Experiment library
Test a pricing increase · method v1
Pricing increase with conversion and retention checks
Raise the price for a random share of new buyers and judge it on retained contribution, not first-month revenue.
- Hypothesis
- Buyers value the product above the current price; a higher price raises contribution per visitor without a proportionate fall in conversion or retention.
- Change
- Show the higher price to treated new visitors. Existing customers are not changed.
- Who is included
- New visitors reaching the pricing page during the test.
- Compared with
- New visitors who see the current price.
- Assignment
- Random split per visitor, persisted across visits.
- Measurement
- Retained contribution per pricing-page visitor per arm after the first renewal.
Worked example
3,000 visitors at 3% and $99 gives $8,910 a month. At $129 conversion can fall to about 2.3% before monthly revenue drops. Retention must also be checked at the real renewal timeline before deciding.
Evidence you need
- Price shown per visitor
- Paid conversions and refunds
- Renewals per cohort
Implementation steps
- Fix the test price and duration
- Persist assignment per visitor
- Record price paid per customer
- Wait through at least one renewal cycle
- Compare retained contribution per visitor
Cost components
- Possible lost conversions
- Support time for price questions
Stop if
- Treated conversion below the break-even rate for three consecutive weeks
Scale if
- Higher retained contribution per visitor after renewal
Does not apply when
- Very low traffic
- Markets where price discrimination is restricted
- Prices visible to existing customers who would see two prices
Adapt this experiment to your business
Your values stay in this browser until you sign in, then become a private protocol that records this method and version.