{"id": "examples/customer-synthetic.md::document", "title": "examples/customer-synthetic.md", "text": "# Customer example: a complete service journey\n\n**Entirely synthetic.** Every organisation, record, result and proposed decision is fictional. No live checkout, account, participant, payment, approval or correction has been tested.\n\n## Profile and remit\n\nA fictional online service has a favourable front-end automated scan but an inaccessible authentication/payment handoff, a chatbot that impedes human support and a video explaining how to cancel. Some fictional customer accounts describe confidence loss despite an improving aggregate satisfaction measure.\n\nThe proposed remit covers discovery through cancellation and redress, including authentication, payment, chat, video and support suppliers. Actual entity, jurisdictions, legal scope, owners, funding and data permission remain unknown. Do not assume the EAA or any other regime applies merely because the service is customer-facing.\n\n## Proposed programme configuration\n\n| Work package | Shared/customer route | Proposed human R / A | Useful AI task | Entry/evidence/exit |\n| --- | --- | --- | --- | --- |\n| Scope and applicability | P01–P02; C01/C02 | Coordinator R / sponsor A; competent applicability reviewer | A1 scoped source retrieval and gaps | Confirm service/entity/jurisdiction, authority and actual commitments before claims. |\n| Map the whole journey | P03; C01–C03 | Research/CX R / service A | A1 map existing approved findings and research questions | Include failure, support and exit; record missing people and contexts. |\n| Plan recovery and supplier correction | P04–P05; C04; R01/R03/R11 | Procurement/operations R / service A | A1 compare options and dependency impacts | Assess availability, privacy, cost, delay and customer control of interim assistance. |\n| Design, build and produce | P05–P06; C05; R12 | Delivery/content R / delivery A | A2 draft candidate code, content and media assets if authorised | Brief video/caption/AD practitioners early; preserve exact versions and destinations. |\n| Evaluate and decide | P07–P08; C05; R05/R10 | Test/research/editorial R / authorised release A | A2 supported checks and A1 evidence package | Complete authentication, payment, support and exit evidence; no pass inferred from the front-end scan. |\n| Monitor and improve | P09; C06; R09/R14 | Operations/reporting R / service A | A1 qualified analysis; A3 routing only under adopted mandate | Preserve disparities, complaint routes, source changes and AI review burden. |\n\nSelect S08–S10 conditionally for regulatory discovery; S13/S14 for QA; S16/S17/S19 for comprehension, brand and ethical research; S21–S25 for content/media; S27 for chat. Use the [catalogue](../sources/README.md) and the owning guidance rather than copying tests here.\n\n## Worked decision trace\n\n| Record | Fictional illustration | Disposition |\n| --- | --- | --- |\n| OUT-C-01 | Understand, obtain, use and cancel the service with meaningful choice and control. | Proposed end-to-end outcome. |\n| EV-C-01 | Illustrative front-end automated pass. | Synthetic and scoped; not evidence for the handoff or complete journey. |\n| EV-C-02 | Illustrative failed authentication/payment task. | Synthetic failure; a real equivalent would remain open despite EV-C-01. |\n| EV-C-03 | Illustrative higher aggregate satisfaction with a reported excluded context. | Synthetic and causally unproven; preserve the disparity under R09. |\n| DEC-C-01 | Whether the service can transition with proposed recovery and corrective actions. | Draft, no actual owner or authenticated approval. |\n\nThe assistant routes the failed task to correction/retest under R05 and sends an unsubstantiated accessibility claim to R10. It proposes R11 evaluation of assisted access and the chatbot-to-human transition. The video enters R12 at briefing; review editorial meaning and actual distributed outputs, not only generated scripts.\n\n## Source-change exercise\n\nSuppose the owning guidance or supplier's authentication model changes. Record the old baseline, newly retrieved candidate, changed meaning, affected requirement/outcome IDs and retest needs. Mark the affected decision `reopened`; do not relabel old evidence as current. A human owner reviews adoption. No actual source change or approval is asserted by this hypothetical.\n\n## Proposed reporting and next decisions\n\nTeams receive complete-journey defects, exact configurations, suppliers, content versions and retest actions. Leadership receives material exclusion, the mismatch between aggregate indicators and scoped failures, options, funding and authority gaps. Actual outcomes are **not assessed**.\n\nNext: confirm service facts/authority; plan accessible research and safe data use; agree recovery; assign supplier and delivery work; verify real outcomes and content. Do not infer disability from abandonment, force diagnostic customer labels or use these fictional findings as evidence.\n\n## Design/brand system planning exercise\n\nThe fictional service has no verified foundation-to-component-to-QA map. Route R15 to plan system owners, adoption/remediation and evidence using the [assurance plan](../templates/system-assurance-plan.md). The generic scan cannot establish the quality of labels, alternative text or content order; assign appropriate human procedures and meaningful automated checks. Require actual reference implementation and user ACT-style script references for authentication/payment/support patterns. Plan brand recognition, trust, expectations, value, values and resonance without removing distinctive identity. All assets, tests and outcome evidence remain hypothetical/unassessed.\n", "record_type": "synthetic_example_not_real_evidence", "source_repository": "Double-FW/Accessibility_Programme_Runbooks", "source_path": "examples/customer-synthetic.md", "source_commit": "a8ec268b29e6f8612c958b183b9d33ab0747efa7", "source_url": "https://github.com/Double-FW/Accessibility_Programme_Runbooks/blob/a8ec268b29e6f8612c958b183b9d33ab0747efa7/examples/customer-synthetic.md", "source_start_line": 1, "source_end_line": 48, "source_document_sha256": "96cb24a5b08cd64faf1f025ab8da7030d40318e8a840e4fd70ea7fbbf4b80de1", "text_sha256": "96cb24a5b08cd64faf1f025ab8da7030d40318e8a840e4fd70ea7fbbf4b80de1", "heading_path": [], "metadata_json": "{}", "declared_source_license": "not_declared_in_document_front_matter", "repository_license": "MIT", "source_status": "draft_for_local_review_and_pilot_adoption", "dataset_version": "1.0.0", "source_commit_date": "2026-11-10", "language": "en-GB", "required_context_paths": ["AGENTS.md", "README.md", "RIGHTS.md", "ADOPTION.md", "framework/programme-cycle.md", "framework/decision-routing.md", "framework/human-ai-responsibilities.md", "sources/catalogue.yaml"], "review_status": "faithful_extraction_checked_not_independent_programme_or_content_review"}
{"id": "examples/employee-synthetic.md::document", "title": "examples/employee-synthetic.md", "text": "# Employee example: payroll and work-tool barriers\n\n**Entirely synthetic.** Organisation, people, findings and decisions below are fictional. No real participant, tool execution, approval, supplier or outcome is implied. All actions and benefits are proposals until a real adoption and evaluation.\n\n## Profile and remit\n\nA small fictional employer uses a purchased payroll service and several collaboration tools. Staff report that an essential payroll activity is inaccessible and that repeated cross-tool authentication adds burden. No dedicated accessibility team exists. Actual jurisdiction, legal applicability, named decision owners, budget and individual support needs are unknown.\n\nThe proposed remit is employee services, initially the payroll task and its authentication/support dependencies. The intended outcome is independent access to essential payroll information with understandable recovery and sustainable effort. Role labels below are proposals for local assignment.\n\n## Proposed programme configuration\n\n| Work package | Shared/employee route | Proposed human R / A | Useful AI task | Entry/evidence/exit |\n| --- | --- | --- | --- | --- |\n| Establish scope and ownership | P01–P02; E01 | Coordinator R / sponsor A; both unassigned | A1 intake and requirement-source routing | Need entity/jurisdiction, service facts and actual delegation; exit is confirmed scope or explicit gaps. |\n| Urgent access and confidential support | P09/P03; E02; R01/R04 | Service owner for defect; separate confidential support owner | A1 minimal handoff drafting; no sensitive case upload | Record blocked task, agreed safe support, receiving owner and restricted case reference if needed. |\n| Task and capability baseline | P03; E01/E03 | Research/evaluation R / service owner A | A1 structure approved observations | Actual task, configurations, worker input, support/effort and coverage; no productivity ranking. |\n| Supplier and improvement options | P04–P05; E04; R03 | Procurement/delivery R / sponsor or delegated service A | A1 options and renewal dependencies | Compare correction, configured alternatives and replacement; funding and dates remain proposed. |\n| Deliver and evaluate | P06–P08; E05 | Delivery/evaluation R / authorised service A | A2 candidate changes/checks only after real permission | Exact build/configuration, technical and complete-task evidence, retest and actual decision. |\n| Operate and review | P09; E06 | Operations R / service A | A3 routine issue routing only if adopted | Review participation, cumulative burden and support quality; retain open supplier defects. |\n\nSelect source routes S01/S02 for coordination and responsibility; S08 for applicability discovery; S11/S12 for work and task evaluation; S13/S14 for layered QA; S19/S20 for ethical data and AT/AI flows; S21/S22 for payroll communication/report outputs. Resolve local source-use baselines before application through the [catalogue](../sources/README.md).\n\n## Worked decision trace\n\n| Record | Fictional illustration | Disposition |\n| --- | --- | --- |\n| OUT-E-01 | Complete the payroll information activity independently in the relevant configuration. | Proposed outcome; acceptance method needs local agreement. |\n| EV-E-01 | Illustrative account: a task is blocked; no actual evidence file or participant exists. | `synthetic: true`; raw result is illustrative failure; excluded from real claims. |\n| ISSUE-E-01 | Shared service defect with private support handled separately. | Route R01/R04; support does not close the defect. |\n| DEC-E-01 | Decide interim support, supplier correction and required funding. | Draft; no actual decision owner or authenticated approval reference. |\n| REVIEW-E-01 | Re-evaluate the configured workflow after supplier correction and review cumulative burden. | Planned, not executed. |\n\n## Decision exercise\n\nIf the employer has no confirmed execution mandate, the assistant can draft the programme, supplier questions and decision brief at A1. It cannot send them, change payroll or access individual records merely because the example mentions those actions. Confirm actual permissions before A2/A3.\n\nIf manual/task evidence later fails while an automated scan passes, keep the failure under R05, return to P06/P07 and block the unsupported claim. An authorised human makes any permissible transition decision with the remaining defect visible. An exception does not change the failed result or waive an obligation.\n\n## Proposed reporting and next decisions\n\nTeam view: affected task/configuration, known unknowns, supplier dependencies, proposed correction and retest owner. Leadership view: impact on essential work, interim access quality, options, funding/authority gaps and decision consequences.\n\nProposed outcome measures are access to the critical task, independence/control, cumulative effort, confidence, support quality and meaningful resolution. Actual baseline/current values are **not assessed**. Agree sample, disclosure controls and evidence before collection. Keep adjustment/health/performance records outside the shared programme.\n\nNext: confirm actual service owner and support route; verify relevant facts and source baselines; commission proportionate evidence; approve feasible work. This exercise is a documented process illustration, not evidence of operational effectiveness.\n\n## Design/brand system planning exercise\n\nThe fictional employer has not supplied a design system, internal brand system or linked QA library. Under R15, propose an owned capability assessment and a [system-assurance plan](../templates/system-assurance-plan.md). Inventory foundations such as payroll labelling/content order and the configured patterns/components, linking actual guidance, methods, reference implementations and user ACT-style scripts when available. Every missing reference remains a gap, not a generated pass. Evaluate recognition, trust, expectations and value of internal communication without ranking workers. No actual system adoption, tests or brand evaluation have occurred.\n", "record_type": "synthetic_example_not_real_evidence", "source_repository": "Double-FW/Accessibility_Programme_Runbooks", "source_path": "examples/employee-synthetic.md", "source_commit": "a8ec268b29e6f8612c958b183b9d33ab0747efa7", "source_url": "https://github.com/Double-FW/Accessibility_Programme_Runbooks/blob/a8ec268b29e6f8612c958b183b9d33ab0747efa7/examples/employee-synthetic.md", "source_start_line": 1, "source_end_line": 50, "source_document_sha256": "e736e308187832504094e400fe921a341e19fc664a3678c591a8aaebf768a55a", "text_sha256": "e736e308187832504094e400fe921a341e19fc664a3678c591a8aaebf768a55a", "heading_path": [], "metadata_json": "{}", "declared_source_license": "not_declared_in_document_front_matter", "repository_license": "MIT", "source_status": "draft_for_local_review_and_pilot_adoption", "dataset_version": "1.0.0", "source_commit_date": "2026-11-10", "language": "en-GB", "required_context_paths": ["AGENTS.md", "README.md", "RIGHTS.md", "ADOPTION.md", "framework/programme-cycle.md", "framework/decision-routing.md", "framework/human-ai-responsibilities.md", "sources/catalogue.yaml"], "review_status": "faithful_extraction_checked_not_independent_programme_or_content_review"}