Pricing
Set prices that are sustainable, customer-friendly, and iterative. Charge something from day one.
Use This Skill When
- User is setting initial pricing for a product/service
- User is considering raising or lowering prices
- User is deciding between pricing models
- User wants to understand pricing psychology
- User says phrases like "what should I charge", "pricing strategy", "is this price fair", "how do I price"
When NOT to Use
- User is building pricing page design (use design skill)
- User is setting up complex billing/invoicing (use payments skill)
- User is analyzing competitor pricing only (use market research skill)
Context: Pricing Maturity
Manual pricing: Asking each customer "what can you pay?" or guessing randomly
Cost-based: Costs + 20–50% margin; one price
Market-aligned: Pricing similar to alternatives; justified by value
Tiered/optimized: Multiple tiers; prices raised over time; LTV-optimized → Target
Core Principle
Charge something. Always. There is a psychological cliff between free and $1 (the "zero-price effect"). Customers will line up for free products but won't engage with them. Pricing signals value. Without pricing, you can't survive or learn what customers truly value.
Instructions
Understanding Pricing Psychology
The Zero-Price Effect
- Free: Unlimited perceived demand, zero commitment, low engagement
- $1+: Creates selection friction (good!), signals value, enables commitment
- Implication: Never offer free as your default; always require payment
Anchor Pricing
- First price you see anchors all subsequent price judgments
- Start at a reasonable price (not too low, which signals low value; not absurd)
- Customers compare to this anchor, not to absolute value
Tiered Pricing Psychology
- Customers will choose the middle tier (not cheapest, not most expensive)
- This means offering 3 tiers lets you capture more value from different segments
- Example: Spotify Free → Premium → Premium Family
Pricing Models
Choose one, then evolve:
Model 1: Cost-Based Pricing
Formula: Costs + Margin = Price
| Cost Category | Example | Notes |
|---|---|---|
| Variable | Payment processing (2.9% + 30¢), hosting per user, materials | Scales with sales |
| Fixed | Your salary, base hosting, tools, domain | Constant regardless of volume |
| Margin | 20–100% above costs | Higher for digital products (100%+); lower for physical (20–30%) |
When to use:
- Physical products with clear material costs
- Services with hourly labor
- Need to ensure baseline profitability
Example:
- Your cost: $3 per unit
- Retail margin: 50% markup
- Price: $4.50 per unit
Model 2: Value-Based Pricing
Formula: Price = Customer Perceived Value
| Customer Situation | Value | Price Justification |
|---|---|---|
| Saves 5 hours/week at $50/hr | $250/week value | Charge $50–100/month |
| Increases revenue by $10k/year | $10k value | Charge $1–3k/year |
| Solves critical problem (pain avoidance) | Pain cost | Charge based on pain magnitude |
When to use:
- Software/digital products (low marginal cost)
- High-value services (consulting, training)
- Clear ROI or pain-based value
Example:
- Customer loses $500/week if system is down
- Your product saves them 50% downtime risk = $250/week value
- Price: $50–100/week ($2.6k–5.2k/year) is still a 5–10x ROI for customer
Model 3: Hybrid
Use both: Cost-based floor + value-based ceiling
- Price must cover costs (floor)
- Price is capped by customer willingness to pay (ceiling)
- Negotiate within this band based on customer value
Setting Your Initial Price
Step 1: Know Your Costs
Variable cost per customer: $X
Fixed costs per month: $Y
Target profit margin: Z%
Price = (Variable cost × (1 + Z%)) + (Fixed cost / Expected customers)
Step 2: Research Alternatives
- What do competing/alternative solutions charge?
- What would customers currently pay (directly or indirectly) to solve this?
- What's the "not-a-no-brainer" price? (Above that, they'll just suffer and not buy)
Step 3: Calculate Customer Payback
How much revenue do you need per month? $X
At your price point, how many customers is that? $X ÷ Price = Y customers
At 1 new customer per business day, when do you reach Y?
(260 business days per year = 20/month)
Example: $10/month price, need $2,000/month = 200 customers = 10 months
Step 4: Validate with Customers
- Ask early customers: "Would you pay $X for this?" (Not "Is this price fair?")
- Offer pricing tiers; see which people choose
- A/B test prices with new customer cohorts
- Listen to price objections carefully
Pricing Strategies
Strategy 1: Start Low, Raise Over Time
- Rationale: You'll improve the product; prices going up is normal and expected
- Implementation: Price increases every 6–12 months as you add value
- Example: Price stays $5/month for existing customers; new customers pay $10/month
Strategy 2: Free Trial + Paid Conversion
- Best practice: 7–14 day free trial (long enough to see value)
- Clear conversion: Email reminder before trial ends
- Pricing transparency: Show price before trial, not after
- Pitfall to avoid: Free trials shouldn't be so easy that paid is a shock
Strategy 3: Freemium (Free + Paid Tiers)
- Free tier: Limited features; enough value to understand product but not solve entire problem
- Paid tier: Adds premium features, support, higher limits
- Conversion rate: Expect 1–5% of free users to convert to paid
- Risk: Freemium can trap you in low prices if free users dominate
Strategy 4: Tiered Pricing (Recommended for Scale)
- Tier 1 (Budget): Lower price, basic features, self-serve
- Tier 2 (Standard): Mid price, core features, email support
- Tier 3 (Premium): Higher price, all features, priority support, customization
Psychology: Customers tend to choose the middle tier, so:
- Price Tier 1 at cost + minimal margin (to compete)
- Price Tier 2 as your target (where most customers land)
- Price Tier 3 much higher (shows value of full product)
Pricing Over Time
| Timeline | Action | Rationale |
|---|---|---|
| Launch | Set initial price (conservative) | Get customers; gather feedback |
| 3 months | Review feedback; consider +20% for new customers | Learning what's valued |
| 6 months | Raise prices; grandfather existing customers | Product improved; market validates value |
| 12 months | Introduce tiered pricing if successful | Capture different customer segments |
| 2+ years | Continuously raise top tier; keep budget tier stable | Extract more value from power users |
Key rule: Raise prices incrementally and transparently. Existing customers get grace period; new customers pay new price.
Profitability Math
Help users understand when they can stop trading time for money:
Monthly revenue needed: $X
Price point: $Y/month
Customers needed: X ÷ Y = C
At 1 new customer per business day (20/month):
Months to profitability: C ÷ 20 = M months
At 2 new customers per business day (40/month):
Months to profitability: C ÷ 40 = M months (faster)
Example:
- Need $3,000/month
- Price: $30/month
- Customers needed: 100
- At 1/day: 5 months to profitability
- At 2/day: 2.5 months to profitability
Common Pricing Mistakes
- Too low: Signals low value; insufficient margin for growth
- Too high: Prevents customers from saying yes; not aligned with perceived value
- No tiering: Leaves money on the table; one-size-fits-all
- Constantly changing: Confuses customers; makes them avoid commitment
- Apologizing for price: Undermines value; confuses customers on what's worth
- Freemium without exit strategy: Traps you in low prices; hard to convert free users
Output
For each pricing decision, provide:
Pricing model (Cost-based / Value-based / Hybrid with explanation)
Initial price point with breakdown:
- Variable costs per unit
- Fixed costs
- Margin
- Final price + rationale
Profitability math:
- Customers needed for financial independence
- Timeline at current sales pace
- Breakeven timeline
Tier structure (if applicable):
- Budget tier
- Standard tier (recommended)
- Premium tier
Launch strategy:
- Free trial? (Length and rules)
- Freemium? (Free tier limits)
- None of the above? (Paid from day one)
Price increase schedule: When and how much to raise