Published July 2026 · 8 min read

Screenshot captions
that actually convert.

A store browser gives your listing about three seconds. In that window they read two captions, maybe three, at thumbnail size. Those captions are the highest-leverage copy you will ever write for your app, and most indies write them last, in the upload form, at midnight before release. Here's how to write them on purpose.

Visible in search first 2 to 3 shots Caption length 3 to 6 words Reading distance thumbnail scale One idea per screenshot

Screenshots are billboards, not documentation

The mental model that fixes most bad screenshot copy: nobody is using your app yet. They're driving past it. A billboard doesn't explain the gearbox; it tells you why to pull over. Your screenshot's caption exists to make one promise, in one glance, at a size where body text is unreadable.

That has a hard implication. The screenshot's UI is evidence, and the caption is the claim. If the caption just narrates the UI ("Home screen with your tasks"), it wastes the claim. If the caption makes a promise the UI doesn't visibly support ("Get promoted faster"), it reads as marketing gas. The pair only works when the caption says why it matters and the pixels prove it's real.

The first three screenshots do all the work

In App Store search results, your listing shows up to three portrait screenshots inline, before anyone taps through. On Google Play, search cards show fewer, and the listing page shows a horizontal strip. Either way, the numbers are brutal: most installs happen without the user ever seeing screenshot four.

So sequence the first three like an argument, not a feature tour:

Shots four through ten are for the tapped-through reader who is already half-convinced. Depth features, settings, edge cases, the second platform, they live here, in whatever order supports the story.

Caption formulas that survive thumbnail scale

Almost every caption that converts fits one of four shapes. Weak captions are usually a good shape buried under extra words.

Rewrites, weak to strong

Same app, same feature
ShapeWeakStrong
Verb-first command A powerful way to track your habits easily Track habits in one tap
Outcome, not feature Advanced sleep-cycle analysis engine Wake up at the right moment
Number as proof Supports many currencies 40 currencies, live rates
Kill the objection Your data is handled securely Everything stays on your phone

Notice what got deleted in every rewrite: easily, powerful, advanced, seamlessly, and the word app itself. These words are dead weight, every listing uses them, so no listing benefits from them. A caption earns its adjectives with numbers or loses them.

!

One caption, one idea. The moment a caption contains "and", split it into two screenshots, you have ten slots, and clarity beats economy.

The type-size floor

Captions fail at the render stage more often than the writing stage. A store search card is roughly a sixth of your export's width, a 1290×2796 screenshot displays at around 215 points wide. At that scale, caption text below roughly 90 to 100px in the export is decoration, not copy.

The practical test: export the screenshot, view it at one-sixth size, and squint. If the caption isn't instantly legible, don't shrink the claim, shorten it. Four big words beat nine small ones every time. This is also the strongest argument for the 3 to 6 word budget: it's not a style preference, it's what physically fits at a legible size.

Caption and UI: division of labor

The most common layout mistake is redundancy: a caption that says "Beautiful statistics" above a screen full of statistics. The reader gets the same information twice and the "so what" never arrives. Assign the jobs cleanly:

If your UI screenshot contains its own prominent text, an empty-state message, a big onboarding headline, make sure it doesn't fight the caption. Two headlines on one billboard means no headline.

Sentence case, punctuation, and other small calls

Before you upload: the ten-second test

Put your first three screenshots side by side at thumbnail size and ask someone who has never seen your app: "What does this do, and why would you want it?" If they can answer both in ten seconds, ship. If they answer with your app's category, "it's a to-do app?", your captions are describing the genre, not the reason to pick you.

Captions are also the highest-return thing to localize, because they carry the whole argument. If you translate one thing in your listing, translate these, the localization guide covers the workflow.