Pricing · Worked example
An annual discount is a bet on retention
At $20 per month, a $144 annual plan equals 7.2 monthly payments. The eighth monthly payment takes receipts to $160, so the annual price is already $16 behind for a customer who would have kept paying.
That is the retention bet inside the discount. A short paywall test can show that the annual offer brings in more cash now. It cannot show how many monthly payments those same customers would have made.
Start with the retention break-even
In this hypothetical example, monthly access costs $20 and the annual plan costs $144, 40% below twelve monthly payments.
Retention break-even: $144 annual price ÷ $20 monthly price = 7.2 monthly payments
If a future annual buyer would have made fewer than 7.2 monthly payments, annual collects more from that person. If they would have made eight or more, monthly collects more.
| Payments | Monthly | Annual | Change |
|---|---|---|---|
| 7 | $140 | $144 | +$4 |
| 8 | $160 | $144 | −$16 |
| 12 | $240 | $144 | −$96 |
Hypothetical prices. The comparison ignores refunds, payment fees, servicing costs and the time value of earlier cash.
This calculation applies only to someone who would otherwise have paid monthly. An annual offer can also attract a customer who would not have bought monthly at all. That is why the formula sets a useful threshold but cannot predict the result of the whole offer by itself.
Compare offers, not the people who selected them
A customer willing to prepay for a year may already expect to use the product for a long time. They might have stayed longer than the average monthly subscriber even without the discount. If you use average monthly tenure to justify the annual price, you could end up discounting the customers who would have kept paying full price.
Randomize who sees the annual offer, preserve a control that does not see it, and compare cumulative net contribution across all assigned users. Keep non-buyers in both groups. This captures both mechanisms: money pulled forward from would-be monthly customers and new money from people who would not have bought monthly.
Microsoft Research's work on long-term experiments describes the broader problem: a short-term revenue increase can diverge from lifetime value when later behavior changes.
What to decide before retention matures
Cash flow can be a sufficient business reason. Earlier cash may extend runway or bring acquisition payback inside a financing window. In that case, make the decision explicit: accept unresolved long-term value in exchange for a specific near-term cash benefit, account for refund liability, and limit exposure.
Keep the randomized holdout. Month eight is the first useful check against the 7.2-payment threshold, not the end of the analysis. Continue through the full horizon that matters to the pricing decision. Until then, report the early result as upfront cash per assigned user, not higher LTV.