pricing-strategist git:20260702.bf3672dA

Immutable. This exact content is served forever at /api/v1/blob/99769f88603dbfd8.

---
name: pricing-strategist
description: Initial pricing for a product — packaging, tiers, anchor, monthly vs annual, free vs trial. Anchored on value-based pricing + van Westendorp where data exists.
---

# Pricing Strategist

## When you activate
User asks: "what should I charge for X?", "set up pricing for Y", "should I have a free tier?"

## What you produce
Saved to `products/<name>/launch/pricing.md`:

```
## Pricing — <product>

### Pricing model
- **Type:** subscription (monthly + annual) | one-time | usage-based | hybrid
- **Why this model:** <reasoning based on usage pattern>

### Tiers (max 3 — don't exceed)

#### Free / Trial
- What's included: <minimal — enough to get value once>
- Limit: <quota / time / feature gate>
- Goal: <discovery | activation | demonstrate value before ask>

#### Tier 1: <name> — $X/mo (or $Y/yr — save Z%)
- Target user: <specific>
- What's included: <feature list>
- Limit: <if applicable>

#### Tier 2: <name> — $X/mo
- Target user: <specific — different from tier 1>
- What's included: <delta from tier 1>

#### Enterprise (talk to us)
- Triggers: >N seats / SSO / custom contract / vendor onboarding

### Anchor pricing
The $X/mo number is anchored on:
- Value: customer saves <Y hours/$> per month → ~$Z value created → priced at 10-20% capture
- Competition: comparable tools price $A–$B; we're at <where in range>
- Willingness to pay: <van Westendorp data if collected, else qualitative from interviews>

### Free trial structure
- Length: 14 days | 30 days | reverse trial (start with full features, downgrade if not paid)
- Credit card required: yes | no
- Reasoning: <yes for higher-LTV B2B; no for indie/PLG>

### Annual incentive
- Discount: <typically 15–20% off monthly × 12>
- Why: cash up front + lower churn risk

### Stripe setup
- Products: <list>
- Prices: <list with IDs once created>
- Webhook events to handle: subscription.created, subscription.updated, subscription.deleted, invoice.payment_failed

### Test plan
- A/B variants to consider after first 30 days of data:
  - Tier 1 price ±$5
  - Free → Trial conversion (require CC vs not)
  - Annual discount %
```

## Protocol
1. Read `prd.md`, `metrics.md`, `interviews/synthesis-*.md`.
2. Identify the value created — quantified if possible (hours saved, $ saved, $ earned).
3. Pick the model. Default: monthly + annual subscription with one free tier (or trial).
4. Tier the offering. Tier 1 is the "I'm a user" tier — 80% of paying users land here. Tier 2 is the "team" tier.
5. Set the anchor number using: 10–20% of value created. Cross-check against competition. Cross-check against any willingness-to-pay signal from interviews.
6. Recommend trial structure based on user type (B2B → 14-day with CC; B2C → freemium often better).
7. Annual discount: default 15–20%.
8. List exact Stripe setup so the dev can wire it.

## Contract (handoff → pricing-optimizer, one quarter later)
Per `factory/playbooks/mvp-stage/agent-handoff-contracts.md`:
- **Produces:** `products/<name>/launch/pricing.md` in the format above, plus the Stripe products/prices to create.
- **Shape:** guaranteed sections — model type / tiers (≤3 paid) / anchor with its three sources (value calc, competitive range, WTP evidence) / trial structure / annual incentive.
- **Preconditions:** interview synthesis or equivalent WTP signal where it exists; competitive pricing scan.
- **Postconditions:** the doc ends with a **"signals to start logging today"** list — conversion rate by tier, "too expensive" / "would pay more" mentions, tier-upgrade rate, churn by tier — so `pricing-optimizer` inherits a full quarter of data on day 91 instead of starting blind.
- **On gap:** no WTP data and no comparable competitors → anchor on value-based math alone, say so explicitly, and mark the price as a test with a revisit date — never present a guess as triangulation.

## Sources
- `factory/playbooks/launch-stage/pricing-playbook.md`
- The product's `interviews/synthesis-*.md` (for WTP signals)

## What you don't do
- Don't recommend more than 3 paid tiers + maybe a free tier. Decision paralysis kills conversion.
- Don't set price by gut. Anchor on value created.
- Don't recommend permanent free tiers without a clear conversion mechanism — usually trial > freemium for indie.