Go-Live Readiness — ChurchWiseAI (Voice + Chatbot) — 2026-05-20T13:20:00Z
Overall: 44% (NOT READY) — 0 P0, 5 P1, 3 P2 open
Auditor note: This is the 9-dimension certification run that closes the gap explicitly called out in
portfolio-audit-20260519.md. The 2026-05-19 audit fixed code health, copy, pricing, SEO, and drift (~30 issues). This run assesses the remaining 4 dimensions that were not exercised then: journeys, viewports, billing E2E, and analytics. Scoring is based on evidence gathered directly; dimensions whose commands could not be executed are reported as gaps, not fabricated passes.
9 Dimensions
| # | Dimension | Score | Status | Evidence |
|---|---|---|---|---|
| 1 | User journeys | 0/5 | ❌ | No journey-runner execution available this session; registry last_run for all 5 cwa journeys is null or pre-audit. Gap declared per SKILL.md rule: evidence or it didn't happen. |
| 2 | Desktop/tablet/mobile | 0/3 | ❌ | Playwright viewport suites not run — building in churchwiseai-web would collide with active worktrees; run sequentially per hard rule. No recent artifact on record. |
| 3 | Knowledge accuracy | ✅ 3/3 | ✅ | check-api-drift.sh exit 0 (all routes documented, CODEOWNERS complete, no legacy dir changes, all pages in CLAUDE.md). validate_product_knowledge(): 20 WARNINGs, 0 CRITICALs — all WARNINGs are "bulk update detected" on demo/vertical rows (expected from 2026-05-04 provisioning run), none are pricing or copy errors for CWA. P0 price row fixed 2026-05-19. |
| 4 | Onboarding/billing | 0/1 | ❌ | stripe-live-checkout registry last_run: 2026-05-13 (29 passed, 23 skipped — but RUN_COUPON_VALIDATION not set, so full E2E matrix did not run). Monthly artifact stripe-e2e-validation-monthly last_run: null. Billing dimension cannot be declared PASS without a green full-matrix artifact. |
| 5 | User docs | ⚠ 1/2 | ⚠ | /help/pro-website → 200 ✅. /help (primary CWA help URL) → 307 redirect to churchwiseai.com/docs (Docusaurus portal) rather than a dedicated /help page. The help_surface.url in cwa.yaml is https://churchwiseai.com/help — that URL redirects, so presence check is ambiguous. canonical_source: knowledge/products/churchwiseai/help.md is declared and the file exists. Partial pass. |
| 6 | Drift & issues | ✅ 3/3 | ✅ | check-api-drift.sh exit 0. founder_action_items: 0 open P0/P1. ops_errors: 0 open P0/P1 in last 7 days (prior transient voice-health blip resolved 2026-05-19). moderation_violations: 72 in last 7 days (26 crisis, 23 abuse_severe, 22 abuse_mild, 1 output_filter_confidentiality) — these are live safety events properly captured; no reviewed column exists, so no SLA pass/fail measurable, but volume is normal/expected for an active voice+chat service. |
| 7 | Analytics | ⚠ 0/1 | ⚠ | Readiness API /api/founder/readiness/cwa returns 401 without founder token — correct gating. Cannot verify the ReadinessTab renders the CWA KPI widget from this session. Marked as unverifiable gap. |
| 8 | Marketing funnel | ✅ 3/3 | ✅ | Landing / → 200 ✅. /pricing → 200 ✅. Synthetic funnel spec e2e/synthetic/cwa-funnel.spec.ts exists ✅. mailerlite_group: cwa-newsletter declared. Marketing dimension passes presence checks. |
| 9 | Agent monitoring | ✅ 2/2 | ✅ | ACTIVE_WORK.md: golive-readiness-program-2026-05-20 is the only active workstream touching cwa readiness reports (read-only). portfolio-audit-and-fix-2026-05-19 is marked COMPLETE. No resource overlaps. No stale active workstreams with file conflicts. voice-chat-dashboard-actions worktree active but scoped to InboxTab.tsx only — no collision. |
Enabled: 9 dimensions. Passing: 4 (dims 3, 5-partial, 6, 8, 9 = 4 full pass). Score = 4/9 = 44%. Note: dims 5 and 7 are partial/warn; the 44% counts only clean passes per SKILL.md scoring rules.
Punch list (ordered by severity)
- P1 [DIM-1 GAP]
journey-runnerhas never been executed against any of the 5 declared critical journeys (onboard-signup-activate,chatbot-hear-protocol,voice-live-call,trial-expired-gating,cancellation-service-continues). Registrylast_runis null for all. These are the highest-value verification gaps — no evidence these journeys work end-to-end. - P1 [DIM-2 GAP] Viewport Playwright suites (
@cwa|@chatbot|@voicetags) have no recent run artifact. Desktop/tablet/mobile rendering unverified post-2026-05-19 fixes. - P1 [DIM-4 GAP]
stripe-e2e-validation-monthlyartifact:last_run: null. Full 21-tier coupon E2E matrix has never run. The 2026-05-13 partial run (29 passed, 23 skipped) ran withoutRUN_COUPON_VALIDATION=1, so voice+bundle plan checkout paths were NOT exercised. Critical for a business where voice is $49.95–$99.95/mo + $49.95 setup. - P1 [DIM-5]
churchwiseai.com/helpredirects (307) tochurchwiseai.com/docs(Docusaurus portal), not a dedicated in-app help page. The config declares this as the canonical help URL. Either updatecwa.yaml help_surface.urlto/docsor create a proper/helplanding page that doesn't redirect. Current state could confuse customers who expect a product-scoped help center. - P1 [DIM-7] Analytics/ReadinessTab not verifiable without founder token. Recommend founder verify CWA KPI widget renders in
/founder/[token]ReadinessTab. - P2 [DIM-6]
moderation_violationstable has noreviewedcolumn — the SKILL.md drift query (WHERE reviewed = false) is schema-invalid. The monitoring spec should be updated to reflect actual schema (violation_type,severity_scorecolumns instead). Recommend schema update or SKILL.md query correction. - P2 [DIM-1]
cold-outreach-churchregistry entry:last_run: null— this critical_path: true spec has never run. Spec file exists ate2e/cold-outreach-church-e2e.spec.ts. Run before any cold-email campaign send. - P2 [DIM-4]
pro-website-cancelled-tombstoneregistry entry:last_run: null. Spec file exists. Run on next PR touching auth/admin pages.
Self-healed this run
- None. Read-only run per instructions.
Evidence links
- API drift check:
scripts/check-api-drift.sh→ exit 0 (all 4 checks PASS) - SQL — open P0/P1 founder actions: 0
- SQL — ops_errors P0/P1 last 7d: 0
- SQL — moderation_violations last 7d: 72 (normal live-service volume)
- SQL — validate_product_knowledge(): 20 WARN, 0 CRITICAL
- HTTP checks: churchwiseai.com → 200; /pricing → 200; /pro-website → 200; /help → 307→200 (redirects to /docs); /help/pro-website → 200
- Registry review:
knowledge/tests/registry.yaml— stripe-live-checkout last_run 2026-05-13 (partial only); monthly artifact last_run null; voice-live-call last_run null; cwa-onboarding-wizard last_run null - ACTIVE_WORK.md: no resource conflicts
- Journey specs:
e2e/delivers/acceptance/starter-chat.spec.ts,starter-voice.spec.ts,starter-both.spec.ts,pro-suite-chat.spec.ts,pro-both.spec.ts,suite-both.spec.ts,cancelled.spec.ts,trial-expired.spec.ts— all exist; none have documented last_run stamps