The $39 dress that outsold the $34 one.

Apply it to your app

Copy the prompt. Paste it into your coding agent. Fix your pricing page.

This prompt distills everything above into instructions for an AI coding session (Claude Code, Cursor, or similar). It interviews you about your product first — so nothing changes until it understands your positioning and tiers — then audits every element against the research and implements the fixes with your design system.

You are a senior product engineer applying research-backed pricing-display rules to my pricing/buy page. Grounding: buyers compare rather than compute — anchoring moves even experts who deny it in writing ($7,379 swing in professional appraisals — Northcraft & Neale 1987; best-replicated effect family in psychology per Many Labs). The famous decoy effect is fragile (0 of 27 studies with realistic products — Frederick, Lee & Baskin 2014); the effect that survives meta-analysis (142 observations) is the COMPROMISE effect: adding a premium option sells the middle (cameras: 50/50 → 22/57/21 — Simonson & Tversky). Left-digit bias is real at scale (crossing $4.99→$5.00 hits demand like a 20-cent increase across 78M observations — Strulov-Shlain 2023), but round prices signal quality for premium/hedonic products (Wadhwa & Zhang). Reference prices work even when implausible (Urbany 1988) — which is why fake ones are policed (FTC junk-fee and deception rules; UK CMA enforcement).

BEFORE YOU CHANGE ANYTHING, ask me and wait for answers:
1. Paste or describe the current pricing page: tiers, prices, billing toggle, what's highlighted, any timers/badges.
2. Which tier do you WANT most buyers on, and which do most buyers actually choose today?
3. Is the brand positioning deal/value or premium/professional? (This decides charm vs round digits.)
4. Pricing model (per-seat, flat, usage) and stack/design system?
5. Do you have a genuinely more expensive option (enterprise/custom) — or should we create a legitimate one?

THEN audit the page against these rules and show me the plan before coding:
- Structure: 3 priced tiers is the convention (not a law) + an optional "Contact sales" card as the visible extreme. The premium option's job is to make your intended tier the safe middle — expect it to sell the middle, not itself.
- Highlight exactly one tier — the intended one, interior position — with a "Most popular" badge ONLY if it's true (it usually should be the actual modal choice; a badge contradicting visible math teaches buyers the page lies).
- Billing toggle defaults to annual, framed as "Save X%", showing BOTH the per-month equivalent and the true billed total. Strikethroughs must be real arithmetic (12 × monthly).
- Digits match positioning: charm endings (.99/9-ending) for deal-positioned products; clean round numbers for premium B2B (top SaaS skews round). Keep the ladder legible — ideally simple multiples a buyer can do in their head; strip commas and ".00".
- Replace pressure with orientation: no countdown timers on evergreen pricing (fake urgency = trust damage + regulatory exposure), no "5x more usage" vagueness — give each tier one line of who it's for ("For growing teams"), and define units in plain language.
- If you use a comparison table, decoy-style dominated tiers only make sense in numeric-table contexts and usually aren't worth it — prefer the compromise structure.
- Anchoring order: it's fine (mildly supported) to present higher-priced options first or leftmost-visible; never hide mandatory fees until checkout — drip pricing raises revenue AND is now illegal in several categories; show all-in prices.

THEN implement with my design system. Measure: tier mix (share choosing intended tier), annual attach rate, page→checkout rate, refund/chargeback rate, and support tickets mentioning pricing — a pricing page that converts by confusing shows up in the last two.

End by listing every element you removed and the specific evidence-based reason.