Skip to main content

Free developer tool · Updated September 2026

App Store A/B Test Duration Calculator

Find out how long a Product Page Optimization test needs before a variant can reach Apple’s 90% confidence, and whether it can finish inside the 90-day limit.

Free, no sign-up Updates as you type Planned for 90% confidence

Plan before you test

How long will your test take?

About Apple’s Product Page Optimization

/ day

Product page views per day for the localizations in the test, from App Store Connect Analytics.

%

Share of product page visitors who download your app today.

%

A relative change: +10% turns a 3% conversion rate into 3.3%.

Number of treatments

Variants compared against your original page. Apple allows up to three.

%

Split equally among the treatments; everyone else sees your original page.

Estimated duration

Turn on JavaScript to see the estimate.

Impressions needed per variant
–
Total test impressions
–
Daily impressions per variant
–

Apple doesn’t publish the exact statistical method behind Product Page Optimization. This is a planning estimate using a standard two-proportion test at 90% confidence and 80% power.

How it’s calculated
  1. p₁ is your current conversion rate, and p₂ = p₁ × (1 + lift) is the rate you want to detect.
  2. Impressions needed by each version (the original and every treatment), with zα = 1.645 for 90% confidence (two-sided) and zβ = 0.8416 for 80% power:
    n = (zα + zβ)² × (p₁(1 − p₁) + p₂(1 − p₂)) / (p₂ − p₁)²
  3. Each treatment gets daily impressions × traffic share ÷ treatments per day. The original gets daily impressions × (1 − traffic share).
  4. The duration is the number of days the slower of the two needs to collect n impressions, rounded up.
  5. The smallest detectable lift is the lift whose n fits into 90 days of the slower version’s impressions.
  6. A test finishes soonest when it gets treatments ÷ (treatments + 1) of your traffic: 50% for one treatment, 67% for two, 75% for three.

Worth running? Screenshot Studio designs the variant and uploads it straight into a Product Page Optimization test.

Download for macOS →
Know a developer planning an App Store test? Send them this free calculator.
Share on X

Test too slow?

Three ways to finish a test sooner

Duration depends on how big the difference is, how many ways you split your traffic, and how much traffic you have. Change any of them and the estimate above moves.

Lever 1

Test a bigger change

Small tweaks produce small lifts, and small lifts need huge samples: halving the lift you want to detect roughly quadruples the impressions needed. Test a new first screenshot, message, or style, not a color tweak.

Lever 2

Test fewer variants

Every extra treatment takes a share of the traffic. With modest traffic, run one bold treatment against the original, then test the next idea against the winner.

Lever 3

Send more traffic to the test

Raise the share of visitors in the test and include every localization that uses the same screenshots. Promoting the app while the test runs helps too.

From plan to live test

Build your test variants in minutes

Screenshot Studio makes new screenshot variants fast, exports every size and localization, and uploads them straight into a Product Page Optimization treatment in App Store Connect.