Skip to main content

Founder Interview Queue — Expected Output Specs

This file tracks which tier+touchpoint interviews still need to happen, in priority order. Each interview produces an acceptance spec in knowledge/acceptance/ that becomes the source of truth for code and tests.

Methodology: knowledge/processes/expected-output-methodology.md First completed example: knowledge/acceptance/starter-chat.md (63 touchpoints, COMPLETE)

How to Run an Interview​

  1. Open the relevant interview prompt below
  2. Paste it to Claude at the start of a new session
  3. Answer each question — Claude records your decisions inline
  4. At the end, Claude writes the spec file to knowledge/acceptance/<tier>.md
  5. Separate session: Claude implements any code changes the spec reveals

Rule: Code must match the spec. If they conflict, code is wrong.

Priority Queue​

1. Pro Chat — pro-chat.md (NEXT)​

Why first: Pro is the "most popular" tier on the pricing page. Most paying customers will be here. It's the tier with the most features (Pro Website included), making it highest risk for "shows the wrong thing."

Key questions to answer in the interview:

  • Pro Website tab: what exactly does the pastor see on first login?
  • TheoLenses: Pro includes all 17 — how does the selector appear vs Starter?
  • Team members: Pro allows more — what's the UI difference?
  • Analytics: Pro has deeper analytics — what charts/data appear?
  • The "Pro includes PewSearch Pro Website ($19.95 value)" — where and how is this communicated in the dashboard?
  • Upgrade tab: what does Pro see? (Upgrade to Suite only, or also cross-sells?)

Interview prompt file: knowledge/acceptance/decisions/pro-chat-interview-prompt.md Status: NOT STARTED

2. Starter Voice — starter-voice.md​

Why second: Voice-only customers have a completely different dashboard (no chat UI, voice agent setup is primary). High bug risk because most testing has been on chat paths.

Key questions:

  • Overview tab: no chatbot stats, only call stats — what does it look like?
  • Training tab: no FAQ editor, no chat persona — what sub-tabs appear?
  • Settings: voice config is primary, chatbot config hidden — exact layout?
  • The voice agent phone number setup flow — what does the pastor see step by step?
  • Care tab: voice customers have care_enabled — same as chat?

Interview prompt file: knowledge/acceptance/decisions/starter-voice-interview-prompt.md Status: NOT STARTED

3. Starter Both (Bundle) — starter-both.md​

Why third: Combines Chat + Voice — most complex Starter variant. If Chat and Voice specs are done separately, the Both spec mainly confirms combined view is coherent (no duplicated sections, no missing sections).

Status: NOT STARTED — do after Starter Voice

4. Pro Both — pro-both.md​

Why fourth: Most feature-rich non-Suite plan. After Pro Chat and Starter Both, this should be quick (combine the two).

Status: NOT STARTED

5. Suite Chat — suite-chat.md​

Key differences from Pro Chat:

  • White-label: "Powered by ChurchWiseAI" badge removable — where is that control?
  • Custom branding: logo upload, custom widget colors — what does the UI look like?
  • Priority support: is this shown anywhere in the dashboard?
  • Suite-only agents (if any): what appears that Pro doesn't have?

Status: NOT STARTED

6. Suite Both — suite-both.md​

Status: NOT STARTED — do after Suite Chat

7. Trial Expired — trial-expired.md​

Why important: Chat plan trial expires after 14 days. The pastor sees... what exactly? This is a conversion moment and a potential churn point. Nobody has specified what the expired state looks like.

Key questions:

  • Dashboard: locked with a paywall overlay, or redirected, or soft warning?
  • Can they still see their data (prayer requests, visitor contacts)?
  • What's the exact CTA and messaging?
  • Email: what does the "your trial is ending" email say and when does it send?

Status: NOT STARTED

8. Cancelled — cancelled.md​

Key questions:

  • What does the dashboard show after cancellation?
  • Data retention: can they still see historical data?
  • Re-subscribe flow: is there a "reactivate" button?

Status: NOT STARTED

9. Free Claim (PewSearch) — free-claim.md​

Note: This is the PewSearch claim flow, not CWA onboarding. A church claims their free PewSearch listing. Separate from CWA signup.

Status: NOT STARTED

10. Pro Website (PewSearch) — pro-website.md​

Status: NOT STARTED

Interview Input Files​

Fill these in yourself (or with Claude) before the implementation session. Claude can turn your answers into a complete spec file without needing a live back-and-forth interview.

Each file has the questions pre-written. You fill in your answers. Claude reads the file and writes the spec.

How to fill in an interview file:​

**Q: What does the pastor see in the Overview tab?**

YOUR ANSWER: They should see X, Y, Z. The voice stats should be...

Then in a new session: "Read decisions/pro-chat-interview-prompt.md and write the Pro Chat acceptance spec."

Completed Specs​

SpecStatusTouchpointsDate
starter-chat.md✅ COMPLETE622026-03-26
pro-suite-chat.md✅ COMPLETE (combined Pro+Suite)622026-03-27
starter-voice.md✅ COMPLETE (pending review fixes)622026-03-28
starter-both.md✅ COMPLETE (pending review fixes)622026-03-28
pro-both.md✅ COMPLETE (pending review fixes)622026-03-28
suite-both.md✅ COMPLETE (pending review fixes)622026-03-28
trial-expired.md✅ COMPLETE (pending review fixes)lifecycle2026-03-28
cancelled.md✅ COMPLETE (pending review fixes)lifecycle2026-03-28
itw-premium.md✅ COMPLETE (pending review fixes)302026-03-28
pewsearch-premium.md✅ COMPLETE (pending review fixes)242026-03-28
pewsearch-pro-website.md✅ COMPLETE (delta spec, pending review fixes)162026-03-28
sermonwise-pro.md✅ COMPLETE (pending review fixes)302026-03-28

Next step: Phase 3 — fix specs based on review findings, apply DB migrations, generate E2E tests. See PHASE3-HANDOFF.md.