App Store screenshot A/B testing: what to test first

Test the order before the content

If your first three screenshots are all strong, rearranging them costs nothing and can be worth more than rewriting any of them. The same artwork, in a different order, tells you which claim your audience actually responds to.

Start there. A reorder is one export and one upload; a rewrite is a design cycle.

Then test the headline on slide one

Once the order is settled, the first slide's headline is the next test. Keep the artwork identical and change only the words. If the result moves, you have learned something about your market that is worth more than the click itself.

Change one thing per test. A new headline and a new colour at the same time tells you nothing.

Then the artwork, not the whole slide

By this point you know what the headline should say. Now test whether the picture behind it helps. Swap a device frame for a full-bleed canvas, or replace a busy screenshot with a simpler one. The words stay put.

What a test cannot tell you

A screenshot test tells you which version of a thing people respond to. It does not tell you why, and it does not tell you anything about the part of the listing above the screenshots — the icon, the title, the first three lines of the description.

It also does not generalise. A headline that wins in the United States may lose in Germany, where the same words are longer and the reading order is different. Test a market, not a keyword.

How long to run it

Long enough to be a signal, short enough that you can still act on it. For most listings, a test needs to see a few thousand impressions on each side before the difference means anything — roughly one to two weeks of meaningful traffic.

If your app does not get enough traffic for that, a test is not the right tool yet. Ship the obvious improvement, watch the store listing data, and come back when you have volume.