April 2026 pricing overhaul — Voice $49.95/$99.95, setup fee, annual = 11 months

Status

DECIDED

Context

Pricing research against GoHighLevel and comparable agency voice-bot offerings revealed ChurchWiseAI was severely underpriced. GoHighLevel agencies charge $500–997/mo for generic voice bots. Our purpose-built ministry agent with 39 tools, HEAR protocol, crisis detection, and per-denomination calibration was priced at $39.95/$69.95 — 10–25x below comparable solutions.

The prior voice pricing was also a unit-economics problem: per-minute Telnyx and Deepgram costs, LLM token costs, and LiveKit infrastructure costs were not adequately covered at the old rates, particularly at the Starter tier.

Annual billing existed for chat plans but not voice or bundle plans. Churches lack the SaaS usage patterns that make voice-only monthly appropriate — committed annual customers reduce churn and offset per-minute cost uncertainty.

Decision

Shipped 2026-04-12. All 11 new Stripe prices created (test + live), all 26 downstream files updated (blog, docs, emails, prompts, pricing.ts).

Voice: $39.95 → $49.95/mo (Starter), $69.95 → $99.95/mo (Pro)
Bundles: $49.95 → $54.95/mo (Starter), $79.95 → $119.95/mo (Pro), $99.95 → $139.95/mo (Suite)
Setup fee: $49.95 one-time, charged on first invoice for all monthly voice/bundle plans. Waived entirely on annual plans.
Annual billing added to ALL tiers: annual = 11 months (1 month free). Annual plans do not carry the setup fee — the annual commitment replaces it.
Chat plans unchanged.

Rationale

Churches spend $800–1,500/mo on part-time receptionists. At $49.95–$99.95/mo the product is still 80–90% cheaper than human staffing, and 10x more sophisticated than generic agency voice bots. The price was corrected to a value-appropriate level before public launch.

The setup fee reflects real provisioning cost (Telnyx number acquisition, LiveKit SIP trunk creation, agent configuration) and filters for churches that are genuinely ready to deploy. Waiving it on annual plans rewards commitment.

Annual = 11 months was chosen over a percentage discount to make the value concrete to non-finance-trained pastors: "one month free" is clearer than "~8.3% off."

Consequences

Good: Unit economics viable at current call volumes. Price positioned for value not cost. Setup fee creates a natural qualifier for serious buyers. No customers were on old voice/bundle pricing at the time of change — zero migration impact.
Bad: Higher entry price may lengthen the sales cycle for the smallest churches. Two rounds of QA caught 3 critical stale-price bugs (knowledge-check cron, founder-costs.ts, Header.tsx "save $24.95" → "save $20") — indicates pricing is referenced in many places and future price changes carry drift risk.
Reversible? Yes, but reverting Stripe prices requires creating new price objects (Stripe prices are immutable). Old prices archived in PRICING.md for existing-subscriber reference.

Alternatives considered

10% incremental increase — rejected; insufficient to reach value-appropriate pricing and would require another revision within 6 months.
Per-minute usage billing — rejected; unpredictable bills are a major churn driver for church administrators. Flat-rate is a feature for this buyer.

Status​

Context​

Decision​

Rationale​

Consequences​

Alternatives considered​

Links​