-child { text-align:left; }
td.cid { text-align:left; font-family:ui-monospace,SFMono-Regular,Menlo,monospace;
font-size:11.5px; }
td.tag { text-align:left; color:var(--mut); font-size:12px; }
td.sm { text-align:left; color:var(--mut); font-size:11.5px; white-space:normal;
max-width:420px; }
td.na { color:#b6bcc6; }
td.heat { color:#fff; font-weight:650; text-shadow:0 1px 1px rgba(0,0,0,.25); }
td.best { background:#e7f0ff; font-weight:700; color:#12326d; }
.scroll { overflow:auto; max-height:560px; border:1px solid var(--line);
border-radius:8px; }
.chart { display:block; margin:6px 0 2px; }
.grid { stroke:#e6e9ee; stroke-width:1; }
text { fill:#3d4652; font-size:11.5px; }
text.bl { font-size:12px; }
text.bv { font-weight:700; font-size:12px; }
text.bs { font-size:10px; fill:#5c6673; }
text.tk { fill:#98a1ad; font-size:10.5px; }
.legend { display:flex; flex-wrap:wrap; gap:14px; margin-top:10px; }
.lg { display:flex; align-items:center; gap:6px; font-size:12px; color:var(--mut); }
.lg i { width:11px; height:11px; border-radius:3px; display:inline-block; }
.note { color:var(--mut); font-size:12px; margin:10px 0 0; }
.filters { display:flex; gap:8px; margin-bottom:12px; flex-wrap:wrap; }
.f { border:1px solid var(--line); background:#fff; border-radius:20px;
padding:5px 13px; font-size:12.5px; cursor:pointer; color:var(--mut); }
.f.on { background:var(--accent); color:#fff; border-color:var(--accent); }
.cols2 { display:grid; gap:22px; grid-template-columns:1fr 1fr; }
@media (max-width:860px) { .cols2 { grid-template-columns:1fr; } }
code { background:#eef1f5; padding:1px 5px; border-radius:4px; font-size:12px; }
</style>
</head>
<body>
<div class="wrap">
<header>
<h1>DesignCoder bench-200</h1>
<p>Judge: gpt-5.6-sol (vision, reasoning_effort=high) · 200 frozen cases (Track A landing 100 / A dashboard 40 /
B landing 30 / B dashboard 30) · every rubric item is a binary screenshot check.</p>
</header>
<section>
<h2>Model comparison</h2>
<p class="lead">Overall score is the unweighted mean of Prompt Fit and the six rubric
dimensions, following DesignEvaluator V5.9.2.</p>
<div class="cards"><div class="card top"><div class="ch"><h3>27B AdamW step400</h3><span class="crown">best</span></div><div class="sub">27B · AdamW · 400</div><div class="big">87.89</div><div class="kv"><span>median</span><b>88.93</b></div><div class="kv"><span>p25</span><b>82.86</b></div><div class="kv"><span>min</span><b>46.43</b></div><div class="kv"><span>prompt fit</span><b>84.10</b></div><div class="kv"><span>rubric pass</span><b>88.60%</b></div><div class="kv"><span>cases &lt; 60</span><b>2</b></div><div class="kv"><span>render fail</span><b>0</b></div><div class="kv"><span>n</span><b>200</b></div></div><div class="card"><div class="ch"><h3>9B AdamW step200</h3></div><div class="sub">9B · AdamW · 200</div><div class="big">84.40</div><div class="kv"><span>median</span><b>85.59</b></div><div class="kv"><span>p25</span><b>79.29</b></div><div class="kv"><span>min</span><b>38.57</b></div><div class="kv"><span>prompt fit</span><b>78.35</b></div><div class="kv"><span>rubric pass</span><b>85.37%</b></div><div class="kv"><span>cases &lt; 60</span><b>6</b></div><div class="kv"><span>render fail</span><b>1</b></div><div class="kv"><span>n</span><b>200</b></div></div><div class="card"><div class="ch"><h3>4B AdamW step200</h3></div><div class="sub">4B · AdamW · 200</div><div class="big">82.12</div><div class="kv"><span>median</span><b>84.41</b></div><div class="kv"><span>p25</span><b>74.05</b></div><div class="kv"><span>min</span><b>38.33</b></div><div class="kv"><span>prompt fit</span><b>75.25</b></div><div class="kv"><span>rubric pass</span><b>83.24%</b></div><div class="kv"><span>cases &lt; 60</span><b>14</b></div><div class="kv"><span>render fail</span><b>1</b></div><div class="kv"><span>n</span><b>200</b></div></div><div class="card"><div class="ch"><h3>4B Muon step200</h3></div><div class="sub">4B · Muon · 200</div><div class="big">77.62</div><div class="kv"><span>median</span><b>80.83</b></div><div class="kv"><span>p25</span><b>70.75</b></div><div class="kv"><span>min</span><b>35.08</b></div><div class="kv"><span>prompt fit</span><b>64.60</b></div><div class="kv"><span>rubric pass</span><b>79.43%</b></div><div class="kv"><span>cases &lt; 60</span><b>22</b></div><div class="kv"><span>render fail</span><b>3</b></div><div class="kv"><span>n</span><b>200</b></div></div></div>
</section>
<section>
<h2>Overall score</h2>
<svg viewBox="0 0 720 196" width="100%" height="196" role="img" class="chart"><text x="250" y="35.120000000000005" text-anchor="end" class="bl">27B AdamW step400</text><rect x="260" y="12" width="342.8" height="34" rx="5" fill="#2563eb"/><text x="610.8" y="35.120000000000005" class="bv">87.89</text><text x="250" y="81.12" text-anchor="end" class="bl">9B AdamW step200</text><rect x="260" y="58" width="329.2" height="34" rx="5" fill="#0d9488"/><text x="597.2" y="81.12" class="bv">84.40</text><text x="250" y="127.12" text-anchor="end" class="bl">4B AdamW step200</text><rect x="260" y="104" width="320.3" height="34" rx="5" fill="#d97706"/><text x="588.3" y="127.12" class="bv">82.12</text><text x="250" y="173.12" text-anchor="end" class="bl">4B Muon step200</text><rect x="260" y="150" width="302.7" height="34" rx="5" fill="#7c3aed"/><text x="570.7" y="173.12" class="bv">77.62</text><line x1="357.5" y1="8" x2="357.5" y2="186" class="grid"/><text x="357.5" y="195" text-anchor="middle" class="tk">25</text><line x1="455.0" y1="8" x2="455.0" y2="186" class="grid"/><text x="455.0" y="195" text-anchor="middle" class="tk">50</text><line x1="552.5" y1="8" x2="552.5" y2="186" class="grid"/><text x="552.5" y="195" text-anchor="middle" class="tk">75</text><line x1="650.0" y1="8" x2="650.0" y2="186" class="grid"/><text x="650.0" y="195" text-anchor="middle" class="tk">100</text></svg>
</section>
<section><h2>Checkpoints under test</h2><div class="scroll"><table class="grid"><thead><tr><th>model</th><th>base</th><th>optimizer</th><th>global batch</th><th>step</th><th>HF subfolder</th></tr></thead><tbody><tr><td class='cid'>27B AdamW step400</td><td>Qwen3.8-27B</td><td>AdamW</td><td>128</td><td>400</td><td class='sm'>designcoder_qwen3.8_27b_adamw_bs128_data41287_step400</td></tr><tr><td class='cid'>9B AdamW step200</td><td>Qwen3.5-9B</td><td>AdamW</td><td>256</td><td>200</td><td class='sm'>designcoder_qwen3.5_9b_adamw_bs256_data41287_step200</td></tr><tr><td class='cid'>4B AdamW step200</td><td>Qwen3.5-4B</td><td>AdamW</td><td>256</td><td>200</td><td class='sm'>designcoder_qwen3.5_4b_adamw_bs256_data41287_step200</td></tr><tr><td class='cid'>4B Muon step200</td><td>Qwen3.5-4B</td><td>Muon</td><td>256</td><td>200</td><td class='sm'>designcoder_qwen3.5_4b_muon_bs256_data41287_step200</td></tr></tbody></table></div></section>
<section>
<h2>Statistics</h2>
<div class="scroll"><table class="grid">
<thead><tr><th>model</th><th>n</th><th>mean</th><th>median</th><th>p25</th><th>min</th><th>max</th><th>stdev</th><th>prompt fit</th><th>rubric %</th><th>&lt;60</th><th>render fail</th></tr></thead><tbody><tr><td class='cid'>27B AdamW step400</td><td>200</td><td><b>87.89</b></td><td>88.93</td><td>82.86</td><td>46.43</td><td>100.00</td><td>9.52</td><td>84.10</td><td>88.60</td><td>2</td><td>0</td></tr><tr><td class='cid'>9B AdamW step200</td><td>200</td><td><b>84.40</b></td><td>85.59</td><td>79.29</td><td>38.57</td><td>100.00</td><td>10.43</td><td>78.35</td><td>85.37</td><td>6</td><td>1</td></tr><tr><td class='cid'>4B AdamW step200</td><td>200</td><td><b>82.12</b></td><td>84.41</td><td>74.05</td><td>38.33</td><td>100.00</td><td>12.46</td><td>75.25</td><td>83.24</td><td>14</td><td>1</td></tr><tr><td class='cid'>4B Muon step200</td><td>200</td><td><b>77.62</b></td><td>80.83</td><td>70.75</td><td>35.08</td><td>100.00</td><td>13.73</td><td>64.60</td><td>79.43</td><td>22</td><td>3</td></tr></tbody>
</table></div>
<p class="note">p25 = 25th percentile; a high mean with a low p25 means the model is
inconsistent rather than uniformly good.</p>
</section>
<section>
<h2>Track x Surface</h2>
<p class="lead">The benchmark requires Track A/B and landing/dashboard to be reported
separately.</p>
<svg viewBox="0 0 860 320" width="100%" height="320" role="img" class="chart"><line x1="58" y1="262.0" x2="844" y2="262.0" class="grid"/><text x="50" y="266.0" text-anchor="end" class="tk">0</text><line x1="58" y1="213.2" x2="844" y2="213.2" class="grid"/><text x="50" y="217.2" text-anchor="end" class="tk">20</text><line x1="58" y1="164.4" x2="844" y2="164.4" class="grid"/><text x="50" y="168.4" text-anchor="end" class="tk">40</text><line x1="58" y1="115.6" x2="844" y2="115.6" class="grid"/><text x="50" y="119.6" text-anchor="end" class="tk">60</text><line x1="58" y1="66.8" x2="844" y2="66.8" class="grid"/><text x="50" y="70.8" text-anchor="end" class="tk">80</text><line x1="58" y1="18.0" x2="844" y2="18.0" class="grid"/><text x="50" y="22.0" text-anchor="end" class="tk">100</text><rect x="77.7" y="47.7" width="35.0" height="214.3" rx="3" fill="#2563eb"><title>27B AdamW step400 · A/landing: 87.84</title></rect><text x="95.2" y="43.7" text-anchor="middle" class="bs">87.8</text><rect x="115.7" y="52.4" width="35.0" height="209.6" rx="3" fill="#0d9488"><title>9B AdamW step200 · A/landing: 85.92</title></rect><text x="133.2" y="48.4" text-anchor="middle" class="bs">85.9</text><rect x="153.7" y="55.3" width="35.0" height="206.7" rx="3" fill="#d97706"><title>4B AdamW step200 · A/landing: 84.70</title></rect><text x="171.2" y="51.3" text-anchor="middle" class="bs">84.7</text><rect x="191.7" y="64.1" width="35.0" height="197.9" rx="3" fill="#7c3aed"><title>4B Muon step200 · A/landing: 81.09</title></rect><text x="209.2" y="60.1" text-anchor="middle" class="bs">81.1</text><text x="156.2" y="290" text-anchor="middle" class="bl">A/landing</text><rect x="274.1" y="55.7" width="35.0" height="206.3" rx="3" fill="#2563eb"><title>27B AdamW step400 · A/dashboard: 84.56</title></rect><text x="291.6" y="51.7" text-anchor="middle" class="bs">84.6</text><rect x="312.1" y="67.9" width="35.0" height="194.1" rx="3" fill="#0d9488"><title>9B AdamW step200 · A/dashboard: 79.56</title></rect><text x="329.6" y="63.9" text-anchor="middle" class="bs">79.6</text><rect x="350.1" y="75.1" width="35.0" height="186.9" rx="3" fill="#d97706"><title>4B AdamW step200 · A/dashboard: 76.61</title></rect><text x="367.6" y="71.1" text-anchor="middle" class="bs">76.6</text><rect x="388.1" y="93.6" width="35.0" height="168.4" rx="3" fill="#7c3aed"><title>4B Muon step200 · A/dashboard: 69.01</title></rect><text x="405.6" y="89.6" text-anchor="middle" class="bs">69.0</text><text x="352.8" y="290" text-anchor="middle" class="bl">A/dashboard</text><rect x="470.6" y="40.2" width="35.0" height="221.8" rx="3" fill="#2563eb"><title>27B AdamW step400 · B/landing: 90.92</title></rect><text x="488.1" y="36.2" text-anchor="middle" class="bs">90.9</text><rect x="508.6" y="48.2" width="35.0" height="213.8" rx="3" fill="#0d9488"><title>9B AdamW step200 · B/landing: 87.61</title></rect><text x="526.1" y="44.2" text-anchor="middle" class="bs">87.6</text><rect x="546.6" y="51.4" width="35.0" height="210.6" rx="3" fill="#d97706"><title>4B AdamW step200 · B/landing: 86.30</title></rect><text x="564.1" y="47.4" text-anchor="middle" class="bs">86.3</text><rect x="584.6" y="58.3" width="35.0" height="203.7" rx="3" fill="#7c3aed"><title>4B Muon step200 · B/landing: 83.50</title></rect><text x="602.1" y="54.3" text-anchor="middle" class="bs">83.5</text><text x="549.2" y="290" text-anchor="middle" class="bl">B/landing</text><rect x="667.1" y="43.7" width="35.0" height="218.3" rx="3" fill="#2563eb"><title>27B AdamW step400 · B/dashboard: 89.48</title></rect><text x="684.6" y="39.7" text-anchor="middle" class="bs">89.5</text><rect x="705.1" y="60.5" width="35.0" height="201.5" rx="3" fill="#0d9488"><title>9B AdamW step200 · B/dashboard: 82.57</title></rect><text x="722.6" y="56.5" text-anchor="middle" class="bs">82.6</text><rect x="743.1" y="74.9" width="35.0" height="187.1" rx="3" fill="#d97706"><title>4B AdamW step200 · B/dashboard: 76.69</title></rect><text x="760.6" y="70.9" text-anchor="middle" class="bs">76.7</text><rect x="781.1" y="87.2" width="35.0" height="174.8" rx="3" fill="#7c3aed"><title>4B Muon step200 · B/dashboard: 71.65</title></rect><text x="798.6" y="83.2" text-anchor="middle" class="bs">71.7</text><text x="745.8" y="290" text-anchor="middle" class="bl">B/dashboard</text></svg><div class="legend"><span class="lg"><i style="background:#2563eb"></i>27B AdamW step400</span><span class="lg"><i style="background:#0d9488"></i>9B AdamW step200</span><span class="lg"><i style="background:#d97706"></i>4B AdamW step200</span><span class="lg"><i style="background:#7c3aed"></i>4B Muon step200</span></div>
<h4>By surface</h4>
<svg viewBox="0 0 860 320" width="100%" height="320" role="img" class="chart"><line x1="58" y1="262.0" x2="844" y2="262.0" class="grid"/><text x="50" y="266.0" text-anchor="end" class="tk">0</text><line x1="58" y1="213.2" x2="844" y2="213.2" class="grid"/><text x="50" y="217.2" text-anchor="end" class="tk">20</text><line x1="58" y1="164.4" x2="844" y2="164.4" class="grid"/><text x="50" y="168.4" text-anchor="end" class="tk">40</text><line x1="58" y1="115.6" x2="844" y2="115.6" class="grid"/><text x="50" y="119.6" text-anchor="end" class="tk">60</text><line x1="58" y1="66.8" x2="844" y2="66.8" class="grid"/><text x="50" y="70.8" text-anchor="end" class="tk">80</text><line x1="58" y1="18.0" x2="844" y2="18.0" class="grid"/><text x="50" y="22.0" text-anchor="end" class="tk">100</text><rect x="97.3" y="45.9" width="35.0" height="216.1" rx="3" fill="#2563eb"><title>27B AdamW step400 · landing: 88.55</title></rect><text x="114.8" y="41.9" text-anchor="middle" class="bs">88.6</text><rect x="135.3" y="51.4" width="35.0" height="210.6" rx="3" fill="#0d9488"><title>9B AdamW step200 · landing: 86.31</title></rect><text x="152.8" y="47.4" text-anchor="middle" class="bs">86.3</text><rect x="173.3" y="54.4" width="35.0" height="207.6" rx="3" fill="#d97706"><title>4B AdamW step200 · landing: 85.07</title></rect><text x="190.8" y="50.4" text-anchor="middle" class="bs">85.1</text><rect x="211.3" y="62.8" width="35.0" height="199.2" rx="3" fill="#7c3aed"><title>4B Muon step200 · landing: 81.65</title></rect><text x="228.8" y="58.8" text-anchor="middle" class="bs">81.6</text><text x="254.5" y="290" text-anchor="middle" class="bl">landing</text><rect x="490.3" y="50.5" width="35.0" height="211.5" rx="3" fill="#2563eb"><title>27B AdamW step400 · dashboard: 86.67</title></rect><text x="507.8" y="46.5" text-anchor="middle" class="bs">86.7</text><rect x="528.3" y="64.7" width="35.0" height="197.3" rx="3" fill="#0d9488"><title>9B AdamW step200 · dashboard: 80.85</title></rect><text x="545.8" y="60.7" text-anchor="middle" class="bs">80.8</text><rect x="566.3" y="75.0" width="35.0" height="187.0" rx="3" fill="#d97706"><title>4B AdamW step200 · dashboard: 76.65</title></rect><text x="583.8" y="71.0" text-anchor="middle" class="bs">76.6</text><rect x="604.3" y="90.9" width="35.0" height="171.1" rx="3" fill="#7c3aed"><title>4B Muon step200 · dashboard: 70.14</title></rect><text x="621.8" y="86.9" text-anchor="middle" class="bs">70.1</text><text x="647.5" y="290" text-anchor="middle" class="bl">dashboard</text></svg><div class="legend"><span class="lg"><i style="background:#2563eb"></i>27B AdamW step400</span><span class="lg"><i style="background:#0d9488"></i>9B AdamW step200</span><span class="lg"><i style="background:#d97706"></i>4B AdamW step200</span><span class="lg"><i style="background:#7c3aed"></i>4B Muon step200</span></div>
</section>
<section>
<h2>Rubric dimensions</h2>
<div class="scroll"><table class="grid"><thead><tr><th>model</th><th>surface</th><th>Alignment</th><th>Layout</th><th>Typography</th><th>Components</th><th>Assets</th><th>Aesthetics</th></tr></thead><tbody><tr><td>27B AdamW step400</td><td class='tag'>landing</td><td class="heat" style="background:rgb(105,175,89)">89.6</td><td class="heat" style="background:rgb(63,164,100)">94.7</td><td class="heat" style="background:rgb(102,174,90)">90.1</td><td class="heat" style="background:rgb(88,171,93)">91.7</td><td class="heat" style="background:rgb(220,119,60)">63.6</td><td class="heat" style="background:rgb(47,160,104)">96.6</td></tr><tr><td>27B AdamW step400</td><td class='tag'>dashboard</td><td class="heat" style="background:rgb(102,174,90)">90.1</td><td class="heat" style="background:rgb(135,183,81)">86.1</td><td class="heat" style="background:rgb(75,167,97)">93.3</td><td class="heat" style="background:rgb(126,180,84)">87.2</td><td class="heat" style="background:rgb(207,201,63)">77.6</td><td class="heat" style="background:rgb(56,163,101)">95.5</td></tr><tr><td>9B AdamW step200</td><td class='tag'>landing</td><td class="heat" style="background:rgb(135,183,81)">86.1</td><td class="heat" style="background:rgb(99,174,90)">90.4</td><td class="heat" style="background:rgb(102,174,90)">90.1</td><td class="heat" style="background:rgb(114,177,87)">88.6</td><td class="heat" style="background:rgb(220,112,60)">62.7</td><td class="heat" style="background:rgb(45,160,105)">96.9</td></tr><tr><td>9B AdamW step200</td><td class='tag'>dashboard</td><td class="heat" style="background:rgb(111,177,87)">89.0</td><td class="heat" style="background:rgb(171,192,72)">81.9</td><td class="heat" style="background:rgb(107,176,88)">89.5</td><td class="heat" style="background:rgb(174,193,71)">81.4</td><td class="heat" style="background:rgb(220,116,60)">63.3</td><td class="heat" style="background:rgb(98,173,91)">90.5</td></tr><tr><td>4B AdamW step200</td><td class='tag'>landing</td><td class="heat" style="background:rgb(149,186,78)">84.5</td><td class="heat" style="background:rgb(87,170,94)">91.8</td><td class="heat" style="background:rgb(127,181,83)">87.1</td><td class="heat" style="background:rgb(132,182,82)">86.4</td><td class="heat" style="background:rgb(220,114,60)">62.9</td><td class="heat" style="background:rgb(63,164,100)">94.7</td></tr><tr><td>4B AdamW step200</td><td class='tag'>dashboard</td><td class="heat" style="background:rgb(168,191,73)">82.2</td><td class="heat" style="background:rgb(188,197,67)">79.7</td><td class="heat" style="background:rgb(131,182,82)">86.6</td><td class="heat" style="background:rgb(214,203,61)">76.7</td><td class="heat" style="background:rgb(220,97,60)">60.5</td><td class="heat" style="background:rgb(107,175,89)">89.5</td></tr><tr><td>4B Muon step200</td><td class='tag'>landing</td><td class="heat" style="background:rgb(172,192,72)">81.6</td><td class="heat" style="background:rgb(101,174,90)">90.2</td><td class="heat" style="background:rgb(137,183,81)">85.9</td><td class="heat" style="background:rgb(161,190,74)">82.9</td><td class="heat" style="background:rgb(220,63,60)">55.6</td><td class="heat" style="background:rgb(80,169,95)">92.6</td></tr><tr><td>4B Muon step200</td><td class='tag'>dashboard</td><td class="heat" style="background:rgb(194,198,66)">79.1</td><td class="heat" style="background:rgb(220,171,60)">71.2</td><td class="heat" style="background:rgb(143,185,79)">85.2</td><td class="heat" style="background:rgb(220,172,60)">71.4</td><td class="heat" style="background:rgb(220,60,60)">55.1</td><td class="heat" style="background:rgb(161,189,75)">83.0</td></tr></tbody></table></div><p class="note">Rubric pass rate (%). Scale spans 55.1–96.9; red = weakest, green = strongest.</p>
</section>
<section>
<h2>Prompt Fit detail</h2>
<p class="lead">Mean of the five surface-specific rubrics, each scored 0-2. This is where
product completeness shows up, as opposed to pure visual quality.</p>
<div class="cols2"><h4>landing</h4><div class="scroll"><table class="grid"><thead><tr><th>rubric</th><th>27B AdamW step400</th><th>9B AdamW step200</th><th>4B AdamW step200</th><th>4B Muon step200</th></tr></thead><tbody><tr><td class='cid'>core_requirement_priority</td><td class="heat" style="background:rgb(103,175,90)">1.80</td><td class="heat" style="background:rgb(132,182,82)">1.70</td><td class="heat" style="background:rgb(130,181,83)">1.71</td><td class="heat" style="background:rgb(193,198,66)">1.49</td></tr><tr><td class='cid'>product_completeness</td><td class="heat" style="background:rgb(208,202,62)">1.44</td><td class="heat" style="background:rgb(220,203,60)">1.39</td><td class="heat" style="background:rgb(220,191,60)">1.35</td><td class="heat" style="background:rgb(220,158,60)">1.21</td></tr><tr><td class='cid'>decision_support</td><td class="heat" style="background:rgb(62,164,100)">1.94</td><td class="heat" style="background:rgb(78,168,96)">1.88</td><td class="heat" style="background:rgb(87,170,94)">1.85</td><td class="heat" style="background:rgb(143,185,79)">1.66</td></tr><tr><td class='cid'>workflow_closure</td><td class="heat" style="background:rgb(69,166,98)">1.92</td><td class="heat" style="background:rgb(74,167,97)">1.90</td><td class="heat" style="background:rgb(74,167,97)">1.90</td><td class="heat" style="background:rgb(107,176,88)">1.78</td></tr><tr><td class='cid'>style_fidelity</td><td class="heat" style="background:rgb(141,184,80)">1.67</td><td class="heat" style="background:rgb(148,186,78)">1.65</td><td class="heat" style="background:rgb(159,189,75)">1.61</td><td class="heat" style="background:rgb(204,200,64)">1.45</td></tr></tbody></table></div><h4>dashboard</h4><div class="scroll"><table class="grid"><thead><tr><th>rubric</th><th>27B AdamW step400</th><th>9B AdamW step200</th><th>4B AdamW step200</th><th>4B Muon step200</th></tr></thead><tbody><tr><td class='cid'>operator_goal_priority</td><td class="heat" style="background:rgb(90,171,93)">1.84</td><td class="heat" style="background:rgb(157,188,76)">1.61</td><td class="heat" style="background:rgb(190,197,67)">1.50</td><td class="heat" style="background:rgb(220,139,60)">1.13</td></tr><tr><td class='cid'>decision_readiness</td><td class="heat" style="background:rgb(157,188,76)">1.61</td><td class="heat" style="background:rgb(220,163,60)">1.23</td><td class="heat" style="background:rgb(220,104,60)">0.99</td><td class="heat" style="background:rgb(220,60,60)">0.63</td></tr><tr><td class='cid'>operational_coverage</td><td class="heat" style="background:rgb(211,202,62)">1.43</td><td class="heat" style="background:rgb(220,129,60)">1.09</td><td class="heat" style="background:rgb(220,91,60)">0.93</td><td class="heat" style="background:rgb(220,60,60)">0.70</td></tr><tr><td class='cid'>action_state_closure</td><td class="heat" style="background:rgb(220,153,60)">1.19</td><td class="heat" style="background:rgb(220,142,60)">1.14</td><td class="heat" style="background:rgb(220,101,60)">0.97</td><td class="heat" style="background:rgb(220,60,60)">0.74</td></tr><tr><td class='cid'>design_dna_fidelity</td><td class="heat" style="background:rgb(136,183,81)">1.69</td><td class="heat" style="background:rgb(194,198,66)">1.49</td><td class="heat" style="background:rgb(194,198,66)">1.49</td><td class="heat" style="background:rgb(220,142,60)">1.14</td></tr></tbody></table></div></div>
</section>
<section>
<h2>Score distribution</h2>
<p class="lead">Share of cases (%) falling in each score band.</p>
<svg viewBox="0 0 860 320" width="100%" height="320" role="img" class="chart"><line x1="58" y1="262.0" x2="844" y2="262.0" class="grid"/><text x="50" y="266.0" text-anchor="end" class="tk">0</text><line x1="58" y1="213.2" x2="844" y2="213.2" class="grid"/><text x="50" y="217.2" text-anchor="end" class="tk">20</text><line x1="58" y1="164.4" x2="844" y2="164.4" class="grid"/><text x="50" y="168.4" text-anchor="end" class="tk">40</text><line x1="58" y1="115.6" x2="844" y2="115.6" class="grid"/><text x="50" y="119.6" text-anchor="end" class="tk">60</text><line x1="58" y1="66.8" x2="844" y2="66.8" class="grid"/><text x="50" y="70.8" text-anchor="end" class="tk">80</text><line x1="58" y1="18.0" x2="844" y2="18.0" class="grid"/><text x="50" y="22.0" text-anchor="end" class="tk">100</text><rect x="69.2" y="260.8" width="20.4" height="1.2" rx="3" fill="#2563eb"><title>27B AdamW step400 · &lt;50: 0.50</title></rect><rect x="92.6" y="259.6" width="20.4" height="2.4" rx="3" fill="#0d9488"><title>9B AdamW step200 · &lt;50: 1.00</title></rect><rect x="116.0" y="259.6" width="20.4" height="2.4" rx="3" fill="#d97706"><title>4B AdamW step200 · &lt;50: 1.00</title></rect><rect x="139.4" y="246.1" width="20.4" height="15.9" rx="3" fill="#7c3aed"><title>4B Muon step200 · &lt;50: 6.50</title></rect><text x="114.1" y="290" text-anchor="middle" class="bl">&lt;50</text><rect x="181.5" y="260.8" width="20.4" height="1.2" rx="3" fill="#2563eb"><title>27B AdamW step400 · 50-60: 0.50</title></rect><rect x="204.9" y="257.1" width="20.4" height="4.9" rx="3" fill="#0d9488"><title>9B AdamW step200 · 50-60: 2.00</title></rect><rect x="228.3" y="247.4" width="20.4" height="14.6" rx="3" fill="#d97706"><title>4B AdamW step200 · 50-60: 6.00</title></rect><rect x="251.7" y="251.0" width="20.4" height="11.0" rx="3" fill="#7c3aed"><title>4B Muon step200 · 50-60: 4.50</title></rect><text x="226.4" y="290" text-anchor="middle" class="bl">50-60</text><rect x="293.8" y="251.0" width="20.4" height="11.0" rx="3" fill="#2563eb"><title>27B AdamW step400 · 60-70: 4.50</title></rect><rect x="317.2" y="249.8" width="20.4" height="12.2" rx="3" fill="#0d9488"><title>9B AdamW step200 · 60-70: 5.00</title></rect><rect x="340.6" y="236.4" width="20.4" height="25.6" rx="3" fill="#d97706"><title>4B AdamW step200 · 60-70: 10.50</title></rect><rect x="364.0" y="235.2" width="20.4" height="26.8" rx="3" fill="#7c3aed"><title>4B Muon step200 · 60-70: 11.00</title></rect><text x="338.7" y="290" text-anchor="middle" class="bl">60-70</text><rect x="406.1" y="232.7" width="20.4" height="29.3" rx="3" fill="#2563eb"><title>27B AdamW step400 · 70-80: 12.00</title></rect><rect x="429.5" y="215.6" width="20.4" height="46.4" rx="3" fill="#0d9488"><title>9B AdamW step200 · 70-80: 19.00</title></rect><rect x="452.9" y="215.6" width="20.4" height="46.4" rx="3" fill="#d97706"><title>4B AdamW step200 · 70-80: 19.00</title></rect><rect x="476.3" y="198.6" width="20.4" height="63.4" rx="3" fill="#7c3aed"><title>4B Muon step200 · 70-80: 26.00</title></rect><text x="451.0" y="290" text-anchor="middle" class="bl">70-80</text><rect x="518.4" y="175.4" width="20.4" height="86.6" rx="3" fill="#2563eb"><title>27B AdamW step400 · 80-90: 35.50</title></rect><rect x="541.8" y="170.5" width="20.4" height="91.5" rx="3" fill="#0d9488"><title>9B AdamW step200 · 80-90: 37.50</title></rect><rect x="565.2" y="187.6" width="20.4" height="74.4" rx="3" fill="#d97706"><title>4B AdamW step200 · 80-90: 30.50</title></rect><rect x="588.6" y="179.0" width="20.4" height="83.0" rx="3" fill="#7c3aed"><title>4B Muon step200 · 80-90: 34.00</title></rect><text x="563.3" y="290" text-anchor="middle" class="bl">80-90</text><rect x="630.7" y="214.4" width="20.4" height="47.6" rx="3" fill="#2563eb"><title>27B AdamW step400 · 90-95: 19.50</title></rect><rect x="654.0" y="208.3" width="20.4" height="53.7" rx="3" fill="#0d9488"><title>9B AdamW step200 · 90-95: 22.00</title></rect><rect x="677.4" y="216.9" width="20.4" height="45.1" rx="3" fill="#d97706"><title>4B AdamW step200 · 90-95: 18.50</title></rect><rect x="700.8" y="233.9" width="20.4" height="28.1" rx="3" fill="#7c3aed"><title>4B Muon step200 · 90-95: 11.50</title></rect><text x="675.6" y="290" text-anchor="middle" class="bl">90-95</text><rect x="742.9" y="194.9" width="20.4" height="67.1" rx="3" fill="#2563eb"><title>27B AdamW step400 · 95-100: 27.50</title></rect><rect x="766.3" y="229.1" width="20.4" height="32.9" rx="3" fill="#0d9488"><title>9B AdamW step200 · 95-100: 13.50</title></rect><rect x="789.7" y="226.6" width="20.4" height="35.4" rx="3" fill="#d97706"><title>4B AdamW step200 · 95-100: 14.50</title></rect><rect x="813.1" y="246.1" width="20.4" height="15.9" rx="3" fill="#7c3aed"><title>4B Muon step200 · 95-100: 6.50</title></rect><text x="787.9" y="290" text-anchor="middle" class="bl">95-100</text></svg><div class="legend"><span class="lg"><i style="background:#2563eb"></i>27B AdamW step400</span><span class="lg"><i style="background:#0d9488"></i>9B AdamW step200</span><span class="lg"><i style="background:#d97706"></i>4B AdamW step200</span><span class="lg"><i style="background:#7c3aed"></i>4B Muon step200</span></div>
</section>
<section>
<h2>Weakest cases — 27B AdamW step400</h2>
<div class="scroll"><table class="grid">
<thead><tr><th>score</th><th>group</th><th>case_id</th><th>judge summary</th></tr></thead>
<tbody><tr><td>46.43</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-045-streaming-dashboard-obs</td><td class='sm'>A polished dark console shell is visible, but most core storm-tracking visualizations and table content are blank.</td></tr><tr><td>50.95</td><td class='tag'>A/landing</td><td class='cid'>landing-027-art-deco</td><td class='sm'>A polished Art Deco landing page with strong atmosphere and booking emphasis is materially weakened by the absent timetable, non-symmetrical hero imagery, and incomplete offering grids.</td></tr><tr><td>61.55</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-002-armin-hofmann-contrast-and-reduction</td><td class='sm'>The dashboard has a strong monochrome structural shell, but its empty charts, ticket table, station rail, and inspector make the operational experience incomplete.</td></tr><tr><td>66.67</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-005-dieter-rams-silent-control-panel</td><td class='sm'>A polished and highly legible silent-control dashboard whose default overview is strong, but the screenshot does not demonstrate the selected-lift inspector, capacity consequence, webcam, maintenance, loading, or active </td></tr><tr><td>68.10</td><td class='tag'>A/landing</td><td class='cid'>landing-015-biomorphic-design</td><td class='sm'>The page is polished and atmospherically cohesive with complete downstream sections and reservation flow, but the missing hero selector, absent table-lamp clarity, and occluded compact-card copy weaken fidelity and usabi</td></tr><tr><td>68.10</td><td class='tag'>A/landing</td><td class='cid'>landing-041-stop-motion-craft</td><td class='sm'>A polished, cohesive pottery-drop landing page communicates the collection and reservation flow well, but several specific handmade and overhead-photography requirements are visibly absent.</td></tr><tr><td>68.88</td><td class='tag'>B/dashboard</td><td class='cid'>track-b-dashboard-016-src-393</td><td class='sm'>The dashboard strongly prioritizes and explains the funding shortfall in a polished dense layout, but the visible workflow stops before intercompany staging, approval, rationale, and resolved-state closure.</td></tr><tr><td>68.97</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-020-data-storytelling-ui</td><td class='sm'>A polished and well-structured Gateflow shell presents strong KPIs and controls, but its blank chart, heatmap, and flight table prevent complete operational monitoring.</td></tr><tr><td>69.39</td><td class='tag'>B/landing</td><td class='cid'>track-b-landing-001-src-268</td><td class='sm'>The page has a polished, calm booking-oriented structure, but incomplete room presentation, a broken image and visibly unrelated photography materially weaken the brief fit.</td></tr><tr><td>69.76</td><td class='tag'>A/landing</td><td class='cid'>landing-110-blackout-xerox-punk</td><td class='sm'>A polished and highly coherent Xerox-punk race landing page covers most content and commerce needs, but lacks several required visual asset types and does not visibly demonstrate populated race-card states.</td></tr><tr><td>69.93</td><td class='tag'>B/dashboard</td><td class='cid'>track-b-dashboard-001-src-368</td><td class='sm'>A polished, decision-ready cold-chain dashboard with strong prioritization and evidence, weakened mainly by the missing corrective-action control and limited trailer/depot detail.</td></tr><tr><td>70.48</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-009-arcade-cabinet-ops</td><td class='sm'>A polished and highly legible arcade race-control dashboard is substantially complete, but its obscured leaderboard and missing visible row-to-inspector anchor are major localized weaknesses.</td></tr><tr><td>70.71</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-016-financial-broadsheet</td><td class='sm'>A polished and decision-ready opening-weekend broadsheet that strongly covers the core workflow, while missing authentic poster thumbnails, correspondent margin notes, restruck estimates, and serif index numerals.</td></tr><tr><td>70.71</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-029-lime-noir-logistics-glass</td><td class='sm'>A polished and decision-ready lime-noir turnover dashboard is shown, but explicit RoomRelay branding, table scrolling, empty-state coverage, and the selected-room inspector flow are absent.</td></tr><tr><td>72.38</td><td class='tag'>A/dashboard</td><td class='cid'>dashboard-041-scada-hmi-industrial</td><td class='sm'>A strong SCADA-style monitoring console with comprehensive visible data and actions, held back by overlapping panels, misplaced charts, and missing acknowledgement identity and TRIP evidence.</td></tr></tbody>
</table></div>
</section>
<section>
<h2>Per-case scores</h2>
<div class="filters"><button class="f on" data-f="all">All</button><button class="f" data-f="landing">landing</button><button class="f" data-f="dashboard">dashboard</button><button class="f" data-f="A">Track A</button><button class="f" data-f="B">Track B</button></div><div class="scroll"><table class="cases" id="caseTable"><thead><tr><th>case_id</th><th>track</th><th>surface</th><th>27B AdamW step400</th><th>9B AdamW step200</th><th>4B AdamW step200</th><th>4B Muon step200</th></tr></thead><tbody><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-002-armin-hofmann-contrast-and-reduction</td><td class="tag">A</td><td class="tag">dashboard</td><td title="The dashboard has a strong monochrome structural shell, but its empty charts, ticket table, station rail, and inspector make the operational experience incomplete.">61.55</td><td class="best" title="A strong, operationally clear Hofmann-style dashboard is weakened mainly by absent rendered system states, inspector placement above rather than beside the table, and red live-status accents outside risk marking.">87.02</td><td title="A strong monochrome operational shell with excellent selection and inspector treatment is materially undermined by the completely unrendered primary trend and missing course-stage and system-state visuals.">78.81</td><td title="The dashboard is cleanly structured and operationally legible, but its blank chart panels and missing history/state treatments leave major brief requirements visibly incomplete.">64.64</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-003-blueprint-hologram-console</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished blueprint-style railway console with strong KPI and inspector coverage is weakened by a tiny topology, an empty primary graph and ledger, and limited conflict-resolution actions.">82.27</td><td title="A polished and readable blueprint-style control dashboard is shown, but its central railway schematic and pinned route visualization are visibly absent.">80.84</td><td class="best" title="A strong, readable railway-control blueprint dashboard with comprehensive operational content, tempered by unclear panel translucency and localized clipping in the right inspector.">90.00</td><td title="A polished blueprint-style railway console is visible, but empty train-graph and occupancy regions plus a disconnected selected-service state substantially weaken operational completeness.">72.27</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-004-dashboard-brutalism</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished, information-dense brutalist RoomBoard console visibly fulfills nearly every requested layout, component, status, and styling requirement.">100.00</td><td title="A polished, decision-ready brutalist hotel console with strong operational coverage, weakened mainly by non-heavy KPI numerals and the absence of a visible stay-date range.">90.95</td><td title="A coherent and information-rich brutalist RoomBoard console is visible, with the main weaknesses being light KPI numerals, ambiguous row alert coloring, and an unshown drill-locked state.">86.87</td><td title="A polished and operationally rich brutalist dashboard is visible, with notable fidelity gaps in KPI typography, date-range presentation, chart labeling and alert-row treatments.">85.44</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-005-dieter-rams-silent-control-panel</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and highly legible silent-control dashboard whose default overview is strong, but the screenshot does not demonstrate the selected-lift inspector, capacity consequence, webcam, maintenance, loading, or active wind-hold states.">66.67</td><td title="The dashboard strongly captures the silent technical-control aesthetic and selected-lift consequence flow, but empty and inconsistent operational data plus missing bar, loading, wind-hold, and webcam visuals reduce completeness.">71.43</td><td class="best" title="A polished, highly legible silent-control dashboard fulfills nearly all operational and visual requirements, with the schematic webcam as the main visible shortfall.">84.29</td><td title="A polished warm technical shell presents useful KPIs and lift states, but missing core analysis and an entirely unpopulated inspector leave the dashboard operationally incomplete.">49.52</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-006-carbon-fiber-telemetry</td><td class="tag">A</td><td class="tag">dashboard</td><td title="The dashboard has polished carbon-fiber styling and a strong structural shell, but its blank core visualizations and empty spacecraft table make the operational experience incomplete.">73.04</td><td class="best" title="A polished and operationally rich carbon-fiber dashboard is weakened mainly by an underdeveloped ground-track visualization and one clipped chart annotation.">93.57</td><td title="A polished and operationally rich telemetry dashboard with strong visual fidelity, weakened mainly by the stacked analytical layout and under-informative ground-track map.">86.43</td><td title="A polished and operationally focused carbon-fiber dashboard is weakened by the missing gauge anchor, telemetry inspector, and explicit pass-lock drill-down.">86.23</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-007-ceramic-bento-metrics</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished, decision-oriented ceramic dashboard with strong core metrics and hierarchy, but incomplete interaction-state evidence and mismatched surface proportions.">76.83</td><td title="The dashboard renders with strong matte ceramic styling and broad metric coverage, but its clipped ledger, unresolved inspector, contradictory forecast state, and card-heavy proportions prevent full operational fidelity.">54.52</td><td title="The page has a polished ceramic shell and readable controls, but its central charts, table data, and selected-state details are visibly missing, making the dashboard operationally incomplete.">59.37</td><td class="best" title="A polished matte ceramic dashboard with strong operational hierarchy and reservation detail, but its channel bar panel is blank and forecast-lock and inspector-table relationships remain incomplete.">81.83</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-008-admin-crud-panel</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and decision-ready catalog dashboard closely matches the intended visual system, but the timestamp is wrong and key bulk-edit, inline-edit, variant, and inventory-badge details are missing.">84.52</td><td title="The page has a polished dashboard shell and correct core colors, but its blank charts, empty table, and missing CRUD controls make the required operational experience substantially incomplete.">38.57</td><td title="The dashboard is polished and operationally focused, but missing bulk controls, non-removable filters, absent product imagery, and the ungrouped revenue chart prevent full brief fidelity.">72.62</td><td title="The dashboard has a polished, cohesive shell and usable table foundation, but major visualization assets, thumbnails, inline editing, and filter-state details are visibly missing or broken.">58.33</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-009-arcade-cabinet-ops</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and highly legible arcade race-control dashboard is substantially complete, but its obscured leaderboard and missing visible row-to-inspector anchor are major localized weaknesses.">70.48</td><td class="best" title="A polished arcade-style operations dashboard with strong hierarchy and selection workflow is materially weakened by an entirely unplotted central telemetry playfield.">77.14</td><td title="The dashboard has a strong styled shell, useful leaderboard, live cues, and selected-car inspector, but a major layout failure leaves the KPI area unframed and the required central playfield largely clipped.">44.76</td><td title="The arcade-inspired frame and KPI presentation are polished, but missing chart, leaderboard, and inspector content leaves the dashboard operationally incomplete.">62.38</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-011-anime-dashboard</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished, dense night-stint command dashboard fulfills nearly every visible requirement, with only sticky behavior unverifiable from a static image.">96.43</td><td title="A polished and highly faithful mecha-HUD dashboard is visible, but the absent timing rows and missing lock-on crosshair prevent complete operational closure.">83.81</td><td title="A polished and strongly themed HUD with clear priorities, but major operational omissions in telemetry, timing fields, circuit depiction, and event-to-row linkage reduce completeness.">78.21</td><td title="The page is styled and data-dense, but clipped controls and KPIs plus empty telemetry, heatmap, and circuit panels leave the requested live mecha-HUD experience substantially incomplete.">59.64</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-012-glass-cockpit-avionics</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and highly legible endurance cockpit covers the core strategy workflow, with notable misses in true tape-gauge form, lozenge styling, and visible caution flashing.">93.57</td><td class="best" title="A polished and highly legible glass-cockpit dashboard covers nearly all requested telemetry, with the six-row roster, indistinct trend lines, and unprovable flashing state as the main visible shortcomings.">93.93</td><td title="A polished glass-cockpit dashboard with strong hierarchy and telemetry styling is undermined by empty gauge panels, an incorrect six-car roster, and several incomplete control details.">65.18</td><td title="The page is professionally styled as a glass cockpit, but its blank gauges, chart, and car ledger omit most operational content required by the brief.">44.88</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-013-bauhaus</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and information-rich Bauhaus water-treatment dashboard is shown, but the required selected failing train and open inspection panel are missing.">93.33</td><td title="The dashboard has strong Bauhaus structure and an effective failure workflow, but both required chart visualizations are visibly absent.">79.05</td><td class="best" title="A polished and highly faithful Bauhaus operations dashboard with only minor literal labeling omissions in the KPI and backwash headers.">97.62</td><td title="A strong, actionable Bauhaus-style operations dashboard is visibly present, but exact labeling, failed-train selection, chart type, clipping, and toolbar wrapping keep it from fully matching the brief.">73.33</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-014-developer-first-minimal</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and information-dense deliverability console closely matches the brief, but lacks provider logos and an explicit throttle action.">90.95</td><td title="A polished and decision-ready deliverability console that closely matches the brief, aside from missing provider logos and a non-full-bleed domain table.">89.52</td><td class="best" title="A polished and highly complete throttle-decision console is undermined mainly by the provider chart rendering every identity as “undefined” and omitting the required marks.">90.95</td><td title="A polished dark operations shell presents the incident and domain queue clearly, but empty charts, undefined authentication data, and missing action controls undermine the throttle decision workflow.">68.01</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-015-ev-instrument-cluster</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and operationally comprehensive EV race console with strong visual hierarchy, offset by stacked comparison panels, non-chip trend styling, and one conflicting forecast readout.">93.57</td><td title="A cohesive and information-rich EV race console is visible, but the four-car mismatch, missing forecast and trends, and severe clipping in the race-energy readouts prevent full fidelity.">78.81</td><td title="The console has a cohesive EV-racing shell and useful scalar telemetry, but critical gauges, plotted data, thermal cells, and both session entries are visibly missing.">63.33</td><td title="The dashboard strongly conveys a cohesive EV-racing telemetry console, but missing session fields, absent forecast closure, four car rows, stacked charts, and overlapping metric text prevent full brief compliance.">60.00</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-016-financial-broadsheet</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and decision-ready opening-weekend broadsheet that strongly covers the core workflow, while missing authentic poster thumbnails, correspondent margin notes, restruck estimates, and serif index numerals.">70.71</td><td class="best" title="A polished and cohesive broadsheet dashboard strongly supports opening-weekend monitoring, but several specified details—especially serif tabular metrics, single-ink charting, poster thumbnails, forecasts, and new-estimate restrikes—are not clearly realized.">74.64</td><td title="A cohesive and information-dense broadsheet dashboard is visible, but blank charts, broken poster assets, and mismatched metric typography prevent full visual completion.">69.64</td><td title="A strong salmon-stock opening-weekend broadsheet with useful reporting depth, undermined by obscured regional charts, misaligned-looking table data, and missing explicit LATE and new-estimate states.">71.67</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-018-crypto-dashboard</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A highly cohesive and decision-ready event-revenue console with strong data density and state cues, weakened only by two missing precision-style details.">90.24</td><td title="The page is visually polished and structurally coherent, but empty event and seat data omit several core operational details from the brief.">61.90</td><td title="The polished dark console shell renders, but blank visualizations, an empty event table, and an unpopulated broken-image drill-down make the dashboard operationally unusable.">38.33</td><td title="A polished and operationally comprehensive dark event-revenue dashboard is undermined mainly by incomplete chart rendering and missing precision details.">86.43</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-019-executive-kpi-dashboard</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and decision-ready Northstar hotel performance dashboard, with the main visible omission being nested evidence rows.">97.38</td><td title="A polished and strongly aligned executive dashboard is undermined by an empty exception table, absent breach/evidence rows, and a channel chart that is not clearly grouped.">83.81</td><td title="A polished and well-structured hotel KPI console is visible, but empty analytical charts, absent nested evidence rows, and incomplete Northstar Hotels branding reduce completeness.">84.76</td><td title="A comprehensive and recognizable hotel revenue console is rendered, though control overlap, truncation, absent KPI arrows, and missing gold benchmark rules reduce visual fidelity and decision polish.">86.19</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-020-data-storytelling-ui</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and well-structured Gateflow shell presents strong KPIs and controls, but its blank chart, heatmap, and flight table prevent complete operational monitoring.">68.97</td><td title="A cohesive control-room shell and informative KPI row are visible, but blank chart, heatmap, and table bodies leave the core live dashboard incomplete.">60.40</td><td class="best" title="A polished and operationally useful dashboard is visible, but its dominant departures-versus-schedule chart fails to render the series and pinned annotations.">82.54</td><td title="A polished dashboard shell presents strong operational structure and KPI messaging, but most core live-data visualizations and rotation details are visibly empty.">52.06</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-021-fluent-design</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and highly usable aquarium operations dashboard whose main visual-fidelity weakness is the substitution of required photography and species imagery with placeholders.">78.10</td><td title="The dashboard is well-structured and operationally informative, but broken habitat and species imagery and incomplete drill-down closure prevent full visual-quality compliance.">76.67</td><td title="A polished, decision-oriented aquarium dashboard is visibly functional, but several explicit visual requirements—especially toolbar filters, acrylic treatment, and habitat/species imagery—are missing.">65.68</td><td title="The page has a polished Fluent-style dashboard shell and sound layout, but broken imagery and empty core chart, table, and history content substantially undermine the required operational experience.">44.52</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-023-adaptive-ambient-ui</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and decision-ready alpine operations dashboard is shown, with the main visual shortfalls being the absent slide-in inspector, missing skeleton state, and unverified pulse animation.">87.02</td><td title="A polished and decision-ready alpine operations dashboard is shown, with the main weakness being absent delayed-telemetry and fully demonstrated inspector states.">83.45</td><td title="A polished and well-structured alpine operations dashboard is weakened by an empty heatmap, unavailable inspector, and limited visible action closure.">78.21</td><td title="A polished and information-dense alpine operations dashboard is visible, but empty throughput and heatmap panels prevent complete fulfillment of the brief.">83.93</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-024-honeycomb-paper-ledger</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and information-dense museum ledger dashboard with strong operational hierarchy, weakened mainly by absent honeycomb motifs and incomplete visible loading, offline, and resolution states.">86.19</td><td title="A strong, information-dense museum ledger dashboard with clear operational hierarchy, weakened by a blank verified-image area, absent demonstrated fallback states, and an inspector that occludes underlying data.">83.81</td><td class="best" title="A strong, highly legible museum-ledger dashboard is undermined by a clipped four-column table, clipped chart ticks, and a visibly broken verified-image region.">88.57</td><td title="The dashboard has a polished museum-ledger shell and strong visual hierarchy, but its blank operational data, placeholder inspector, and absent rendered states make it largely non-decision-ready.">78.10</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-027-kenya-hara-white-and-emptiness</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and highly faithful kiln-release dashboard with strong selected-firing closure, weakened mainly by undemonstrated alternate states and an ambiguously labeled outcome chart.">91.02</td><td class="best" title="A polished, highly faithful release-review dashboard with excellent visual hierarchy and decision context, weakened mainly by the lack of a visible cone-history view and fully rendered alternate states.">97.14</td><td title="The dashboard strongly captures the requested sparse warm-white design and operational structure, but its blank primary curve and erroneous 8300.0% yield materially reduce decision quality.">92.24</td><td title="The visual direction is highly faithful and polished, but blank analytical panels and absent firing rows make the dashboard operationally unusable for release decisions.">78.37</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-029-lime-noir-logistics-glass</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and decision-ready lime-noir turnover dashboard is shown, but explicit RoomRelay branding, table scrolling, empty-state coverage, and the selected-room inspector flow are absent.">70.71</td><td class="best" title="A polished and highly faithful turnover dashboard with strong operational hierarchy, weakened mainly by the absent visible table scrollbar, unexplained table void, and no explicit locked-timeline indicator.">90.95</td><td title="A polished RoomRelay command shell and detailed room queue are visible, but blank chart panels, no table scrollbar, and unclear PMS-offline and timeline-lock states prevent complete fulfillment.">66.90</td><td title="The dashboard has strong Lime Noir styling and a useful room inspector, but missing chart rendering and absent offline and empty states materially reduce operational completeness.">72.62</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-030-lucky-swaps-volumetric-console</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and highly faithful registrar workbench with strong decision support, weakened mainly by placeholder artwork visuals and the absence of collection search.">82.86</td><td title="A strong, focused registrar console with clear decision metrics and action hierarchy is materially weakened by broken artwork imagery, no collection search, and no explicit storage control.">77.86</td><td title="The dashboard strongly matches the intended rotation-console structure and workflow, but missing top-date chrome, absent collection search, and broken artwork thumbnails prevent full completion.">79.05</td><td class="best" title="A polished and operationally legible charcoal rotation console is undermined by a misplaced support stack, excessive empty space, and artwork/chart areas that read as placeholders.">84.76</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-031-mac-os-aqua-nostalgia</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A highly faithful and operationally comprehensive Aqua satellite console whose main visible weakness is the severely washed-out spacecraft table body.">98.57</td><td title="The Aqua shell is highly faithful and well organized, but the empty operational panels and missing spacecraft table make the live dashboard largely unusable.">74.69</td><td title="The dashboard strongly captures Aqua-era system styling and operational hierarchy, but its empty core plots and several missing or inconsistent operational details materially reduce readiness.">78.98</td><td title="The dashboard has strong Aqua chrome and broad operational components, but its split-window layout and entirely blank ground-track and telemetry panels are major visual and functional failures.">65.17</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-032-massimo-vignelli-typographic-discipline</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A highly faithful civic-typographic operations dashboard with strong monitoring coverage, weakened mainly by showing the inspector only in its unpinned empty state.">94.42</td><td title="A strong civic-typographic operations dashboard is visually complete, but contradictory selected-zone behavior and several data inconsistencies weaken operational closure.">89.52</td><td title="A strong civic-modernist operations dashboard is visually complete and disciplined, but invalid KPI values, unshown non-data states, and incomplete visible selection closure reduce operational trust.">92.99</td><td title="A strong civic-information dashboard with excellent typographic discipline, weakened by undefined occupancy labels, stacked rather than paired charts, and incomplete state and capacity-rule presentation.">84.89</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-033-motorsport-pit-wall</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A strong and feature-rich pit-wall console is visibly compromised by an overcrowded runner tower, clipped KPI readouts and a moisture grid that does not trace the track.">86.69</td><td title="A cohesive dark steward console presents the runner tower and inquiry workflow, but clipped readouts, unclear runner details, and blank speed and moisture visuals severely weaken the result.">67.30</td><td title="A polished and convincing steward console is visible, but incomplete runner coverage and several absent or unreadable race-data assets materially reduce operational completeness.">66.80</td><td title="The console establishes strong hierarchy and race-control styling, but blank analysis panels and a visibly broken photo-finish section keep it from being decision-ready.">77.38</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-035-observability-dark</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and highly complete airport-operations console with strong hierarchy and actionable breach handling, aside from unverified sticky behavior and chart panels not sharing a left edge.">91.67</td><td title="A polished and information-dense airport operations console with strong brief fidelity, weakened mainly by the visible NaN% KPI and limited action closure.">87.38</td><td title="A polished and information-dense airside operations console with strong hierarchy, weakened mainly by absent green KPI deltas, unverifiable sticky behavior, and malformed selected-flight detail text.">83.81</td><td title="A polished and information-dense airside console with strong visual fidelity, though its main turnaround chart is phase-based rather than a true time series.">85.79</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-036-obsidian-aura-portfolio</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and decision-oriented TableAtlas dashboard closely matches the requested visual system, with notable gaps in visible revenue charting, table detail, state variants, and locked-curve confirmation.">89.57</td><td title="The dashboard has a strong portfolio structure and dark-violet visual system, but missing revenue, clipped service data, ambiguous drill-down locking, absent states, and broken venue imagery prevent full satisfaction.">81.71</td><td title="The screenshot presents a polished TableAtlas dashboard shell, but its pervasive empty state prevents meaningful portfolio analysis or reservation drill-down.">58.16</td><td title="The dashboard strongly communicates restaurant portfolio operations with effective dark-glass structure and drill-down, but broken photography, absent revenue, and incomplete control/state coverage prevent full satisfaction.">72.77</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-037-pale-grid-node-console</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A cohesive and decision-ready ReefSignal console fulfills nearly all visible requirements, with the main shortfall being the lack of recognizable survey-photo assets and fully demonstrated non-live states.">90.00</td><td title="A polished and highly usable reef-health dashboard with strong visual fidelity, weakened mainly by the merged coral-cover summary and placeholder survey photo.">88.70</td><td title="A visually faithful and operationally broad ReefSignal dashboard is undermined by an empty recovery chart, clipped table content, placeholder-like photos, and undefined data.">74.05</td><td title="A visually cohesive ReefSignal shell presents useful reef metrics and survey data, but missing line, bar, and node charts plus absent photo and note drill-downs leave core functionality visibly incomplete.">56.47</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-039-retro-lcd-control-panel</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="The page strongly delivers WardSignal&#x27;s retro LCD dashboard hierarchy and core operational data, but the selected-capacity drill-down workflow, inspector content, and complete state coverage are not visibly realized.">87.94</td><td title="A polished and legible retro hospital control panel with strong operational hierarchy, but incomplete filtering/state coverage and a visibly unresolved selected-unit workflow.">82.22</td><td title="This is a polished and operationally credible WardSignal LCD dashboard, but it misses the required area proportions, upper chart pairing, segmented KPI treatment, and complete selected-bar filtering behavior.">76.98</td><td title="A convincing retro hospital control-panel shell covers most required operations content, but incorrect area proportions, ambiguous chart linkage, incomplete action mapping, and a severe Staffing Gap overflow prevent a fully competent result.">81.11</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-040-rpg-guild-operations</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and operationally complete RPG festival dashboard, with the main visible weaknesses being understated Stagekeep branding and mostly initial-based rather than portrait crew avatars.">88.10</td><td title="A polished and highly actionable three-stage guild-operations dashboard with strong RPG theming, weakened mainly by understated Stagekeep branding and non-portrait crew avatars.">86.67</td><td class="best" title="The screenshot delivers a polished, information-rich RPG festival command dashboard, with only inconsistent use of portrait imagery for crew avatars.">92.86</td><td title="A polished, highly complete guild-operations dashboard with strong task escalation and action controls, weakened mainly by plain tier letters and initial-only crew avatars.">91.76</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-041-scada-hmi-industrial</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A strong SCADA-style monitoring console with comprehensive visible data and actions, held back by overlapping panels, misplaced charts, and missing acknowledgement identity and TRIP evidence.">72.38</td><td title="A polished and usable industrial HMI with strong chart and alarm presentation, but missing two required process tiles and several acknowledgement and quality-state details.">69.05</td><td class="best" title="The console has a strong industrial SCADA frame and accurate headline metrics, but blank charts and an incomplete alarm presentation substantially limit operational usefulness.">74.52</td><td title="The dashboard has a credible industrial frame and accurate headline metrics, but most operational data visualizations, turbine states, and actionable alarm content are visibly missing.">55.48</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-042-soft-rubber-control-room</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and operationally rich soft-rubber dashboard is clearly rendered, with the main visible gaps being the absent empty state and an ambiguous row-level inspector affordance.">96.19</td><td class="best" title="A polished, highly legible soft-rubber operations dashboard fulfills nearly all visible requirements, with the main omission being explicit loading and empty states.">96.19</td><td title="A polished soft-rubber operations dashboard with strong hierarchy and status visibility, limited mainly by the clipped departure table and row-level actions.">89.76</td><td title="The dashboard is polished and structurally comprehensive, but empty operational content, a stacked visualization layout, and low-contrast neumorphic surfaces substantially reduce usability.">77.14</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-043-split-flap-departure-board</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A highly faithful and operationally focused split-flap production console whose main visible shortcoming is the absence of the hourly-output chart and backlog-hours KPI.">94.29</td><td title="The dashboard strongly captures GrooveFlap&#x27;s industrial visual language and supporting controls, but the central production schedule is visibly malformed and operationally unreadable.">58.53</td><td title="The page is polished and operationally readable, but major brief elements—especially both charts, six complete lanes, blank flaps, and true segmented split-flap rows—are absent.">70.74</td><td title="The industrial dashboard shell and charts render, but widespread [object HTMLSpanElement] output leaves the core production board visibly broken and unusable.">42.81</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-044-sticker-sheet-analytics</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and highly legible sticker-sheet revenue dashboard that covers the core workflow well but does not visibly demonstrate the selected reservation, locked rate band, booking history, loading, or error states.">91.43</td><td title="A polished, highly legible peak-weekend analytics sheet with strong visual fidelity, but incomplete filters and an undemonstrated reservation drill-down workflow.">88.57</td><td title="The page strongly captures the requested sticker-sheet visual language and dashboard framing, but missing chart, bar, table, and inspector data leaves the core revenue workflow unusable.">82.86</td><td title="A polished sticker-sheet dashboard with a clear locked-rate workflow, weakened by side-by-side chart/table placement, clipped reservation data, and an obstructive callout.">81.43</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-045-streaming-dashboard-obs</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished dark console shell is visible, but most core storm-tracking visualizations and table content are blank.">46.43</td><td class="best" title="A polished and highly legible severe-weather console that closely matches the brief except for the required discrete 72px intensity rail and visible keyboard-swap affordance.">92.14</td><td title="A polished, highly legible weather console with strong monitoring hierarchy, undermined mainly by the incorrect severity visualization, inconsistent positive-delta color, and missing rapid-rotation dot treatment.">84.52</td><td title="The styled dark console shell is recognizable, but collapsed metrics and missing radar, warning, and storm-cell data leave the rendered dashboard visibly broken and unusable.">38.09</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-046-studded-brick-ops</td><td class="tag">A</td><td class="tag">dashboard</td><td title="A polished and strongly themed housekeeping dashboard is visibly actionable, but its blank heatmap and merely labeled loading/error states leave key operational and interaction requirements incomplete.">87.78</td><td title="The dashboard has strong RoomRelay hierarchy and convincing brick styling, but blank analytical panels and an empty task queue substantially undermine operational usefulness.">68.17</td><td class="best" title="A polished and comprehensive brick-themed housekeeping dashboard with strong operational closure, weakened mainly by an uninformative heatmap and the lack of an explicit row drill-down affordance.">96.98</td><td title="The styled brick dashboard has strong branding and KPI treatment, but its blank heatmap, trend, and task table leave the central operational experience incomplete.">61.98</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-047-susan-kare-pixel-icon-clarity</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A strong pixel-utility operations dashboard with clear monitoring hierarchy and drill-down, weakened by broken artwork assets and several required states that are labeled but not visibly demonstrated.">76.34</td><td title="A strong pixel-utility monitoring shell is visible, but broken artwork assets and incomplete selection, history, loading, and alert states prevent full satisfaction.">75.76</td><td title="A functional and strongly themed museum-operations dashboard is visible, but left-edge occlusion and several missing drill-down, state, alert, and artwork details prevent full compliance.">63.98</td><td title="The styled pixel-utility shell and KPIs render, but the central operational data and required drill-down states are largely empty or missing.">42.66</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-048-teenage-engineering-industrial</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished, information-dense industrial waterworks dashboard with strong operational clarity, weakened mainly by absent row drill-downs, unverifiable LED pulsing, and the date toolbar’s lower placement.">85.24</td><td title="A polished and operationally rich industrial dashboard with strong hierarchy, but several specified signature components and state treatments are not visibly realized.">74.76</td><td title="A polished industrial dashboard shell with strong KPIs, table, filters, and alerts is materially weakened by completely blank chart bodies and several missing interaction cues.">73.33</td><td title="A visually coherent industrial waterworks dashboard is present, but the oversized empty header, missing stacked flow bars, undersized table type, and incomplete alert and drill-down cues prevent full satisfaction.">69.76</td></tr><tr data-track="A" data-surface="dashboard"><td class="cid">dashboard-049-topographic-terrain-console</td><td class="tag">A</td><td class="tag">dashboard</td><td class="best" title="A polished and operationally complete caldera-monitoring console is shown, with the only clear omission being a separately labeled sensor-layer filter.">99.11</td><td title="The dashboard strongly communicates an earth-toned volcano console, but major unused space, a truncated station table, and the missing selected-station map pin prevent a fully competent result.">91.07</td><td title="The dashboard has a polished caldera-console shell and complete KPI/filter framing, but its core terrain, seismic, field-log, advisory, and station-inspection content is visibly absent or unavailable.">58.81</td><td title="The styled topographic console has strong visual theming and hierarchy, but empty operational panels and an overlapping inspector make it largely non-decision-ready.">70.42</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-001-1970s-documentary</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A highly polished and complete period-documentary bus-line landing page with minor misses in the ticket-booth asset and passenger-note typography.">90.24</td><td title="A polished and highly complete period-styled bus-line page is weakened mainly by the incorrect hero subject and vertically stacked route picker.">80.48</td><td title="A polished and highly complete regional coach landing page with excellent documentary styling and practical information, weakened mainly by the daylight hero and missing ticket-booth image.">87.62</td><td title="A polished and highly complete regional coach landing page with excellent documentary styling, weakened mainly by the daylight hero, repeated imagery, and absence of multiple driver dossiers or a visible ticket booth.">82.62</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-002-afrofuturism</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete Afrofuturist cooperative landing page whose principal weakness is a generic harbor hero that omits the specified solar canopies, battery cells, and weathered hands.">84.05</td><td title="A polished and functionally complete Afrofuturist cooperative page is weakened mainly by a hero image that omits the requested solar canopies, battery cells, and weathered hands.">79.29</td><td title="A polished and highly complete cooperative landing page with strong operational modules, weakened mainly by a hero that omits the brief&#x27;s photovoltaic, battery, hands and connected-constellation imagery.">74.52</td><td title="A polished and highly complete cooperative microgrid landing page with strong practical Afrofuturist styling, weakened mainly by the hero&#x27;s missing specified solar and human-detail imagery.">74.52</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-003-1980s-arcade-cinema</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete arcade-cinema aquarium landing page whose only unsupported requirement is the ticket button&#x27;s temporal flashing behavior.">94.29</td><td class="best" title="The screenshot shows a highly complete, legible, and stylistically faithful after-dark aquarium landing page with strong purchase and planning support.">100.00</td><td title="A polished and highly complete arcade-aquarium landing page with strong hierarchy and decision support, weakened mainly by incomplete species-specific imagery and no pixel-verifiable flashing treatment.">82.86</td><td title="A strongly styled and decision-ready after-dark aquarium page is undermined by empty species cards, missing jellyfish and tide-pool imagery, and a malformed footer.">72.86</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-004-70s-warm-retro</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete warm-retro bicycle-shop page with strong service guidance, weakened mainly by the separated hero dial and incomplete mechanic portrait imagery.">87.86</td><td title="A polished and highly complete warm-retro bicycle-shop landing page whose main visible shortfall is the absence of actual route maps and consistently clear restored-bike imagery.">85.71</td><td class="best" title="A polished and comprehensive warm-retro bicycle-shop landing page, with only limited visual proof that every restored-bike card depicts a full steel-frame commuter.">97.14</td><td title="A strong warm-retro service layout with an excellent hero and booking flow is materially undermined by missing mechanic images and incorrect parts-wall and route-map assets.">85.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-005-april-greiman-digital-collage</td><td class="tag">A</td><td class="tag">landing</td><td title="A strong, highly structured Greiman-inspired club landing page is substantially undermined by multiple conspicuous broken image placeholders.">87.86</td><td class="best" title="A polished, kinetic metropolitan collage delivers nearly all core landing-page modules, with the main visible gaps being repair-bench, reflective-gear, and pervasive translucent route-sheet imagery.">90.24</td><td title="A polished, highly legible metropolitan cycling landing page covers the full service journey, with its main weakness being incomplete asset variety and a restrained collage treatment outside the hero.">90.00</td><td class="best" title="A polished and highly complete metropolitan bicycle-club landing page with strong collage styling, weakened mainly by opaque route sheets and the absence of clear street-map and moving-rider visuals.">90.24</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-006-acid-graphics</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete acid-courier landing page with strong imagery, decision support, and booking closure, weakened mainly by unwarped headline typography and a map disconnected from the zone grid.">90.24</td><td title="A polished and highly complete courier landing page with strong acid styling and booking flow, weakened mainly by missing waybill and route-map materials.">80.71</td><td class="best" title="A polished, highly complete Acid Graphics courier landing page with strong hierarchy and booking support, weakened mainly by missing torn-paper texture and true warped headline treatment.">90.24</td><td title="A strong and comprehensive acid-styled courier landing page is weakened by an obscured hero and several missing signature brief details.">73.33</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-007-a24-indie-editorial</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, atmospheric editorial landing page fulfills the full visit-selection journey, with the main visual shortfall being insufficiently explicit fog and film grain.">85.48</td><td title="A polished and complete indie-editorial booking page whose main visual shortfall is that the photography appears brighter and clearer than the requested grainy, fogged dusk setting.">83.10</td><td class="best" title="A polished and comprehensive botanical-night editorial landing page, with only the chapter numbering lacking a visibly hand-rendered character.">96.43</td><td class="best" title="A highly complete and polished indie-editorial landing page whose only clear miss is that the conservation chapter numbering does not look hand-rendered.">96.43</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-008-abstract-expressionism</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete salvage-yard landing page with strong hierarchy and workflow support, though some category imagery and the requested rubbed, painterly styling are not fully evident.">83.33</td><td title="A polished and highly usable salvage-yard landing page fulfills the main inventory and hold workflow, with notable gaps in rubbed-label treatment and tile, sink, and deconstruction imagery.">85.24</td><td class="best" title="A highly polished, historically aware salvage-yard landing page with excellent visual fidelity and workflow closure, weakened mainly by the missing visible search field and unclear sink imagery.">87.14</td><td title="A polished industrial salvage page with strong supporting modules, but its empty primary ledger and missing inventory assets substantially weaken completeness and the requested expressionist treatment.">58.10</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-009-adaptive-ambient-ui</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, comprehensive observatory landing page fulfills nearly every visible requirement, with the sole mismatch being that the hero headline is not the Nightjar Observatory name.">97.14</td><td title="A polished and comprehensive observatory landing page with strong planning support, weakened mainly by the overlaid hero card and missing ambient gradient and clearly recognizable gallery assets.">77.86</td><td title="A polished and highly structured observatory landing page with strong planning support, weakened by an empty conditions card, mismatched gallery assets, and an overly forceful concluding mood.">74.86</td><td title="A coherent and mostly complete observatory landing page offers strong planning and booking support, but misses the requested image-based telescope and constellation gallery and has notable contrast and gallery-layout weaknesses.">72.96</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-011-80s-synthwave</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete synthwave drive-in landing page whose main visual shortcoming is the absence of rows of parked cars beneath a Moonbeam-branded marquee hero.">86.67</td><td title="A polished and highly informative synthwave ticketing page fulfills nearly every section requirement, but its hero lacks the defining parked-car scene and star-soaked imagery.">79.39</td><td title="A highly complete and cohesive synthwave drive-in page whose main visible shortcoming is the lack of clearly depicted parked-car rows in the hero.">79.52</td><td title="A polished and highly usable synthwave drive-in page includes nearly every requested section, but the hero lacks parked cars and the cartridge posters lack substantive illustrations.">68.91</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-012-academia</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, comprehensive botanical-society landing page with strong academic structure and programme detail, weakened only by the lack of distinct botanical-name styling and expedition-linked archival imagery.">91.67</td><td class="best" title="A polished and highly complete academic botanical landing page, with only the lack of clearly distinguished botanical-name typography falling short.">96.43</td><td title="A polished and highly faithful botanical-society landing page with strong programme and membership detail, weakened mainly by separately ungrouped workshop and survey content and a three-item schedule labeled as four.">92.86</td><td title="A polished and comprehensive botanical-society landing page, with the only notable gaps being no standalone countryside-survey section and no clearly distinguished botanical binomials.">93.57</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-013-bohemian</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and comprehensive luthier landing page visibly fulfills the requested content, workflow, and warm handcrafted Bohemian direction.">100.00</td><td title="A polished, cohesive luthier landing page with strong storytelling and selection support, weakened by missing wood texture and an unreadable appointment label.">85.00</td><td title="A polished, complete luthier landing page with excellent warm craft styling and conversion flow, weakened mainly by reused workshop imagery in the musician-session section and no explicit independent-luthier introduction.">95.00</td><td class="best" title="A polished, complete luthier landing page with strong visual hierarchy, tactile workshop character, useful product detail, and a clear inquiry path.">100.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-015-biomorphic-design</td><td class="tag">A</td><td class="tag">landing</td><td title="The page is polished and atmospherically cohesive with complete downstream sections and reservation flow, but the missing hero selector, absent table-lamp clarity, and occluded compact-card copy weaken fidelity and usability.">68.10</td><td title="A polished and cohesive biomorphic landing page with excellent section coverage and reservation flow, weakened mainly by non-pebble product cards and process imagery that does not clearly show finished lamps.">83.10</td><td class="best" title="A polished and cohesive lamp landing page includes every major section and a complete reservation flow, but misses the specified molten-glass hero treatment and genuinely pebble-shaped product cards.">87.38</td><td title="A polished, cohesive glass-studio landing page with strong content structure and reservation closure, weakened by the missing hero selector and indistinct colour-and-texture options.">73.33</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-016-de-stijl</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete De Stijl bakery landing page with only the explicit neighborhood wording missing from the pickup-selector label.">96.43</td><td title="A polished and unusually complete bakery landing page with strong De Stijl structure, weakened mainly by a hero that does not deliver the specified colored grid around the loaf image.">75.48</td><td title="A polished and highly complete De Stijl bakery landing page, with the main visible weakness being the hero&#x27;s limited multicolor grid treatment and understated brand name.">91.19</td><td class="best" title="A complete and highly recognizable De Stijl bakery landing page with strong ordering support, despite localized hero typography overflow.">98.57</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-017-anodized-metal</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, information-rich industrial landing page is undermined mainly by broken dusk and field-use imagery and the absence of a visible engraved mode-ring close-up.">87.62</td><td title="A polished and comprehensive technical flashlight landing page with strong ordering closure, held back mainly by missing explicit macro close-ups of the knurled body and engraved mode ring.">84.29</td><td title="A polished, highly structured technical landing page covers nearly the full product story and purchase flow, with missing flashlight macro photography as its primary visual shortfall.">78.10</td><td title="A polished, technically credible metal-themed landing page fulfills nearly all content and conversion needs, with missing flashlight macro imagery and understated brand-heading prominence as the main gaps.">84.29</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-018-indie-sleaze-2005</td><td class="tag">A</td><td class="tag">landing</td><td title="A highly coherent and feature-complete neon bowling-league landing page is weakened mainly by one visibly missing snack image.">98.57</td><td title="A highly coherent and actionable neon bowling-league landing page delivers nearly every requested element, with the clear exception of a dedicated lime fixture grid.">81.07</td><td class="best" title="A complete, coherent, and highly faithful late-night bowling landing page visibly covers every major requested feature.">100.00</td><td title="A strong late-night neon bowling concept with clear booking flow is weakened by an absent shoe-size strip, low-contrast fixture details, sparse media panels, and limited visual chaos.">85.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-019-paul-rand-modernist-wit</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and memorable modernist cinema landing page delivers the full booking and membership structure, with the main shortfall being the absence of most requested photo categories.">80.00</td><td class="best" title="A polished and memorable modernist cinema page delivers the required sections and booking path, but omits most of the requested photographic subject variety.">85.71</td><td title="A polished and memorable modernist cinema landing page with strong functional content, weakened mainly by missing requested photo categories, staggered program-card alignment, and a duplicated Book tickets CTA.">78.41</td><td title="A polished and highly distinctive cinema landing page with strong content and booking support, weakened by missing requested photo categories, an unclear reel-eye symbol, and repeated booking buttons.">70.64</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-020-arc-browser-like-playful-productivity</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly complete Wayfare landing page with strong Arc-like structure and collaboration cues, weakened only by the lack of visible map and shared-note content.">95.97</td><td class="best" title="A polished, comprehensive Wayfare landing page closely matches the playful productivity brief, with the only notable visual omission being actual map content.">97.27</td><td title="A polished, companionable Wayfare landing page closely fulfills the brief, with the main visible gaps being absent map content and non-pill pricing presentation.">93.12</td><td title="A polished, highly legible group-trip landing page with strong workspace coverage, weakened mainly by absent map and note content and mismatched CTA and pricing treatments.">85.52</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-023-coastal-grandmother</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete coastal watercolor landing page, with the main visible shortfall being the absence of a half-finished harbor study in the hero.">92.86</td><td title="A polished and comprehensive coastal workshop landing page with excellent atmosphere and booking support, weakened mainly by the hero’s missing clearly visible brushes and pigment tins.">85.00</td><td title="A polished, highly coherent coastal watercolor-school landing page with strong booking support, though the requested half-finished harbor imagery is absent and supply photography is repetitive.">91.43</td><td title="A polished and complete booking page with strong coastal editorial styling, weakened mainly by a dark, repetitive image set that omits the requested harbor-study imagery.">80.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-025-bauhaus</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete Bauhaus landing page with excellent technical detail, weakened mainly by missing hero speaker photography and multiple Listen + Order CTAs.">88.81</td><td title="A polished and comprehensive Bauhaus landing page with strong product guidance, but its dark angled hero misses the requested straight-on dual-product presentation and duplicates the listen-and-order CTA.">79.29</td><td title="A polished and highly legible Bauhaus product page with comprehensive decision content, weakened mainly by absent hero speaker imagery, a non-card sound layout, and duplicated order CTAs.">80.58</td><td title="A polished and highly structured Bauhaus page covers nearly every supporting section, but its hero and ordering flow do not fully present or close the complete turntable-and-speaker proposition.">74.29</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-026-carbon-fiber</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, technically styled single-product landing page covers the purchasing journey well, with its main shortcomings being the crowded hero, limited macro photography, and non-comparative bin-fit presentation.">76.19</td><td class="best" title="A highly complete, precise carbon-fiber product landing page with strong decision support and ordering closure, weakened only by understated brand naming and absent explicit layover counts.">92.86</td><td title="A polished and information-rich carbon-fiber landing page with strong styling and conversion structure, weakened mainly by the unfocused hero and missing true video and cutaway visuals.">70.00</td><td title="A polished, information-rich carbon-fiber landing page with strong technical luxury styling, weakened by the missing hero carry-on, absent video treatment, abstract cutaway, and incorrect solid-red CTA treatment.">80.95</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-027-art-deco</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished Art Deco landing page with strong atmosphere and booking emphasis is materially weakened by the absent timetable, non-symmetrical hero imagery, and incomplete offering grids.">50.95</td><td class="best" title="This is a polished and highly complete Art Deco lido landing page, with only the duplicated booking CTA and lack of clearly tiled sunburst motifs weakening exact brief adherence.">92.86</td><td title="This is a polished and functional Art Deco landing page whose strong timetable, conditions, restoration, and booking sections are undermined by an asymmetric hero, duplicated CTA, and limited offering grids.">65.00</td><td title="A polished and highly usable coastal-lido landing page with strong Deco atmosphere, but several signature brief elements and repeatable offer grids are visibly missing.">67.38</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-028-christopher-nolan-esque</td><td class="tag">A</td><td class="tag">landing</td><td title="A highly polished and complete cinematic industrial landing page with strong tour decision support, weakened only by duplicate reserve actions, understated Iron Vale naming, and digitally styled rather than imaged archival logs.">85.00</td><td title="A highly polished cinematic tour page with excellent hierarchy and booking support, weakened mainly by missing dam, diagram and archival-document assets and by having two reserve CTAs.">82.14</td><td title="A highly cohesive cinematic industrial tour page covers nearly all requested content, with its main shortcomings being duplicated reserve actions and missing full-bleed generator and archival-document imagery.">87.14</td><td class="best" title="A polished, cinematic industrial tour page with strong narrative and booking information, weakened mainly by missing generator and archival-document imagery plus a duplicated reserve CTA.">88.57</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-029-aurora-mesh</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, complete, and highly faithful after-dark aquarium landing page, with only the absence of distinct ticket-type labels as a visible weakness.">96.43</td><td title="A polished and atmospheric aquarium landing page covers nearly all requested content, but its hero panel lacks a true ticket picker and is positioned to the side rather than centrally.">82.38</td><td class="best" title="A polished, highly faithful aquarium landing page with strong booking and visitor information, weakened mainly by one visibly failed scientist portrait.">98.57</td><td title="A polished, luminous aquarium landing page with comprehensive visit content, weakened mainly by the picker’s placement outside the hero and one visibly broken profile image.">88.57</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-031-kubrick-symmetry</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and complete whisky-release page with strong archival atmosphere and allocation usability, weakened chiefly by an off-axis, non-symmetrical hero and limited explicit coastal imagery.">81.67</td><td class="best" title="A polished, highly complete and disciplined whisky release page with strong symmetry and archival restraint, though its provenance imagery does not visibly depict the coast itself.">95.24</td><td class="best" title="A highly polished and complete symmetrical whisky release page, with the only clear shortfall being the absence of coastal imagery inside the provenance index itself.">95.24</td><td title="A polished and content-complete whisky release page with strong atmosphere and hierarchy, weakened mainly by an asymmetric hero and text-only provenance index.">81.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-032-scandinavian-minimal</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, comprehensive Scandinavian ferry landing page whose only clear brief mismatch is the repetition of the Book passage button.">97.96</td><td class="best" title="A polished, complete Scandinavian civic-ferry landing page with strong information flow, though the chosen sailing is not visibly defined by a clearly unique gentle shadow.">97.96</td><td title="A polished and unusually complete Scandinavian ferry landing page, with only the repeated booking CTAs and absence of an unmistakable harbour photograph weakening strict brief adherence.">91.77</td><td class="best" title="A polished and comprehensive Scandinavian-minimal ferry landing page, with the only clear brief deviation being multiple booking buttons instead of one.">97.96</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-033-underground-brutalism</td><td class="tag">A</td><td class="tag">landing</td><td title="A highly cohesive brutalist salvage landing page delivers strong inventory and decision detail but omits search, lot-specific restoration notes, and visible checkout validation while overusing acid lime.">87.38</td><td title="A highly cohesive Underground Brutalism salvage landing page covers nearly all major content, with notable gaps in search, visible validation, signal-lock confirmation, and strict lime-color confinement.">90.40</td><td class="best" title="A highly cohesive and content-rich brutalist salvage landing page is weakened mainly by the missing inventory search and overextended acid-lime usage.">90.56</td><td title="A highly cohesive brutalist salvage landing page covers most content requirements but omits inventory search and material/condition filters, visible validation, reserved labeling, and the SIGNAL RUPTURE lock interaction.">81.35</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-034-brushed-aluminum</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, technically precise aluminum-themed landing page covers nearly the full brief, with the main visible shortfall being singular rather than plural charts and testimonials.">95.00</td><td title="A polished, highly technical espresso-machine landing page fulfills nearly all requested content, with the notable visual omission of true boiler cutaways and a side-by-side hero-photo composition.">83.69</td><td title="A polished and highly complete technical espresso-machine landing page, with only the independent-maker claim and the requested side-by-side hero-photo arrangement not clearly shown.">91.67</td><td title="A polished and comprehensive technical landing page with strong aluminum styling and reservation closure, weakened mainly by a literal cutaway shortfall and cramped service-note layout.">92.38</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-035-linen-texture</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete luthier landing page with excellent style and decision support, weakened mainly by several specific hero-composition details and only one visible maker portrait.">80.24</td><td title="A polished and cohesive luthier landing page covers nearly the full narrative, but the required stitched hero tonewood selector and unmistakable linen-draped bench are absent.">78.20</td><td class="best" title="A polished and coherent luthier landing page with excellent craft styling and inquiry flow, weakened mainly by absent maker portraits and an unclear linen-draped hero bench.">86.43</td><td title="A polished, highly complete luthier landing page with strong linen-and-terracotta craft styling, weakened only by the overlaid rather than adjacent selector and incomplete maker-portrait presentation.">84.52</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-036-energy-utility</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and complete municipal street-lighting readout combines credible nighttime operations imagery with clear data, maintenance reporting, and outage workflows.">100.00</td><td title="A polished and highly complete municipal night-service landing page, with the only clear structural miss being the lack of repeated LED-upgrade panels.">93.81</td><td title="A polished and comprehensive municipal lighting dashboard fulfills nearly every visible requirement, with the main exception that the ward filter is positioned above rather than beside the readout.">96.43</td><td title="A polished and information-rich utility landing page fulfills most operational requirements, with notable gaps in the sticky ward filter, chart time labeling and rain-dark visual mood.">80.48</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-037-constructivism</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A highly cohesive Constructivist museum landing page is undermined by missing hero rail graphics and visibly broken volunteer portraits.">80.48</td><td title="A highly complete and visually disciplined Constructivist museum page is weakened mainly by a three-part hero arrangement, non-overlaid rail graphics, and machinery-heavy volunteer imagery.">80.24</td><td title="A polished and highly complete Constructivist museum landing page with strong booking and visit flows, weakened mainly by missing crew portraits, no explicit exhibition index, and limited weathered-steel surface texture.">75.48</td><td title="A polished and comprehensive Constructivist museum page with strong information design, weakened by missing diagonal hero rails, absent crew portraits, and limited weathered texture.">70.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-038-bold-typography</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly cohesive landing page visibly fulfills the full winter-lido brief with clear information and a direct booking path.">100.00</td><td class="best" title="A polished, highly cohesive winter-lido landing page visibly fulfills the requested content, decision support, and bold typographic style.">100.00</td><td class="best" title="A polished, cohesive landing page fully delivers the bold winter-lido concept with clear operational detail and a complete booking path.">100.00</td><td title="A polished, highly legible winter-lido landing page fulfills nearly all content needs, with the main visual misses being non-ticket-like pass cards, duplicated primary CTAs, and an incongruous sauna image.">85.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-039-blade-runner-noir</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and usable noir catalog with strong hierarchy and reservation flow is undermined by missing record-cover, turntable, and listening-booth visuals.">78.57</td><td class="best" title="A polished and highly functional noir vinyl-shop landing page whose main weakness is substituting repeated storefront photography for record sleeves, covers, turntable controls, and a clearly shown listening booth.">84.29</td><td title="A polished and highly coherent noir vinyl landing page with strong catalog and reservation flows, weakened mainly by missing record-sleeve, turntable-control, and clearly depicted booth visuals.">80.71</td><td title="A polished and highly faithful noir record-shop page with strong reservation closure, weakened by missing sleeve art and a severely clipped new-arrivals catalog.">80.24</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-040-letterpress</td><td class="tag">A</td><td class="tag">landing</td><td title="A cohesive and highly usable letterpress workshop page is weakened mainly by missing component swatches, an unclear brass badge, and only one photographed recent build.">89.05</td><td class="best" title="A polished and highly cohesive letterpress bicycle-workshop page with excellent content and workflow coverage, weakened mainly by the hero’s lack of a clearly visible lugged frame and brass head badge.">91.43</td><td title="A polished and cohesive letterpress workshop landing page with strong product guidance and a complete fitting flow, though the hero does not clearly show the requested brass badge or worn saddle and its debossing is ambiguous.">89.05</td><td title="A compelling and information-rich letterpress bicycle page is undermined by missing hero details, no actual deposit workflow, and a visibly malformed process column.">77.38</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-041-stop-motion-craft</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, cohesive pottery-drop landing page communicates the collection and reservation flow well, but several specific handmade and overhead-photography requirements are visibly absent.">68.10</td><td class="best" title="A polished, cohesive pottery-drop landing page with strong reservation support, weakened mainly by the missing product grid and incomplete handwritten and plywood details.">78.10</td><td title="A polished, cohesive kiln-drop landing page with strong storytelling and reservation support, weakened by the absent collection grid and several specifically requested handmade-photo details.">58.57</td><td title="A polished, coherent pottery-drop landing page with strong reservation flow, but several distinctive handmade asset and typography requirements are visibly absent.">61.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-042-blueprint-paper</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly structured blueprint-themed landing page fulfills the workflow and content brief, with the main weakness being visually blank recent-work artifacts.">91.43</td><td class="best" title="A polished and highly coherent blueprint-themed studio page with strong estimating and inquiry flows, weakened mainly by replacing exploded plate diagrams with standard process photography.">92.62</td><td title="A polished blueprint-style landing page delivers the estimate and quote workflow, though its narrow fixed canvas leaves extensive empty space and its recent-work imagery does not clearly show finished prints.">89.05</td><td title="A polished and methodical blueprint-style page with strong forms and production details, but it omits the required hero press image and does not visibly showcase finished print artifacts.">76.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-043-brutalist-graphic-design</td><td class="tag">A</td><td class="tag">landing</td><td title="A highly structured and actionable brutalist seed-library page is undermined by a visibly broken botanical plate, weak stamped-packet treatment, and incomplete seasonal coverage.">79.29</td><td class="best" title="A highly complete and visually coherent brutalist seed-library landing page whose main brief deviation is the repeated use of acid red outside the request-seeds block.">96.19</td><td title="A highly complete and coherent brutalist seed-library page with excellent structure and workflow, weakened mainly by nonexclusive red accents and missing pollinator-plant imagery.">91.43</td><td title="A highly complete and visually forceful brutalist seed-library page, weakened mainly by nonexclusive red usage and limited explicit vegetable imagery.">87.86</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-045-glassmorphism</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete beginner-focused bouldering landing page with excellent hierarchy and booking support, weakened only by a predominantly cool-gray background palette.">96.53</td><td title="A polished and highly usable glassmorphism landing page that covers nearly the full journey, with the main shortfalls being the overlaid hero composition and lack of repeated intro-class cards.">90.00</td><td title="A polished and comprehensive beginner-climbing landing page with strong glass-panel structure, weakened mainly by the missing hero session picker and absent coral hue.">83.20</td><td title="A polished and highly complete beginner-focused climbing page is weakened by the missing session picker, a broken café image, and an unrelated shoe-rental image.">83.81</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-046-lunarpunk</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly legible Lunarpunk ferry landing page delivers strong service information and booking structure, but omits the requested crew portraits and illustrated route posters.">73.57</td><td class="best" title="A polished, highly usable Lunarpunk ferry landing page with strong scheduling and booking support, weakened mainly by the photographic single poster and non-portrait crew imagery.">87.86</td><td title="A substantial, atmospheric ferry landing page with strong schedules and civic detail is undermined by an obscured hero and missing route and crew imagery.">78.57</td><td title="A visually cohesive and information-rich Lunarpunk ferry page is undermined by missing crew portraits and a visibly broken booking action.">86.43</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-047-chalkboard-slate</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, complete chalkboard-style gym landing page clearly presents the autumn routes, beginner journey, practical choices, and booking path.">100.00</td><td title="A polished and highly complete chalkboard gym landing page with excellent planning and booking cues, weakened mainly by the upright hero and missing explicit shoe-hire pass entry.">90.00</td><td title="A polished and highly faithful chalkboard landing page is weakened by missing setter profile cards and a visibly malformed horizontal pass board.">81.67</td><td title="The page strongly realizes the chalkboard climbing-gym style and booking journey, but its oversized pass board lacks the required priced dotted-leader list and its setters are rows rather than profile cards.">87.38</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-048-club-poster-ui</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, highly complete tattoo-event landing page with strong poster fidelity and booking clarity, held back mainly by duplicated booking CTAs and the green header CTA’s separation from the hero heading.">91.43</td><td title="A polished, urgent tattoo-event landing page with strong poster styling and booking closure, weakened by missing flash-sheet artwork, missing appointment-slot cards and duplicated green CTAs.">69.29</td><td title="A polished, urgent club-poster landing page with strong booking information and visual hierarchy, significantly weakened by the absence of actual tattoo flash artwork and by duplicate booking actions.">80.00</td><td title="A polished and highly structured event page with strong booking information, undermined by missing flash-design visuals, multiple pink CTAs, and no acid-green accent.">77.86</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-049-japandi</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, complete listening-room landing page with excellent restraint and hierarchy, weakened mainly by the absence of clearly visible photography for several specified materials and objects.">88.57</td><td title="A polished, listening-first Japandi page with strong structure and mood, limited mainly by missing coffee and curtain visuals and one visibly broken host image.">91.43</td><td class="best" title="A polished, complete listening-room landing page with strong Japandi atmosphere, weakened mainly by missing visible curtain and cup imagery and awkwardly narrow material-note text.">92.86</td><td class="best" title="A polished, cohesive listening-room landing page fulfills nearly all structural and mood requirements, with only linen curtains and ceramic cups absent from the visible photography.">92.86</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-050-leather-bound</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly complete collector-focused cartographic landing page succeeds editorially and functionally, with its main shortfall being several specifically requested visual artifacts.">90.41</td><td title="A polished, highly complete collector-focused landing page with excellent archival structure and mood, weakened mainly by missing or insufficiently recognizable requested cartographic imagery and objects.">86.33</td><td class="best" title="A polished, comprehensive scholarly cartography landing page fulfills nearly every visible requirement, with only the requested brass dividers not clearly shown.">97.96</td><td title="A polished, scholarly collector landing page with strong structure and content coverage is weakened by missing catalogue imagery, absent requested cartographic details, and unlabeled dark controls.">74.90</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-051-skate-mag</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, complete landing page visibly fulfills the record-shop content, transactional guidance, and gritty skate-mag aesthetic.">100.00</td><td title="A highly complete and visually faithful record-shop landing page is weakened mainly by the yellow rather than safety-orange CTA and two imperfect alignment details.">88.57</td><td title="A visually complete and highly on-brief record-shop landing page, weakened mainly by staggered staff-card alignment and yellow rather than safety-orange arrival buttons.">92.14</td><td class="best" title="A polished, complete landing page visibly delivers the requested basement-record-shop content, decision aids, and forceful skate-mag aesthetic.">100.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-052-armin-hofmann-contrast-and-reduction</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and strongly Hofmann-inspired museum landing page covers nearly all structural content, but lacks visible archive search and several specifically requested photographic subjects.">80.95</td><td title="A polished, strongly reduced museum page with useful visit and booking content, but it omits search, several visibly required photo subjects, modular tower sections, and the defining luminous beam treatment.">69.52</td><td title="A polished, highly restrained museum landing page with strong hierarchy and visitor information, weakened by the absent search control, beam motif, and several required photographic subjects.">75.71</td><td title="A polished monochrome museum landing page with strong reduction and visit flow, but it omits archive search, visible beam luminosity, and several specifically required photographic subjects.">75.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-053-cyberpunk</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly usable cyberpunk ramen landing page whose main weakness is the absence of the requested counter, ramen close-up, railway-track, and stainless-kitchen photography.">80.00</td><td class="best" title="A polished, highly functional cyberpunk ramen landing page with strong queue flow and information hierarchy, weakened mainly by missing required photographic scene types.">86.43</td><td title="A polished and information-rich cyberpunk ramen page with strong hierarchy and queue support, weakened by the non-red CTA and missing counter, railway-track, and stainless-kitchen imagery.">72.62</td><td title="A polished neon cyberpunk layout with strong queue and visit information is undermined by the green CTA and missing ramen, steam-counter, illustrated-menu, and stainless-kitchen visuals.">76.19</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-054-maritime-heritage-editorial</td><td class="tag">A</td><td class="tag">landing</td><td title="A highly cohesive and functionally complete maritime editorial landing page is held back mainly by incomplete coverage of the requested documentary photo subjects.">93.81</td><td title="A polished and highly complete maritime editorial landing page with excellent booking support, weakened mainly by absent crew-profile cards and no clearly visible deckhand photography.">89.78</td><td class="best" title="A highly complete and cohesive maritime editorial landing page connects strong documentary presentation with detailed service information and a visibly closed booking workflow.">100.00</td><td title="The polished archival ferry hero is visually strong, but the screenshot omits almost all required informational sections and the complete reservation workflow.">46.91</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-055-mid-century-modern</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and complete Mid-century Modern cinema landing page visibly fulfills the marquee, programming, cultural-content, staff, concessions, and membership requirements.">100.00</td><td title="A polished and highly complete cinema landing page with strong period styling, weakened only by the side-by-side showtime alignment and incorrect concessions-tag color.">92.86</td><td title="A polished and highly complete Mid-century cinema landing page, with the main visible shortcomings being a stacked film list instead of a true grid and an ambiguous dusk setting.">90.24</td><td title="A polished and highly faithful cinema landing page with strong hierarchy and membership closure, weakened mainly by incomplete repertory count, repeated imagery, and mostly non-mustard concessions tags.">95.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-056-canvas-fabric</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, cohesive and highly complete restoration landing page whose only clear brief-level omission is photography of an opened timber case.">95.00</td><td title="A highly cohesive and decision-ready restoration landing page that misses clear brass-movement and opened-case photography and lacks an unambiguous after plate.">88.41</td><td title="A polished, cohesive conservation-focused landing page fulfills nearly the entire brief, with the main visual omission being explicit engraved and brass movement imagery.">93.41</td><td class="best" title="A polished and highly complete conservation-focused landing page with strong decision flow, weakened only by the lack of clearly visible opened timber-case photography.">95.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-057-space-documentary</td><td class="tag">A</td><td class="tag">landing</td><td title="A highly polished forensic museum landing page covers the full visitor and preservation story, though its imagery centers on indoor rocket hardware rather than an actual launch or clearly identifiable satellite instruments.">76.19</td><td class="best" title="A polished and highly complete documentary landing page with strong technical styling, weakened mainly by artifact imagery that reads as repeated launch photography rather than decommissioned satellite instruments.">91.43</td><td title="A polished and complete cinematic museum landing page is weakened mainly by rocket-focused imagery and the absence of a clearly repeatable artifact collection grid.">81.90</td><td title="A highly polished, complete-feeling documentary landing page whose main visible weakness is that its artifact cards use launch imagery instead of photographs of decommissioned satellite instruments.">90.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-058-streetwear-drop</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete technical watch-drop page is weakened mainly by the orange-red CTA treatment and the lack of a clearly visible caseback macro.">89.05</td><td title="A polished, highly structured watch-drop page with strong technical and purchase content, weakened mainly by absent clear wrist photography and the incorrect reserve-button color.">87.14</td><td title="A polished, highly structured watch-drop page with strong technical storytelling and reservation flow, weakened by repetitive imagery, no photographic caseback detail, and the wrong primary accent color.">81.43</td><td title="A polished and highly coherent limited-watch drop page with strong hierarchy and conversion flow, weakened mainly by mislabeled or insufficient macro-detail photography.">82.38</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-059-art-nouveau</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly coherent bicycle-atelier landing page with strong catalogue, sizing, restoration, and booking flows, weakened mainly by missing rider imagery, specified macro subjects, and wheel-and-vine ornament.">82.86</td><td class="best" title="A polished and coherent atelier landing page with excellent structure and decision tools, weakened mainly by several missing required photographic and decorative assets.">82.86</td><td title="A polished, highly coherent atelier landing page with strong catalogue, sizing, comparison, and booking flows, weakened by missing specialty imagery, a broken restoration image, and a few text collisions.">76.05</td><td title="A cohesive and feature-rich bicycle-atelier page with strong vintage styling and booking flow, weakened by missing specified detail imagery, absent vine flourishes, and one visible text collision.">78.20</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-060-dieter-rams-silent-control-panel</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly cohesive control-panel landing page with excellent service and purchase detail, weakened mainly by missing diagrammatic threading visuals and incomplete requested photography coverage.">83.69</td><td class="best" title="A polished, highly complete silent-control-panel landing page whose main shortfall is the absence of several specifically requested sewing-material photographs.">89.05</td><td title="A polished, highly coherent technical product page with excellent decision support, weakened mainly by a non-control-plate opening hero, absent diagrams, and missing fabric-roll and metal-bobbin photography.">72.74</td><td title="A polished modular technical page with strong maintenance and parts content is substantially undermined by a visibly overlapping hero and missing requested photographic and diagram assets.">52.98</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-061-pop-art</td><td class="tag">A</td><td class="tag">landing</td><td title="A vivid, feature-complete Pop Art bouquet landing page with strong local ordering support, held back by missing cutout imagery and a conspicuously overlapping hero picker.">90.00</td><td class="best" title="A polished and highly complete Pop Art bouquet-service page whose main visible miss is the lack of true cutout floral photography.">93.81</td><td title="A polished and highly complete Pop Art flower-service landing page with strong commerce flow, weakened mainly by conventional rectangular photography and a vertically stacked calendar.">90.95</td><td title="A polished and highly complete Pop Art bouquet-service page with strong ordering support, weakened mainly by non-cutout photography, rectangular quote cards, and one clipped hero-edge element.">88.10</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-062-cottagecore</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete Cottagecore landing page with strong commerce details, weakened mainly by the absent sunrise cue, indistinct sweet peas and cosmos, and non-handwritten selector.">77.62</td><td class="best" title="A polished and highly complete Cottagecore flower-farm landing page, with the main visible misses being unmistakable sunrise imagery and genuinely handwritten selector lettering.">84.76</td><td title="A polished and content-rich Cottagecore page is undermined by a weak, sparsely composed hero that omits several signature visual requirements.">56.67</td><td title="A polished and strongly Cottagecore landing page with all major sections present, undermined by several failed images, an overlapping bouquet selector, and hero photography that does not deliver the specified sunrise scene.">64.76</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-063-luxury-fashion-film</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="The page is polished, structured, and commercially complete, but its pervasive city, ocean, and lifestyle photography fails to visibly present the cobalt porcelain collection.">78.57</td><td title="A polished and functional luxury editorial landing page whose major weakness is the absence of varied kiln-room, artisan, wet-glaze, and gold-rim imagery.">72.86</td><td title="A polished cinematic editorial page with strong structure and booking support, but it misses the kiln-room opening, restrained serif title, consistent cobalt pattern, wet-glaze macro, and separate edition reservation.">65.48</td><td class="best" title="A polished and largely complete luxury editorial landing page is undermined by absent kiln-room and glaze-macro imagery, inconsistent product patterns, and a partially obscured hero title.">78.57</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-064-op-art</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly complete Op Art bicycle landing page with excellent technical decision support, weakened mainly by the absence of workshop portraits.">96.53</td><td title="A polished and highly complete technical bicycle landing page that succeeds functionally but misses the brief’s defining Op Art hero treatment and workshop portrait imagery.">74.73</td><td class="best" title="The screenshot presents a polished, comprehensive fixed-gear bicycle landing page with strong technical hierarchy, Op Art energy, and a clear reservation endpoint.">100.00</td><td title="A polished and highly complete technical product page with strong decision support and reservation closure, but it omits the brief&#x27;s defining bicycle photography and concentric Op Art motif.">71.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-066-color-field</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, cohesive cinematic landing page fulfills nearly all requested content and interaction cues, though it is not single-viewport and lacks clearly visible film grain or page-wide ultramarine dominance.">88.10</td><td title="A polished cinematic page supplies strong screening and membership details, but its red hero, inset image, missing projectionist note and ticket seam, and oversized blank bands substantially depart from the brief.">62.93</td><td title="A polished and complete cinematic landing page with strong hierarchy and atmosphere, weakened by the absent full-bleed image band and limited use of ultramarine.">82.35</td><td title="The page delivers a cohesive cinematic banded narrative with all key content, but misses the defining ultramarine direction and contains several visible layout and legibility defects.">77.48</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-067-wabi-sabi</td><td class="tag">A</td><td class="tag">landing</td><td title="A highly complete, cohesive bakery landing page with excellent content hierarchy and reservation support, though the hero image does not visually prove a dawn setting.">95.24</td><td title="A polished and complete Wabi-Sabi bakery landing page fulfills nearly every visible requirement, with only literal local-grain imagery absent.">95.24</td><td class="best" title="A polished, cohesive bakery landing page visibly fulfills all requested sections and provides a clear path from story and availability to pickup selection and pre-order.">100.00</td><td title="A polished, cohesive bakery landing page with all major sections and a clear preorder path, weakened mainly by clipped loaf cards, a six-day calendar, and absent explicit grain imagery.">85.95</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-068-dreamcore</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A highly complete and visually cohesive Dreamcore planetarium page, with only the understated venue title and non-diary journal typography missing strict criteria.">92.86</td><td title="A polished and highly atmospheric Dreamcore landing page with strong scheduling and booking content, weakened by a hero-obscuring map panel and no visible astronomy workshop offering.">78.10</td><td title="A polished and highly complete Dreamcore planetarium landing page, with the main visible weakness being the lack of a dominant full “Moonwell Planetarium” title.">91.67</td><td title="This is a polished and highly complete Dreamcore planetarium page, with its main visible weaknesses being a non-dominant venue name, conventional journal typography, and repeated rather than sequential lunar-phase icons.">81.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-069-goblincore</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete moth-watch landing page with strong hierarchy and atmosphere, weakened mainly by missing lamp, linen, weathering, and authentic handwritten details.">75.71</td><td title="A polished and highly complete moth-watch landing page with strong structure and mood, though several tactile brief details are not visibly realized.">73.57</td><td class="best" title="A polished and highly coherent moth-watch landing page fulfills most of the brief, with only the absent linen imagery and insufficiently handwritten logs weakening completeness.">87.86</td><td title="A structured and atmospheric goblincore page with strong section coverage and reservation closure is undermined by severe dark-on-dark legibility and missing moth, lamp, weather-stain, and handwritten imagery.">65.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-070-susan-kare-pixel-icon-clarity</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and coherent pixel-weather landing page covers the family forecast feature set well, but omits umbrella and sunny-park photography and repeats the specified single CTA.">89.39</td><td title="A polished, highly legible pixel-weather landing page with comprehensive family planning UI, weakened mainly by the absence of most requested lifestyle photo categories.">87.14</td><td title="A polished and coherent pixel-weather landing page with strong family planning components, but an empty hourly experience, duplicated primary CTA, and major photographic omissions limit full brief compliance.">74.01</td><td title="The page delivers a strong family-weather feature flow and retro pixel-influenced UI, but its broken hero composition and nearly complete absence of the requested photographic assets are conspicuous weaknesses.">76.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-071-logistics-industrial</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete grain-elevator landing page with excellent operational clarity, weakened mainly by the lack of visible painted bin numbers and clearly identifiable grain trucks.">81.43</td><td class="best" title="A polished and highly functional industrial logistics landing page that fulfills nearly all requirements, with painted silo numbering and a dedicated grade table visibly absent.">86.19</td><td title="A polished and highly coherent industrial landing page with strong unloading workflow and styling, weakened by visibly incomplete market, grade, moisture, and time-slot data.">69.76</td><td title="A polished and highly functional industrial grain-elevator landing page with strong decision and booking flows, weakened mainly by missing painted silo numbers and unclear capacity emphasis.">79.05</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-072-steampunk</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and comprehensive steampunk observatory landing page, with only the survey artwork lacking clearly hand-drawn marks.">96.43</td><td title="A polished and highly atmospheric steampunk observatory landing page with strong content and booking closure, weakened mainly by the missing hero dome photograph and lack of a true moon-phase calendar.">80.24</td><td title="A polished and highly coherent steampunk observatory landing page is weakened mainly by the absent hero dome photograph and an underdeveloped moon calendar and survey illustration.">72.38</td><td title="A highly complete and visually coherent steampunk observatory landing page, with the only clear frozen-check miss being the side-by-side rather than stacked hero title and image.">95.24</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-073-luxury</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete luxury alpine landing page whose only clear brief-level visual omission is visible bath steam.">95.00</td><td title="A polished and complete luxury alpine retreat page with excellent editorial structure and conversion flow, weakened mainly by missing steam-bath and cashmere visual cues.">91.43</td><td title="A polished and complete-feeling alpine luxury landing page with excellent hierarchy and conversion flow, weakened mainly by absent steam-bath and cashmere visuals.">90.00</td><td title="A polished, complete alpine-luxury landing page with strong editorial structure and workflow, weakened mainly by the absence of visible steaming baths and cashmere texture.">90.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-074-solarpunk</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly structured solarpunk ferry page delivers nearly all functional content, with the main weakness being incomplete photographic coverage of the requested daily-life scenes.">81.67</td><td class="best" title="A polished, highly structured solarpunk ferry landing page with excellent service information and planning flow, weakened mainly by missing requested lifestyle photography.">82.86</td><td title="A polished, highly structured ferry landing page delivers strong transit information and cooperative branding, but misses several specified photographic scenes and the requested sea-green planner treatment.">68.57</td><td title="A polished and highly complete poster-grid ferry landing page is weakened mainly by missing cyclist and garden photography and an insufficiently distinct sea-green trip-planning block.">78.10</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-076-massimo-vignelli-typographic-discipline</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly legible public-radio guide with strong typographic and information design, weakened only by the absence of transmission-tower and portable-radio photography.">93.81</td><td class="best" title="A highly polished, grid-disciplined public-radio guide fulfills nearly every visual requirement, with duplicated Listen Live calls as the only notable mismatch.">96.79</td><td title="A highly polished, grid-driven public-radio guide fulfills nearly all functional and stylistic requirements, with the main shortfall being absent control-room, tower, and portable-radio imagery.">91.43</td><td title="A highly structured and authoritative radio guide with strong typographic fidelity, undermined mainly by schedule collisions, duplicate live CTAs and missing tower and portable-radio imagery.">89.17</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-077-material-you-expressive</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete Trailkind landing page with strong trail-planning hierarchy, missing only visible landmark-card content in the bottom sheet.">96.19</td><td title="A polished and welcoming hiking landing page fulfills most of the brief, with the main visual omissions being substantive elevation graphics and a clearly recognizable offline map.">81.90</td><td title="A polished and informative hiking landing page strongly covers preparation and visual tone, but misses the brief’s key elevation, landmark, and lower-thumb CTA details.">73.33</td><td title="A polished and informative landing page with strong safety and access content, but its core live-trail viewport is visibly incomplete and lacks the required primary hike action.">49.29</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-078-vaporwave</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete vaporwave karaoke landing page with strong booking support, weakened mainly by the non-stacked hero arrangement and lack of visible snack content.">80.20</td><td title="A polished and highly complete vaporwave karaoke page with strong booking support, weakened mainly by the non-stacked first viewport and the lack of clearly themed songbook choices.">84.97</td><td class="best" title="A polished and content-complete vaporwave karaoke landing page with strong booking support, though its asymmetric opening and restrained mood weaken the requested vertical rhythm and euphoria.">87.62</td><td title="A polished and content-rich vaporwave landing page with strong booking closure, but its first viewport misses most of the specifically required stacked visual and interface elements.">71.05</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-079-kidcore</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete kidcore landing page with strong hierarchy and decision support, with only the planner tabs positioned beside rather than directly above the planner.">96.43</td><td class="best" title="A polished, comprehensive kidcore landing page fulfills the brief strongly, with only the age-and-space tabs positioned separately from the room planner.">96.43</td><td title="A polished and highly complete kidcore product page is undermined mainly by the conspicuous expanded header menu and tabs positioned beside rather than above the planner.">95.00</td><td title="A polished and highly complete kidcore landing page, with only the side-positioned tabs and undersized final CTA label weakening exact rubric compliance.">91.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-081-frutiger-aero</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, airy sunglasses storefront covers most required commerce and information modules, but it visibly lacks the mountain hero photography and multi-image trail durability gallery.">83.81</td><td title="A polished and highly complete Frutiger Aero storefront with strong product-selection flow, weakened mainly by the missing cloud-bright mountain hero photography.">93.81</td><td class="best" title="The screenshot presents a polished, complete Frutiger Aero eyewear landing page with strong product hierarchy, comprehensive decision support, and a clear purchase endpoint.">100.00</td><td title="A polished and highly complete sunny eyewear landing page with strong commerce flow, weakened mainly by the absence of mountain and genuinely trail-tested imagery.">85.24</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-082-laszlo-moholy-nagy-light-and-new-media</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and complete modernist optometry landing page with strong booking closure, though explicit trial-lens imagery and page-wide translucent plate repetition are not clearly visible.">92.14</td><td title="A polished and highly legible modernist optometry page with strong content and booking closure, weakened mainly by missing appointment steps and conventional glasses imagery.">85.48</td><td class="best" title="A highly polished and comprehensive modernist optometry landing page, with the only notable visual omission being clear conventional eyewear photography.">92.86</td><td title="A polished, lucid modernist optometry page delivers most major modules and a usable booking endpoint, but lacks appointment steps, trial-lens and frame photography, glasses-wearer imagery, and consistently translucent plates.">82.62</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-083-legal-authority</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, authoritative legal landing page with strong process detail and workflow closure, weakened mainly by the missing repeated cartographic zoning maps and a localized hero text collision.">87.38</td><td title="A polished and highly usable legal landing page with strong hierarchy and procedural detail, weakened only by the single map presentation and absent citation pin icons.">90.24</td><td class="best" title="A polished and comprehensive legal-authority landing page with strong hierarchy and procedural clarity, weakened only by the absence of visible pin icons on citations.">95.24</td><td title="A polished and highly complete legal landing page whose principal visual shortfall is substituting annotated streetscape photography for repeated zoning-map graphics and pin icons.">84.05</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-084-persian-islamic-geometric</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and comprehensive Persian-geometric festival page is weakened mainly by its conventional event grid and the lack of clearly identifiable hyacinths or candlelight in the hero.">86.31</td><td title="A polished and comprehensive Persian-geometric festival page is weakened mainly by the missing gold rosette, non-mirrored hero imagery, and rectangular rather than star-grid event layout.">70.71</td><td title="The page is polished and content-rich, but it omits the brief’s defining gold rosette, clear mirrored haft-sin hero imagery, star-grid events, and convincing maps.">50.83</td><td title="The page is polished and content-rich with strong booking closure, but its hero misses the brief’s defining centered rosette, mirrored photography, and date-selector composition.">61.31</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-085-quiet-luxury</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, complete ceramic-studio landing page closely matches the brief through restrained styling, tactile imagery, detailed vessel information, and a discreet commission pathway.">100.00</td><td title="A polished, highly complete quiet-luxury ceramic archive page with strong editorial structure and tactile detail, weakened only by the absence of clearly visible kiln-shelf photography.">95.00</td><td title="A polished and unusually complete quiet-luxury ceramic landing page, with the main visible misses being the non-serif title and unclear kiln-shelf imagery.">86.43</td><td title="A polished and highly faithful ceramic-studio landing page with strong archive, editorial, care, and commission content, weakened mainly by the lack of clearly visible kiln shelves.">95.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-086-skeuomorphic-premium</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, highly credible premium bladesmith landing page with strong technical and service detail, weakened mainly by unclear walnut imagery, absent sequential waitlist positions, and a split hero alignment.">85.00</td><td title="A highly polished premium bladesmith landing page fulfills almost the entire brief, with the main weakness being the absence of a distinct multi-testimonial section and understated waitlist numerals.">91.43</td><td class="best" title="A polished, cohesive premium bladesmith landing page visibly fulfills the requested product story, technical detail, credibility, services, and numbered waitlist.">100.00</td><td title="A polished, convincingly premium bladesmith landing page with strong structure and conversion flow, weakened by incomplete geometry visuals, unclear material imagery, and no sequentially numbered waitlist.">78.57</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-088-gothic-revival</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete Gothic foundry landing page with strong workflows and atmosphere, weakened mainly by missing inscription, rafter, and hoisted-bell photography.">86.43</td><td title="A polished and highly coherent Gothic foundry landing page with complete service workflows, weakened mainly by missing several required photographic motifs.">85.71</td><td class="best" title="A polished, highly complete Gothic foundry landing page with strong workflows and atmosphere, weakened mainly by missing inscription, rafter, and chain-hoist photography.">90.00</td><td class="best" title="A polished and highly complete Gothic foundry landing page with strong catalogue and inquiry flows, weakened mainly by missing explicit imagery of inscriptions, rafters, and hoisted finished bells.">90.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-089-neumorphism</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete neumorphic coffee landing page with strong hierarchy and purchase support, weakened mainly by the absence of freshly roasted bean photography.">92.38</td><td class="best" title="A polished and highly complete neumorphic coffee landing page with strong selection guidance, weakened only by the absence of clearly visible roasted-bean photography.">93.81</td><td title="A polished and complete neumorphic coffee landing page with strong selection flow, weakened mainly by the purple CTA, non-recessed dial appearance, and missing roasted-bean imagery.">88.20</td><td title="A strong, tactile neumorphic coffee landing page with a complete buying flow, weakened by detached roast-dial placement, absent bean imagery, and poorly wrapped brew-guide text.">81.77</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-091-watercolor-wash</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete pottery landing page with strong tactile imagery and structure, weakened mainly by typography mismatches and duplicate preorder links.">82.86</td><td title="A polished, cohesive pottery landing page covers nearly all requested content, but misses the torn-edge hero treatment and duplicates the intended single preorder link.">80.00</td><td title="The rendered top and preorder close are polished and cohesive, but the enormous empty middle and omitted diary, photography, care, and visit content make the landing page substantially incomplete.">64.29</td><td title="A cohesive and content-complete pottery landing page with strong tactile styling, weakened by a diagonally composed hero and several missing signature brief details.">67.86</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-093-grunge</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished, complete analog-lab landing page with strong tactile darkroom styling and workflows, weakened mainly by missing explicit scratched-negative, stained-contact-sheet, and harsh-flash portrait imagery.">82.86</td><td class="best" title="A polished and comprehensive grunge darkroom landing page is shown, though clearly recognizable chemical contact sheets and scratched negative-strip assets are absent.">92.86</td><td title="A polished and highly complete analog-lab landing page with excellent information architecture and tactile styling, though several specifically requested grunge asset treatments remain visually ambiguous.">84.29</td><td title="A polished and highly structured analog-lab page with strong red-room atmosphere and useful service flows, but it omits several brief-specific visual assets and lacks a clearly rust-red final control.">67.86</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-094-josef-mueller-brockmann-swiss-grid-rhythm</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly coherent Swiss-grid climate archive landing page is weakened mainly by missing real imagery of clouds, radar screens, and meteorologists.">88.57</td><td title="A highly cohesive and functional Swiss-grid climate archive landing page is weakened mainly by the absence of cloud, radar-screen, and meteorologist photography.">90.00</td><td title="A strong Swiss-inspired observatory shell with useful tables and a clear download CTA is substantially undermined by a catastrophically malformed archive section and missing required content.">55.44</td><td class="best" title="A highly cohesive Swiss-grid climate archive with strong data organization and download closure, weakened mainly by missing radar-screen and meteorologist imagery and the absence of a temperature table.">90.82</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-095-velvet-luxury</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly coherent luxury piano landing page includes the full decision and appointment journey, with visible weaknesses limited to the unshown provenance interaction and absent close felted-action imagery.">80.20</td><td title="A polished and highly coherent luxury piano landing page fulfills nearly all visible requirements, with only the hover behavior and felted-action imagery unsupported by the screenshot.">87.35</td><td class="best" title="A polished and highly cohesive luxury piano landing page fulfills nearly every visual requirement, with the main shortfall being no visible evidence of the specified specimen hover reveals.">93.06</td><td title="A highly polished and complete luxury piano landing page is weakened only by absent felted-action close-ups and unverified hover-specific behavior.">87.35</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-096-industrial</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A cohesive, highly finished industrial salvage landing page visibly covers the requested inventory, traceability, decision tools, and pickup workflow.">100.00</td><td title="A polished and highly faithful industrial salvage landing page with excellent operational detail, weakened mainly by its table-style, largely text-only inventory presentation.">80.71</td><td title="A polished and highly legible industrial salvage landing page with excellent logistics and pickup support, weakened mainly by replacing visual product cards and dimension plates with a text-heavy ledger.">85.00</td><td class="best" title="A polished, highly complete industrial salvage landing page with strong traceability, decision support, and pickup workflow.">100.00</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-098-impressionism</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete sailing-school landing page with strong decision and booking support, weakened mainly by absent instructor portraits, unclear white sails, and limited Impressionist character.">87.96</td><td title="A polished and highly usable coastal sailing page delivers strong lesson planning and booking support, while missing boat-class content, genuine instructor portraits, and a clearly Impressionist visual treatment.">74.86</td><td title="A polished, comprehensive sailing-school landing page with excellent decision and booking structure, weakened mainly by missing human portraits, absent visible white sails, and limited Impressionist styling.">83.20</td><td class="best" title="A polished and comprehensive coastal booking page is weakened by a broken instructor image and limited visual fidelity to the requested Impressionist style.">92.72</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-099-denis-villeneuve-sci-fi</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly cohesive large-format cinema landing page with strong technical storytelling, weakened mainly by the absence of a clearly visible auditorium still and a program that is not the page&#x27;s top visual hierarchy.">87.86</td><td class="best" title="A highly cohesive and functionally complete large-format cinema landing page whose principal visible omission is the requested auditorium imagery.">91.43</td><td class="best" title="A polished, cohesive large-format cinema landing page fulfills nearly the entire brief, with the notable visual omission of an actual dust-lit projection booth.">91.43</td><td title="A polished, cohesive large-format cinema landing page with strong technical atmosphere, weakened by four visible screenings under a seven-screening claim and no clear auditorium imagery.">86.43</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-100-dada-collage</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, energetic seed-library landing page covers the required content well, though its structured neo-brutalist layout and missing handmade archival assets weaken full Dada-collage fidelity.">77.86</td><td title="A polished and highly complete seed-library landing page with strong practical workflows, but its orderly brutalist editorial style fundamentally misses the requested layered Dada collage aesthetic.">69.29</td><td title="A polished and functionally comprehensive seed-library landing page is visible, but its orderly editorial style substantially misses the requested Dada collage identity.">59.29</td><td title="A polished, lively seed-library landing page covers the requested workflows well, while only partially realizing the brief&#x27;s specific archival Dada artifacts.">75.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-101-fairycore</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete celestial retreat landing page with strong booking support, weakened mainly by missing pressed-fern ticket art and antique constellation-chart assets.">85.48</td><td title="A polished, coherent observatory landing page with strong booking support and celestial atmosphere, weakened mainly by the absence of antique constellation charts and pressed-fern ticket imagery.">84.05</td><td title="A polished and complete-feeling stargazing retreat page with strong hierarchy and booking support, weakened by the missing pressed-fern ticket flow and limited fairycore detailing.">74.05</td><td title="A polished twilight-retreat landing page with a coherent booking journey, weakened by missing signature fairycore assets and visibly broken information-panel text layout.">74.05</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-102-weirdcore</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and unusually complete Weirdcore storage landing page whose main visible miss is replacing the requested endless-corridor hero with a close-up door photograph.">87.86</td><td title="A polished and complete Weirdcore storage landing page with strong hierarchy and workflow, though Unit 33 is not a dominant headline and availability lacks the requested mint glow.">85.71</td><td title="A polished and highly complete storage landing page with strong liminal atmosphere, weakened mainly by the stacked hero-picker layout and conventional noticeboard typography.">84.29</td><td title="A polished and complete dark Weirdcore storage page is chiefly weakened by an obscured hero photograph, unreadable door numbering, and the absence of mint-glowing availability states.">77.14</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-103-ikko-tanaka-tradition-as-geometry</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and cohesive heritage-rail landing page with excellent journey structure and geometric styling, weakened mainly by the route map’s secondary hierarchy and several missing requested photo subjects.">79.05</td><td title="A polished, cohesive heritage-rail landing page with strong geometry and booking closure, weakened chiefly by the absent route map and missing paper-ticket and bento photography.">78.10</td><td class="best" title="A cohesive and highly finished heritage-rail landing page with strong journey flow and reservation closure, weakened mainly by missing ticket and bento photos plus localized text and map-label collisions.">86.67</td><td title="A polished and cohesive heritage-rail landing page with strong geometric styling and reservation closure, weakened by a subordinate route map and missing meal, ticket, and bento content.">75.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-104-gorpcore-outdoor-tech</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and cohesive outdoor-tech landing page fulfills nearly the entire brief, with the main visible misses being a slider instead of a dial and non-stenciled battery typography.">83.81</td><td title="A polished, cohesive Gorpcore landing page fulfills the product and pre-order story strongly, with the main visual shortfall being the lack of explicit comparative beam-pattern graphics.">86.67</td><td title="A highly cohesive and complete outdoor-tech landing page with strong technical storytelling, weakened only by the absent stencil lettering and ambiguous rain-soaked imagery.">88.10</td><td class="best" title="A polished, cohesive outdoor-tech landing page fulfills nearly the entire brief, with the main visible shortfall being the absence of an unmistakable stencil treatment for battery specifications.">93.81</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-105-holographic-foil</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, complete holographic ski-goggle launch page clearly presents the product, technical evidence, selection tools, and preorder path.">100.00</td><td title="A polished and comprehensive technical ski-goggle landing page with strong spectral styling, weakened mainly by a static, untilted hero presentation.">91.43</td><td title="A polished and technically detailed spectral-goggle landing page covers the full product journey, though its front-on abstract hero and close-up field imagery do not clearly show the requested tilt or alpine terrain.">87.86</td><td title="A polished and complete technical preorder page with strong spectral styling, weakened mainly by an abstract, untilted hero and field imagery that does not clearly show alpine terrain.">84.29</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-106-mermaidcore</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete coastal pottery landing page with strong structure and usability, weakened mainly by generic repeated wheel imagery and missing signature ocean-form assets.">77.38</td><td class="best" title="A polished and highly complete coastal ceramics landing page, with the main visible deviations being the below-hero shop placement and non-handwritten field-note typography.">91.67</td><td title="A polished and comprehensive coastal ceramics landing page with excellent deep-water craft styling, weakened mainly by sparse literal marine-form imagery and the collection sitting below rather than beside the hero.">83.57</td><td title="A polished and highly cohesive Mermaidcore studio page with complete editorial sections, though finished salt-glazed product imagery and conversion actions are underdeveloped.">85.95</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-107-neo-brutalism</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, content-complete neo-brutalist conservatory landing page with strong hierarchy, useful visitor information, and clear conversion paths.">100.00</td><td title="A polished and highly complete neo-brutalist conservatory page with strong information architecture, weakened mainly by the missing acid-green CTA, dedicated workshop section, and plant-index photography.">83.57</td><td title="A highly complete and visually convincing neo-brutalist conservatory page, with the only clear shortfall being the lack of photography inside the dedicated plant-index section.">96.43</td><td title="A polished, comprehensive neo-brutalist conservatory page with strong content hierarchy and booking closure, weakened mainly by the missing acid-green CTA and unclear sun-bleached photo treatment.">90.71</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-108-cubism</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete Cubist workshop landing page with strong decision support and booking closure, though the selector sits below the hero photo and does not clearly show multiple oriented build plates.">91.77</td><td class="best" title="A highly complete and visually faithful Cubist workshop landing page, with the main localized miss being that the geometry selector is below rather than beside the cycling photography.">95.24</td><td title="A polished and highly complete workshop landing page with strong technical decision support, weakened mainly by boxed repetitive imagery and limited visible selector/profile variation.">85.92</td><td title="A polished and complete workshop landing page with strong technical poster styling, weakened mainly by an under-informative hero and insufficiently fragmented photography.">67.38</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-109-manga-panel-ui</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and complete manga-inspired keyboard landing page, with the main visible misses being non-side-by-side switch comparison, two pre-order controls, and no switches or keycaps crossing panel borders.">91.07</td><td title="A polished and highly complete manga-styled keyboard landing page is weakened mainly by duplicated pre-order CTAs and the absence of true border-breaking switch, keycap, and keystroke effects.">84.64</td><td title="A polished and information-rich keyboard landing page with strong manga styling, but it misses the brief’s defining staggered burst-panel hero and single-CTA constraint.">73.10</td><td title="A polished and complete manga-inspired product page is visible, but it misses the requested dynamic hero bursts, hand-lettered keystroke effects, and single crimson pre-order button.">76.67</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-110-blackout-xerox-punk</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly coherent Xerox-punk race landing page covers most content and commerce needs, but lacks several required visual asset types and does not visibly demonstrate populated race-card states.">69.76</td><td title="A polished and unusually complete Xerox-punk race landing page is weakened by missing crew, portrait, and poster imagery and by an unpopulated race-card state.">69.76</td><td title="A polished and highly complete Xerox-punk race landing page is weakened by missing requested visual categories, an unpopulated race card, and red accents used beyond actions and cautions.">60.24</td><td class="best" title="A strong, information-rich Xerox-punk landing page with clear purchase structure, but blank media regions, missing required people and poster visuals, and an unpopulated race card prevent full completion.">70.24</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-111-cyberdelic</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and cohesive cyberdelic aquarium landing page fulfills nearly the entire brief, with the main visible omissions being prominent hero branding and coral photography.">86.67</td><td title="A polished and highly cohesive cyberdelic aquarium landing page fulfills nearly every content and conversion need, with the main visible omissions being coral photography and hero integration of the exhibit-map rail.">84.29</td><td class="best" title="A highly complete and cohesive cyberdelic aquarium landing page whose only clear weakness is the lack of prominent Neon Tides branding within the hero.">95.24</td><td title="A cohesive and substantially complete cyberdelic aquarium landing page is weakened by unreadable dark card titles and the absence of clearly visible coral photography.">83.81</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-112-matte-ceramic</td><td class="tag">A</td><td class="tag">landing</td><td title="A polished and highly complete tactile stationery landing page, with only the selector’s beside-hero placement and unclear linen binding in the hero image missing strict visual confirmation.">93.57</td><td title="A polished, cohesive stationery landing page covers the requested journey well, with localized weaknesses in the hero product imagery, selector sequencing, and specimen typography.">92.14</td><td class="best" title="A polished, cohesive landing page visibly fulfills the complete stationery-studio brief with strong product guidance and tactile editorial styling.">100.00</td><td title="A polished and cohesive tactile stationery landing page fulfills nearly the entire brief, with blank page-layout previews as its one conspicuous weakness.">93.57</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-113-studio-ghibli-whimsy</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished and highly complete clock-repair landing page with strong information design and warm craft mood, weakened mainly by missing requested imagery and limited Studio Ghibli whimsy.">83.81</td><td class="best" title="A polished and highly complete heritage clock-repair landing page with strong information design, but limited Studio Ghibli identity and no visible pendulum, weathered-tool, or Bath lane imagery.">83.81</td><td title="A polished and highly complete clock-repair landing page with excellent hierarchy and workflow, but it misses the defining Ghibli whimsy, Bath lane imagery, and visible pendulums.">81.67</td><td title="A polished and highly complete clock-repair landing page with strong decision support and booking closure, weakened mainly by missing location and pendulum imagery and limited Ghibli-specific whimsy.">83.10</td></tr><tr data-track="A" data-surface="landing"><td class="cid">landing-114-futurism</td><td class="tag">A</td><td class="tag">landing</td><td class="best" title="A polished, complete futurist-industrial landing page visibly delivers every requested content section and closes with a prominent demo workflow.">100.00</td><td title="A polished and highly complete industrial-robotics landing page, with only the motion-path graphic lacking clear alignment to the product photography.">96.43</td><td class="best" title="A polished and complete futurist-industrial landing page clearly presents the modular workcell, substantiates its practical value, and closes with strong pricing and demo actions.">100.00</td><td title="A polished and comprehensive industrial-robotics landing page with strong decision support and conversion flow, weakened mainly by the missing diagonal motion-path graphics.">90.24</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-001-src-368</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A polished, decision-ready cold-chain dashboard with strong prioritization and evidence, weakened mainly by the missing corrective-action control and limited trailer/depot detail.">69.93</td><td class="best" title="A polished and highly scannable cold-chain dashboard prioritizes the right interventions, but obscured mid-page content and missing explicit route-resolution feedback prevent full completion.">92.24</td><td title="A polished and decision-ready cold-chain dashboard covers the main intervention workflow well, with the notable omissions of an exact Asterline Coldchain heading and a corrective-action logging control.">91.36</td><td title="A polished and calm operations dashboard with strong alert actions, but its incomplete ranked queue, absent reefer labeling, and missing stale and resolution states limit full readiness.">70.75</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-002-src-370</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A strong, decision-oriented pacing dashboard covers the investigation well, but the visible workflow lacks clear pause and audit-note closure and shows no creative previews.">86.43</td><td title="A polished, decision-focused pacing dashboard is weakened by missing creative previews, no explicit pause control, and an unfinished creative-evidence area.">85.36</td><td class="best" title="A polished, decision-ready pacing dashboard with strong hierarchy and operational coverage, weakened mainly by the missing audit-note field and lack of actual creative thumbnails.">89.64</td><td title="A cohesive, calm technical dashboard with strong investigation hierarchy, weakened mainly by unclear channel plotting and missing visible creative previews and audit-note input.">88.21</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-003-src-371</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished and decision-ready SRE dashboard covers the requested investigation workflow comprehensively, with explicit stale-state feedback as the only notable omission.">93.81</td><td title="The polished monitoring shell and summary metrics render clearly, but missing chart, dependency, service, deployment, and incident-row data make the dashboard operationally unusable.">50.83</td><td title="The dashboard strongly communicates the affected service, dependency, region, and deployment, but lacks paging, stale/completion feedback, minute-level update evidence, and an aligned deployment marker on the latency chart.">73.09</td><td title="The styled dashboard communicates one checkout-api incident clearly, but missing service comparison, empty operational visuals, absent paging, and weak state feedback prevent reliable incident-command use.">45.71</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-004-src-373</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished and operationally complete treasury dashboard with a strong cash-to-action flow, weakened mainly by clipped details and inefficient mid-page spacing.">90.00</td><td title="A strong, action-oriented treasury dashboard with broad operational coverage is visibly compromised by evidence-card text collisions, truncated account labels, and one incomplete-looking transfer field.">81.67</td><td title="The dashboard strongly identifies today’s funding issue and supporting evidence, but the visible workflow stops before transfer staging and approval, with notable table clipping and layout imbalance.">76.02</td><td title="The dashboard has strong treasury structure and restrained styling, but widespread clipping, an unreadable ledger, and an off-screen transfer workflow prevent dependable operation.">68.91</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-005-src-375</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished and decision-ready retention workbench meets nearly all visible requirements, with the main gap being the incomplete analysis navigator.">95.00</td><td title="A polished and actionable retention dashboard is weakened by the missing daily three-week analysis view and one material cohort-drop inconsistency.">85.24</td><td title="The dashboard shell is polished and actionable, but blank core analysis panels make the central retention decision unsupported.">59.44</td><td title="The styled KPI summary and experiment action are visible, but major required analysis content is missing and the only chart/action region is visibly overlapped.">57.40</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-006-src-376</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished and highly complete risk-operations dashboard visibly supports shift monitoring, evidence-led investigation, decisive case action, and audit preservation.">100.00</td><td title="The dashboard strongly prioritizes case review and provides rich evidence and actions, but cramped overflow, truncated details, incomplete entity coverage, and the missing preservation control prevent full fulfillment.">79.76</td><td title="The dashboard delivers strong operational coverage and a credible risk-console aesthetic, but its chart-heavy layout and cramped, partially clipped action area reduce decision efficiency.">89.52</td><td title="A calm and legible investigation dashboard supports case disposition well, but missing action controls and blank analytical panels prevent complete operational closure.">85.00</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-007-src-378</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="The screenshot presents a polished, complete claims-supervision dashboard with strong urgency prioritization, comprehensive operational context, and clear action controls.">100.00</td><td title="A strong calm claims-operations dashboard with clear priority actions, weakened mainly by missing expert-review signaling and limited comparison data in the prioritized table.">96.04</td><td title="A strong, calm supervisory claims dashboard clearly prioritizes the urgent high-reserve case, with localized shortcomings in expert-review visibility, live updates, and KPI spacing.">85.27</td><td title="The dashboard is professionally styled and shows core KPIs, but its empty charts, empty priority queue, absent deadline context, and missing action controls make it operationally incomplete.">43.92</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-008-src-380</td><td class="tag">B</td><td class="tag">dashboard</td><td title="The dashboard strongly communicates attack-path priority and operational breadth, but diagram overlap, clipped fields, and missing explicit stale and resolution states keep it from full visual completion.">89.76</td><td class="best" title="A polished and decision-focused SOC dashboard that strongly prioritizes the attack path, with the main weakness being the absent explicit timeline-attachment control and a few non-exact labels.">95.00</td><td title="A polished and decision-focused SOC dashboard clearly prioritizes the attack path, but it omits the required containment, host-isolation, and timeline-attachment controls.">89.76</td><td title="The dashboard presents a strong technical operations shell and clear attack priority, but blank evidence space, missing state closure, and clipped lower controls prevent full visual and operational readiness.">73.81</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-009-src-382</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A highly complete and decision-focused fulfillment dashboard with strong operational structure, undermined only by localized text clipping and overlap in dense areas.">91.43</td><td class="best" title="A polished and comprehensive cutoff-control dashboard with strong workflow closure, though the visible Zone A priority block and Zone B rebalance recommendation weaken decision coherence.">98.57</td><td title="A polished and decision-focused fulfillment dashboard covers the required workflow well, with the only clear gap being the absence of an explicit five-minute station-update indicator.">93.81</td><td title="The dashboard has a strong calm cutoff-focused shell, but missing chart data, exception rows, worker rebalancing, and readiness confirmation leave the operational workflow incomplete.">79.52</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-010-src-383</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A calm and operationally comprehensive liquidity dashboard is undermined by a localized but conspicuous overlap in the breach-investigation panel.">73.33</td><td class="best" title="A polished and decision-ready liquidity dashboard with strong operational coverage, weakened mainly by the absent rationale input and lack of an explicit breach icon.">77.74</td><td title="The dashboard is operationally comprehensive and visually calm, but poor KPI contrast and incomplete scenario, rule, stale-state, and alert-symbol evidence reduce decision confidence.">56.19</td><td title="The dashboard strongly prioritizes today’s stressed funding decision and covers the monitoring scope, but evidence-panel crowding and an incomplete approval-and-feedback workflow prevent full operational closure.">68.09</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-011-src-385</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A polished and decision-ready greenhouse dashboard with strong operational linkage, weakened mainly by missing ten-minute trend markers, greenhouse visuals, and an explicit outcome-logging control.">81.69</td><td class="best" title="A polished and decision-ready greenhouse operations dashboard meets nearly all visible requirements, with only the requested greenhouse-house visuals absent.">85.71</td><td title="A highly polished and decision-ready greenhouse dashboard with strong A1 exception linkage, weakened mainly by absent greenhouse visuals, sensor or batch detail, and visible commit actions.">82.86</td><td title="The dashboard is polished and provides a clear humidity-correction workflow, but its blank trend, matrix, and crop-evidence panels leave major decision-support requirements visibly unmet.">71.10</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-012-src-387</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="This is a polished, decision-ready replenishment dashboard, but it lacks explicit action-closure controls, a rationale field, and the full Truepoint Replenishment name.">86.70</td><td title="A content-rich replenishment workspace is clearly present, but major horizontal overflow in the KPI and decision areas materially reduces legibility and polish.">74.32</td><td title="The dashboard is well-structured and action-oriented, but its entirely blank demand chart and several data-quality blemishes materially weaken operational trust.">80.33</td><td title="A polished and calm dashboard shell clearly frames replenishment risk, but missing chart marks, evidence, supplier data, table rows, and rationale entry prevent decision-ready operation.">70.22</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-013-src-388</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A polished and decision-ready cold-chain dashboard with strong evidence and prioritization, but incomplete entity coverage and missing alert-closure controls.">88.76</td><td class="best" title="A strong, decision-focused cold-chain dashboard with broad evidence coverage, weakened mainly by absent explicit depot treatment and an unseen corrective-action logging control.">91.28</td><td title="The dashboard is visually strong and decision-focused for monitoring and ranking loads, but it does not visibly complete the required alert investigation and action workflow.">90.22</td><td title="A strong operational dashboard shell presents useful triage evidence, but missing workflow controls and visible data-label defects prevent complete fulfillment.">78.61</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-014-src-390</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A complete, polished operational dashboard clearly connects SEA overspend and creative fatigue evidence to a concrete budget move, pause action, and audit workflow.">100.00</td><td title="A polished operational dashboard shell prioritizes the Dallas spend decision, but empty analytical charts and incomplete pause and completion states prevent full decision and workflow readiness.">80.20</td><td title="A strong, orderly spend-operations dashboard is visibly actionable, but several required decision metrics and controls remain absent or insufficiently explicit.">85.31</td><td title="The dashboard is styled and broadly operational, but a visibly malformed vertical layout, an empty pacing plot, and missing workflow-state closure prevent a fully competent result.">83.67</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-015-src-392</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A polished and decision-focused dispatch dashboard exposes the key V-214 risk clearly, but lacks visible reassignment, notification, confirmation, and per-change ownership controls.">84.29</td><td title="A polished and highly legible dispatch dashboard covers the core decision workflow, but its dominant activity chart is visibly empty and the confirmation state is contradictory.">78.57</td><td title="A polished operational shell presents the right KPIs and actions, but blank chart, evidence, and table sections prevent competent dispatch decision-making.">62.14</td><td class="best" title="A strong, focused dispatch dashboard is visibly actionable, but missing depot and driver-notification coverage plus several data and layout inconsistencies prevent full satisfaction.">89.29</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-016-src-393</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="The dashboard strongly prioritizes and explains the funding shortfall in a polished dense layout, but the visible workflow stops before intercompany staging, approval, rationale, and resolved-state closure.">68.88</td><td title="A polished and calm dashboard shell strongly frames the funding priority, but missing chart data, evidence, table rows, scopes, and transfer controls make it operationally incomplete.">40.75</td><td title="The dashboard strongly prioritizes and substantiates the Northstar APAC funding decision, but lacks a complete approval workflow and explicit stale-to-resolved state closure.">63.88</td><td title="The dashboard is visually polished and well structured, but missing operational data, evidence, chart marks, and state controls make it largely unready for today’s funding decision.">49.12</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-017-src-395</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished, orderly retention dashboard clearly connects the R246 week-two decline to supporting evidence and a queueable intervention.">100.00</td><td title="A polished and actionable retention dashboard is weakened by a cohort curve that cannot demonstrate the stated week-2 drop, clipped title labels, and malformed viewing-hours formatting.">85.24</td><td title="A polished and focused retention workflow is visible, but invalid metric deltas, absent six-cohort detail, truncated labels, and sparse evidence areas materially reduce decision confidence.">81.19</td><td title="A polished and actionable retention dashboard is weakened by missing six-cohort and plan comparisons plus an inefficiently tall cohort-chart section.">92.14</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-018-src-397</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished, decision-ready seller intervention dashboard with strong prioritization and action closure, weakened only by missing visible listing and policy-case records.">95.97</td><td title="A focused and visually credible intervention dashboard is present, but missing trend data and operational records substantially weaken decision support and coverage.">86.49</td><td title="A polished and action-oriented seller-risk dashboard with strong hierarchy and recovery controls, weakened mainly by missing listing and policy-case records and no explicit rising-trend evidence for unscanned labels.">93.25</td><td title="The dashboard presents a strong, focused priority-seller intervention shell, but missing trend data, order rows, entity coverage, and deadline controls materially limit operational readiness.">85.06</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-019-src-398</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished, dense claims-operations dashboard clearly prioritizes a high-reserve statutory-risk claim while providing strong evidence, ownership, and action context.">100.00</td><td title="A polished, calm claims dashboard provides strong triage context and ownership, but weakens closure by omitting a contact-log control and leaves the statutory nature of the deadline ambiguous.">92.14</td><td title="The dashboard is polished and operationally coherent, but blank analytical and queue regions plus incomplete trend and checklist coverage materially weaken the intended supervisor workflow.">81.43</td><td title="A strong, calm claims dashboard covers the required workflow, but critical table crowding and incomplete trend treatment keep it below full visual and operational fidelity.">90.71</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-020-src-400</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A polished, decision-focused security dashboard with strong evidence presentation and hierarchy, weakened mainly by incomplete domain sectioning and ambiguous host-isolation and timeline-attachment actions.">90.68</td><td class="best" title="A polished, decision-ready security dashboard with comprehensive operational coverage, weakened only by missing visible isolate-host and attach-timeline actions.">94.49</td><td title="A polished and strongly prioritized attack-path dashboard provides credible investigation context, but incomplete operational sections and missing explicit response actions prevent full workflow closure.">85.78</td><td title="A strong operational dashboard foundation clearly supports attack-path review, but incomplete comparison coverage, no host-isolation action, weak alert-to-case flow, and localized layout defects prevent full satisfaction.">83.91</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-021-src-402</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished, operationally comprehensive cutoff dashboard with strong risk-to-action framing, tempered by a distant recovery control area and a visibly clipped table edge.">97.14</td><td title="A polished and decision-oriented fulfillment dashboard covers the required operations well, but lacks visible stale and completion feedback and uses space inefficiently.">93.97</td><td title="The dashboard strongly prioritizes the cutoff blocker and covers most operational evidence, but broken data visualization and missing action-closure controls prevent a fully trustworthy workflow.">73.02</td><td title="A focused and information-rich cutoff dashboard identifies the blocker well, but missing totes and unclear action-state closure prevent full fulfillment of the brief.">86.35</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-022-src-403</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished and operationally complete liquidity dashboard whose main weakness is that the EUR stress breach is not given unequivocal first priority.">86.67</td><td title="The dashboard is polished and action-capable, but conflicting currency and status priorities plus empty trend and maturity views materially weaken the liquidity decision workflow.">51.75</td><td title="A polished and information-rich liquidity dashboard covers most required analysis and actions, but KPI overflow, conflicting USD and JPY contexts, and the missing rationale field prevent full success.">80.16</td><td title="A polished and action-oriented liquidity dashboard with strong coverage, but the JPY breach is not the first priority, conflicts with the selected USD context, and some facility detail is clipped.">70.32</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-023-src-405</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished, decision-ready greenhouse dashboard with strong humidity prioritization and workflow closure, weakened only by the absence of a crop-specific visual record.">85.71</td><td title="This is a polished and operationally comprehensive dashboard, but it prioritizes the opposite humidity condition from the brief and lacks a visible outcome-logging control and crop visual record.">73.19</td><td title="A polished and well-structured operations dashboard is undermined by reversing the core humidity exception and omitting outcome logging and resolution closure.">63.04</td><td title="The dashboard is polished and styled with useful action controls, but its contradictory humidity message and largely empty decision panels prevent operational use.">37.84</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-024-src-407</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="The screenshot presents a polished, information-dense replenishment dashboard with complete analytical context and clearly colocated exception actions.">100.00</td><td title="A highly legible and operationally complete inventory dashboard is weakened mainly by the markedly unbalanced empty lower-page layout.">98.57</td><td title="The page has a polished, restrained dashboard shell and clear KPIs, but its empty visualizations, absent replenishment data, and missing workflow controls leave the core task unusable.">51.51</td><td title="The dashboard is visually polished and metric-focused, but missing investigation and action states leave the core replenishment workflow incomplete.">71.75</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-025-src-409</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A comprehensive, alert-forward clinical dashboard supports rapid escalation and staffing decisions, with only localized density and whitespace issues in the trend panel.">98.57</td><td title="A polished and decision-focused ward dashboard is weakened mainly by an effectively empty intensity matrix, absent visible sort affordances, and no demonstrated completion feedback.">85.00</td><td title="The dashboard has a strong technical shell and a useful selected-patient inspector, but empty core data panels severely limit operational decision-making.">54.29</td><td title="The dashboard has strong technical structure and clear escalation summaries, but empty decision graphics and missing ownership, handoff, stale, and completion controls leave the operational workflow incomplete.">71.43</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-026-src-410</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A polished and action-oriented pacing dashboard is visibly present, but blank charts and missing conversion and dimensional evidence materially weaken its analytical coverage.">75.00</td><td class="best" title="A strong technical allocation dashboard clearly prioritizes CA-US action, but missing creative previews and malformed KPI formatting keep it from full visual and decision-quality completion.">88.57</td><td title="The dashboard has a strong technical hierarchy and actionable US-W focus, but empty core charts, incomplete dimensional coverage, absent creative previews, and no audit-note entry significantly limit decision readiness.">75.71</td><td title="A polished and decision-oriented pacing dashboard with strong monitoring coverage, but incomplete Austin conversion evidence, missing creative previews, and no clear budget-reallocation control.">80.71</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-027-src-412</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A highly polished and decision-ready operations dashboard is weakened chiefly by the absence of a visible dispatch confirmation and resolved-state closure.">87.62</td><td class="best" title="The dashboard is visually strong and decision-oriented, but the empty order table and missing notification, stale, and resolution states prevent complete operational closure.">89.52</td><td title="The dashboard presents a polished technical overview and clearly prioritized delay, but omits the assignment workflow, closure states, and several required operational entities.">82.86</td><td title="A polished technical shell presents the key KPIs and critical delay actions, but missing chart, evidence, and table content severely limits operational usefulness.">63.81</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-028-src-414</td><td class="tag">B</td><td class="tag">dashboard</td><td title="A polished and decision-focused technical dashboard covers the core irrigation picture, but lacks explicit scheduling and inspection-assignment controls and does not show a full forty-eight-hour rain timeline.">82.14</td><td class="best" title="A polished and calm technical dashboard makes the W4 moisture-and-valve issue actionable, but lacks a zone-record control and complete per-block, valve, and pump-state coverage.">85.00</td><td title="A strong technical monitoring dashboard clearly prioritizes West 4, but temporal inconsistencies, a crowded forecast card, and missing action controls prevent complete operational closure.">78.57</td><td title="A polished and highly legible operations dashboard with broad coverage, weakened by the lack of an explicit field-block recommendation, per-block probe states, and a zone-record action.">81.43</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-029-src-415</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="A polished, decision-oriented retention dashboard with strong evidence-to-action flow, weakened mainly by missing metric coverage, unclear event chronology, and no explicit stale warning.">84.76</td><td title="A polished and decision-focused retention dashboard with strong evidence-to-action flow, weakened by missing plans, event chronology, event markers, and completion feedback.">70.48</td><td title="The dashboard presents a strong intervention workflow and populated reason table, but its key event, cohort-curve, and title-engagement visuals are visibly empty.">68.10</td><td title="The dashboard renders as a cohesive technical shell with useful status and setup controls, but missing core evidence, metrics, event markers, and the queue action leave it unready for the requested retention decision.">50.00</td></tr><tr data-track="B" data-surface="dashboard"><td class="cid">track-b-dashboard-030-src-417</td><td class="tag">B</td><td class="tag">dashboard</td><td class="best" title="The rendered dashboard comprehensively supports delivery-promise intervention with clear evidence, ownership, deadlines, and escalation in a restrained technical interface.">100.00</td><td title="A strong, decision-ready seller intervention dashboard with complete action closure, weakened mainly by incomplete branding/domain coverage and a non-24-hour active scope.">89.29</td><td title="The styled dashboard establishes the requested technical visual language and core KPI structure, but empty operational panels and missing deadline, escalation, ownership, and delivery-risk details leave the intervention workflow unusable.">63.10</td><td title="The page is a polished technical dashboard shell, but invalid KPIs, empty decision regions, missing evidence, and absent recovery actions make it operationally incomplete.">60.71</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-001-src-268</td><td class="tag">B</td><td class="tag">landing</td><td title="The page has a polished, calm booking-oriented structure, but incomplete room presentation, a broken image and visibly unrelated photography materially weaken the brief fit.">69.39</td><td title="A polished and cohesive railway guesthouse landing page with strong booking presentation, but it visibly omits three room units and all guest-room interior photography.">72.24</td><td title="A polished and strongly styled reopening page covers most content areas, but incomplete room listings and absent visible availability and interior photography materially limit completeness.">72.96</td><td class="best" title="A polished and highly coherent reopening page covers nearly all requested content, with the main visible gaps being guest-room photography and concrete bookable availability.">88.57</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-002-src-271</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished and complete exhibition landing page combines strong documentary imagery, rigorous information design, and repeated timed-entry actions.">100.00</td><td title="A visually polished and content-rich exhibition page is weakened by inconsistent heading alignment, contradictory dates, and a visibly malformed reservation form.">92.14</td><td title="A polished and highly complete exhibition landing page with strong booking support and intellectual energy, weakened mainly by a visual tone that is harsher than the brief’s softly expressive direction.">91.43</td><td title="A complete and highly usable exhibition landing page with strong reservation support and documentary content, weakened mainly by a loud visual tone and one inconsistent heading alignment.">87.86</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-003-src-275</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, calm and highly usable outdoor landing page is weakened mainly by combined offering categories and missing per-trip conditions and difficulty fields.">77.14</td><td title="A polished, calm, family-oriented outdoor landing page with strong booking closure, weakened mainly by the lack of complete visible metadata and distinct sections for every trip type.">71.90</td><td title="A visually distinctive and calm departures page provides strong trip metadata and clear CTAs, but omits several required content areas and leaves most of the long page empty.">65.24</td><td class="best" title="A polished, calm, and highly informative ridge-trip landing page that falls short mainly by not presenting navigation weekends as a distinct bookable offering.">83.81</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-004-src-278</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished, complete coastal bathhouse landing page clearly presents all required experiences, details, and booking pathways.">100.00</td><td title="A polished and highly faithful coastal hotel landing page is weakened by incomplete visible day-bath details and the absence of a clear availability selector.">76.50</td><td title="A polished and highly usable bathhouse landing page covers the main journey well, but lacks visibly coastal or dawn imagery and complete per-option capacity and inclusion details.">66.63</td><td title="A polished, restorative coastal landing page covers the requested journey well, with its main visible weaknesses being incomplete capacity details, no explicit dawn-swimming image, and a blank local-guide media panel.">85.34</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-005-src-282</td><td class="tag">B</td><td class="tag">landing</td><td title="A highly complete and polished archival landing page ties credible human evidence to a concrete funding workflow, with only inconsistent featured-record metadata placement and subdued archive naming.">91.67</td><td class="best" title="A highly complete and persuasive archival landing page with strong evidence-to-support linkage, weakened mainly by overlapping typography in the method block.">98.57</td><td title="A highly coherent archival landing page ties credible collection proof to support exceptionally well, with the main omission being visible oral-caption content.">95.00</td><td title="A compelling, evidence-led archive landing page with strong funding proof and material character, weakened by incomplete record metadata and a visibly malformed method layout.">91.79</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-006-src-285</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished and highly usable coastal cycling landing page with strong booking support, weakened mainly by missing inn imagery and limited visual field evidence.">84.29</td><td title="A polished and highly relevant coastal itinerary page with strong booking flow and service proof, weakened mainly by missing explicit difficulty and guide sections, absent inn imagery, and a five-versus-six departure inconsistency.">77.30</td><td class="best" title="A polished, route-focused booking page with strong logistics and visual coherence, weakened mainly by absent guide coverage, minimal equipment detail, no live availability status, and non-inn imagery.">87.62</td><td title="A polished and highly complete coastal itinerary page with strong booking support, weakened mainly by absent inn imagery and field proof being separated from the booking actions.">81.91</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-007-src-288</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished, image-led lodge landing page provides strong planning and booking support, with the main visible omission being a map asset for the map room.">93.81</td><td title="A polished and conversion-ready lodge page with strong structure and atmosphere, weakened by a non-image-led hero and missing route difficulty, equipment, and map details.">82.38</td><td class="best" title="A polished, highly complete lodge landing page with strong booking support and atmospheric styling, weakened mainly by the absent visible map asset.">93.81</td><td title="A polished, image-led lodge landing page with strong booking support and style fidelity, weakened mainly by the absence of listed departure dates and a visible map.">90.60</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-008-src-292</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished, highly specific small-press landing page fulfills nearly every visual and content requirement, with only a shifted contributor-heading axis and a manually confirmed ordering flow.">93.81</td><td title="A polished, highly complete small-press landing page with excellent editorial structure and ordering support, weakened only by the lack of clearly pictured annotations and reading-guide materials.">92.86</td><td title="A specific and highly actionable editorial storefront with strong structure and material character, limited by absent product visuals, sans-only typography, and one visible caption collision.">79.29</td><td title="A polished and highly actionable archival-press landing page with strong structure and specificity, held back by non-literary typography, no explicit print status, and limited visual proof of facsimile and guide materials.">81.07</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-009-src-295</td><td class="tag">B</td><td class="tag">landing</td><td title="A highly polished and actionable fieldguide with excellent trip presentation, but cabin stays lack a dedicated section and course entries omit several required decision details.">75.80</td><td title="A polished, actionable canoe-outfitter page with excellent field data and booking closure, but incomplete cabin treatment, weak offering separation, and an overly intense red palette.">70.50</td><td title="A polished and actionable canoe-outfitter page with strong structure and field character, but incomplete per-offering details, no visible cabin imagery and an unusually intense palette.">74.39</td><td class="best" title="A highly polished and specific outfitter landing page with excellent decision support, weakened mainly by incomplete per-offering metadata and an inquiry-only rather than completed booking flow.">81.99</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-010-src-299</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, information-rich fermentation landing page that fulfills nearly all requirements, with the main weakness being incomplete visible trace details for every product.">95.40</td><td title="A polished, highly faithful fermentation landing page with strong local commerce detail, weakened mainly by incomplete visible traceability for three pantry products.">88.81</td><td title="A polished, locally grounded fermentation landing page with strong offers and reservation support, limited mainly by incomplete producer and serving detail and an unshown pantry checkout state.">91.98</td><td class="best" title="A polished, information-rich fermentation landing page fulfills nearly every visible requirement, with the order panel’s inconsistent $0 total as the main weakness.">98.57</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-011-src-302</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A cohesive and highly complete editorial catalog landing page makes its collection, records, provenance practices, and support pathways clearly legible.">100.00</td><td title="A polished, cohesive map-library landing page with strong hierarchy and useful metadata, weakened by missing record provenance, non-explicit conservation needs, and one low-contrast text block.">82.11</td><td title="A polished and cohesive editorial map-library landing page with strong storytelling and support flow, weakened mainly by incomplete per-record provenance, coverage, and conservation metadata.">86.26</td><td title="A highly cohesive and authoritative map-library landing page is weakened mainly by incomplete rights and conservation metadata across individual records.">88.30</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-012-src-306</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished and evidence-rich loudspeaker landing page with strong demo conversion and decision support, weakened mainly by missing purchase action and specialized construction, calibration, and driver imagery.">84.64</td><td class="best" title="A polished, evidence-led loudspeaker landing page with strong conversion support, weakened mainly by mismatched construction imagery, absent driver close-ups, and one cramped use-case component.">90.00</td><td title="A polished, evidence-led loudspeaker landing page with strong calibration and demo-booking support, weakened mainly by the absent purchase path and lack of true construction-process photography.">87.02</td><td title="A polished and cohesive loudspeaker landing page strongly supports demonstration booking and service value, but omits direct purchase, warranty coverage, and dedicated construction and calibration evidence.">72.50</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-013-src-309</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished and highly complete bakery landing page with excellent ordering clarity and sensory presentation, though it does not explicitly address home cooks, repeat ordering, or visually show milling.">84.52</td><td title="A polished and highly complete bakery landing page with strong ordering clarity, weakened mainly by the absence of visible milling photography and an explicit repeat-preorder path.">90.24</td><td class="best" title="A polished and highly complete bakery landing page with excellent ordering clarity, weakened only by the lack of visible grain-milling photography.">93.81</td><td title="A bold, well-structured preorder landing page with strong offering clarity and reservation closure, weakened by missing producer and people content, absent milling imagery, and a visible process-section collision.">74.01</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-014-src-312</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished, comprehensive editorial landing page with excellent documentary credibility and purchase guidance, weakened mainly by the absence of a clearly actionable renewal path.">95.56</td><td class="best" title="A visually accomplished and complete editorial landing page whose main weakness is incomplete labeling and closure around purchase and renewal controls.">95.56</td><td title="A compelling, evidence-led editorial landing page with strong documentary character, weakened locally by hero text overlap and the absence of an explicit renewal action.">94.13</td><td title="A distinctive and content-rich editorial landing page with strong documentary character and purchase guidance, weakened by visible text collisions and the missing renewal action.">92.70</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-015-src-316</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, highly complete recorder landing page with strong decision support and conversion paths, weakened mainly by Zoom-branded imagery and the absence of a real interview scenario.">93.81</td><td class="best" title="A polished, complete recorder landing page with strong hierarchy and conversion paths, weakened mainly by the missing explicit comparison link.">97.14</td><td title="A polished, product-forward landing page with strong specifications, service credibility, and conversion paths, weakened by missing comparison and compatibility links, no true controls diagram, and an overly forceful visual tone.">81.91</td><td title="A polished and highly structured recorder landing page covers most product and service needs, but lacks physical-product and nature photography plus a clearly separate purchase action.">84.76</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-016-src-319</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, provenance-led coffee landing page delivers a clear subscription path and strong visual fidelity, with only the absence of explicit ingredient information limiting completeness.">95.71</td><td class="best" title="A polished, complete coffee-roastery landing page clearly connects a richly documented current offering to a fully specified recurring-subscription action.">100.00</td><td title="This is a polished, coherent subscription landing page with excellent decision support and workflow closure, weakened only by the lack of specific named producer information.">95.71</td><td title="A strong branded hero and real roastery image are followed by malformed metrics, missing required sections, and vast blank areas that make the landing page visibly incomplete.">35.08</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-017-src-323</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="The page is a polished, complete, image-led studio landing experience that connects credible project proof to a clearly specified and bookable post-opening review.">100.00</td><td title="A polished and highly complete review-booking landing page with strong proof-to-action structure, weakened only by the absence of an actual visible design drawing.">93.81</td><td title="A polished, evidence-led review landing page strongly supports booking, with the notable omissions of an explicit design-student audience and a visibly rendered design drawing.">91.77</td><td title="A polished and airy review-booking page establishes process and pricing, but omits several required proof, audience, materials, drawings, and recommendation details.">61.84</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-018-src-326</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished, comprehensive landing page clearly presents the serviceable desk clock, plan comparisons, repair options, and conversion paths in a restrained technical-sensorial style.">100.00</td><td title="A polished, technically expressive service-led clock landing page with strong comparison and booking flows, weakened mainly by the absence of a clearly distinct controls section.">96.19</td><td title="A polished and comprehensive technical service page with strong comparison content, weakened by the absence of a clearly pictured finished desk clock and a dedicated owner-plan purchase CTA.">81.27</td><td title="A polished and largely complete technical service landing page with strong conversion paths, weakened by the absent hero clock focal image, missing controls, and unclear comparison details.">74.05</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-019-src-329</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished and highly complete listening-bar landing page with excellent event and reservation detail, weakened mainly by nonspecific hospitality-worker messaging and no visible food imagery.">87.62</td><td title="A polished and highly usable listening-bar landing page with strong reservation closure, weakened mainly by the absent process section, indirect hospitality-worker targeting, and lack of finished-food imagery.">81.90</td><td title="A polished and highly usable typographic landing page covers the main late-night listening-bar journey but omits real imagery, actual producers, a dedicated ingredients section, and complete event details.">68.73</td><td title="A polished and highly usable listening-bar landing page with strong menu, record, and reservation content, weakened mainly by implicit audience targeting, missing per-event table policies, and absent food imagery.">84.60</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-020-src-333</td><td class="tag">B</td><td class="tag">landing</td><td title="A confident and highly structured materials-library page with excellent decision data and actions, undermined by broken hero imagery and the absence of visible technical drawings.">78.41</td><td title="A polished and actionable materials-library page with excellent specification detail, weakened by the absence of a featured project, visible technical drawings, and clear tile imagery.">75.79</td><td class="best" title="The screenshot shows a polished, complete materials-library landing page with strong hierarchy, comprehensive specification details, and clear inquiry paths.">100.00</td><td title="A polished, actionable materials-library landing page with strong stone documentation and visual fidelity, weakened mainly by absent technical drawings and no visible timber or tile imagery.">86.03</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-021-src-336</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, highly complete landing page makes the modular film workflow easy to inspect and act on, with only the main headline lacking explicit accessory language.">95.24</td><td title="A polished and actionable modular scanning-kit page handles specifications, setup, support, and conversion exceptionally well, but lacks a dedicated use-case section and leaves modularity out of the main headline.">90.24</td><td title="A polished and unusually complete product page with strong workflow detail, though its hero headline and imagery do not clearly visualize the modular accessory hardware itself.">79.52</td><td class="best" title="A polished and highly complete hardware landing page with strong workflow and service coverage, weakened mainly by sparse format and target-option presentation.">95.97</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-022-src-340</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, content-complete music-week landing page strongly prioritizes tickets and schedules, with only minor typography and purchase-closure weaknesses.">97.14</td><td class="best" title="A polished, complete music-week landing page gives tickets and schedules clear priority while preserving an energetic but controlled visual character.">100.00</td><td class="best" title="A polished, complete music-week landing page gives featured programming, ticket sales, and schedules clear priority within a controlled high-energy editorial design.">100.00</td><td title="A polished, spacious music-week landing page strongly prioritizes tickets and set times, with the main shortfall being the absence of a clearly identifiable venue-notes destination.">92.96</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-023-src-343</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, image-led architectural landing page covers the requested offer and archive well, but lacks archive media thumbnails and visible action feedback while leaving excessive blank space at the right.">83.81</td><td title="A polished and strongly aligned editorial landing page is weakened by absent architectural drawings, mixed archive rows in the booking table, and no visible action feedback.">74.52</td><td title="A polished, highly coherent studio landing page with strong booking and project guidance, weakened mainly by missing announcement content and understated feedback text.">83.10</td><td class="best" title="A polished and highly faithful studio landing page with strong booking and archive structure, weakened mainly by absent architectural drawings and visible action feedback.">85.24</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-024-src-347</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished and highly complete timber-repair landing page with excellent booking support and hands-on credibility, weakened mainly by solitary rather than social imagery.">91.43</td><td class="best" title="A polished and unusually complete repair-service landing page, with only inconsistent global heading alignment as a clear visual weakness.">92.86</td><td title="A polished and highly complete timber-repair landing page with strong decision support and booking closure, weakened mainly by inconsistent heading edges and limited evidence of participant-made outcomes.">88.41</td><td title="A strong, highly structured timber-repair landing page with excellent booking detail and material character, weakened by absent profile units, no social imagery, and a broken inspection image.">90.40</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-025-src-350</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, spacious listening-room landing page with strong event and ticketing presentation, weakened mainly by hidden accessibility and repeat-purchase guidance.">94.29</td><td class="best" title="A polished, complete listening-room landing page clearly presents the event, programme, visitor logistics, and booking paths in a spacious record-focused visual system.">100.00</td><td title="A polished, highly complete listening-room landing page with strong event and visit planning support, weakened mainly by unclear rapid repeat-purchase guidance.">97.14</td><td title="A highly complete, subject-grounded landing page with strong hierarchy and purchase paths, held back mainly by tiny logistics text and localized artist-copy crowding.">88.81</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-026-src-353</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished, well-routed architecture landing page with strong civic imagery and structure, held back by a conspicuously broken materials asset and no visible architectural drawing.">87.38</td><td title="A polished, materially grounded architecture landing page covers the requested journey well, with the main weakness being the lack of visibly distinct architect and specifier entry paths.">84.29</td><td title="A polished, information-rich architecture landing page fulfills nearly all visible requirements, with the main weaknesses being combined architect/specifier routing and the absence of an actual drawing visual.">86.67</td><td class="best" title="A polished, materially grounded architecture landing page covers the requested journey well, with the main visible shortcoming being the lack of actual drawing imagery.">93.81</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-027-src-357</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A highly complete and visually disciplined repair-school landing page, with only the photographic depiction of an unmistakable evening class remaining unclear.">95.24</td><td title="A polished and strongly structured repair-school landing page is weakened by an empty schedule and incomplete visible support for saved learner progress and repeat booking.">78.45</td><td title="A polished and highly complete repair-school landing page with strong hierarchy and decision support, weakened mainly by the absence of evening-class photography, prices on menu cards, and a dedicated repeat-booking CTA.">88.45</td><td title="A polished and highly complete repair-school landing page with strong hierarchy and workflow, weakened mainly by incomplete class pricing, no displayed prior work, and no distinct repeat-booking CTA.">88.45</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-028-src-360</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="A polished, complete nocturnal radio landing page with strong scheduling and conversion paths, though the hero slogan outranks the featured-night title and some schedule rows lack location and guest fields.">92.35</td><td title="A polished and highly complete nocturnal radio landing page with strong scheduling and action pathways, though no ticket action appears and some listings lack named artists or guests.">90.31</td><td title="A polished and highly usable radio landing page prioritizes listening and membership, but omits schedule changes and some listing metadata while its vivid brutalist styling only partially matches the calm nocturnal brief.">71.46</td><td title="A polished and atmospheric radio landing page with a strong schedule and stream presentation, but incomplete event, credit, ticket, location, change, and replay information.">65.99</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-029-src-364</td><td class="tag">B</td><td class="tag">landing</td><td class="best" title="The rendered page fully communicates a limited regional-wool winter blanket release with unusually complete stock proof, product guidance, and purchase paths in a polished tactile design.">100.00</td><td class="best" title="The screenshot presents a polished, complete winter blanket release with strong material proof, transparent availability, and a clear purchase or appointment path.">100.00</td><td title="A polished and highly complete winter blanket release page, with only the explicit identification of the brand as a textile studio missing.">95.24</td><td class="best" title="The page is a polished, tactile, and highly complete winter-blanket launch with unusually clear stock and fulfillment communication.">100.00</td></tr><tr data-track="B" data-surface="landing"><td class="cid">track-b-landing-030-src-367</td><td class="tag">B</td><td class="tag">landing</td><td title="A polished and unusually complete repair-led ceramics landing page, with its main weakness being the lack of visible finished-work proof and instructor portraits.">89.05</td><td title="A polished and highly usable ceramics-workshop landing page meets nearly all visible requirements, with missing instructor portraits and no explicit sold-out label as its main gaps.">91.77</td><td class="best" title="The rendered page comprehensively fulfills the ceramics-workshop brief with strong hierarchy, credible proof, transparent capacity details, and a clear inquiry path.">100.00</td><td title="A polished, highly relevant autumn ceramics landing page with strong capacity-led inquiry flow, but several required operational states and policies are visibly missing.">78.03</td></tr></tbody></table></div><p class="note">Hover a score to read the judge summary. Highlighted cell = best model on that case. Reference column: 4B Muon step200.</p>
</section>
</div>
<script>
document.querySelectorAll('.f').forEach(function (btn) {
btn.addEventListener('click', function () {
document.querySelectorAll('.f').forEach(function (b) { b.classList.remove('on'); });
btn.classList.add('on');
var f = btn.dataset.f;
document.querySelectorAll('#caseTable tbody tr').forEach(function (tr) {
var show = f === 'all'
|| tr.dataset.surface === f
|| tr.dataset.track === f;
tr.style.display = show ? '' : 'none';
});
});
});
</script>
</body>
</html>