Games sell a moment, utilities sell an outcome. Game screenshots should lead with the most visually striking scene and minimal text, because players are judging whether the world looks fun. Utility screenshots should lead with a captioned benefit, because the user is checking whether a specific problem gets solved. Copying a productivity listing's caption-heavy layout onto a game listing is one of the most common and expensive ASO mistakes.
Why is the same advice wrong for both?
Nearly every ASO article treats screenshots as one problem with one answer. Add captions, lead with your strongest benefit, keep text large.
That advice is correct for productivity apps and actively harmful for games, and I've watched a studio tank a launch by following it faithfully.
The difference isn't cosmetic. It comes from what the person scrolling is actually doing. A user searching for an expense tracker has a problem and is checking whether you solve it. A user browsing games has no problem. They're looking for something that looks fun, and "fun" is communicated visually or not at all.
What is the user actually deciding?
Utility user: "Does this do the specific thing I need?" They arrived from a search with intent. They have a task. Your screenshots are evidence, and they're reading them the way you'd read a product spec.
Text works here because they want information. A caption saying "Scan receipts with your camera" answers a question they came with. Interface clarity matters because they're evaluating whether it'll be usable.
Game user: "Does this look fun?" They arrived from browsing, a chart, or an ad. They have no task. Your screenshots are a mood board, and they're scanning them the way you'd scan a film poster.
Text mostly gets in the way. A caption saying "Battle epic monsters" tells them nothing they can't see, and it covers the art that was doing the actual selling. What they want is a glimpse of the world, the characters, and the visual style.
That single distinction drives everything else. A game listing full of captioned UI shots reads as a game with weak visuals, which is a fatal impression regardless of how good the game is.
The mechanics of getting either type framed cleanly are the same, which is why the device frame library is worth a look whichever category you're in.
Where should each one put the hook?
Games should open with the single most visually impressive scene in the entire product. Not the tutorial, not the menu, not a balanced representation of the game. The best looking moment.
Frame two should establish variety, showing that the world is bigger than one screen. Different environment, different character, different visual register.
Resist the urge to explain mechanics in frames one and two. Mechanics can live in frames four and five for the people who scrolled, and those people are already interested.
Utilities should open with the primary benefit stated as a caption over a clean interface shot. The caption does the work, the screenshot proves it's real.
Frame two should show the second most common use case, not a secondary feature you're proud of. There's a difference, and teams get it wrong constantly because they pick based on engineering effort rather than user frequency.
For both, one habit is universal: look at your listing at actual thumbnail size on a phone, not at full size on a monitor. Most screenshot design fails at the size it's actually viewed, which is a point I've laboured before in the design mistakes writeup.
How much text is too much?
"Track every expense automatically" works. "Our powerful expense tracking engine categorises your spending across accounts" does not, because nobody reads that at 200 pixels wide.
For games, either no text at all, or two to four words used as a stylistic element rather than an explanation. Text in a game screenshot should look like part of the art direction, matching the game's own typography and colour, not like a caption bolted on top.
The failure mode I see most in game listings is a white sans-serif caption on a semi-transparent black bar sitting over hand-painted fantasy art. It looks like a stock template because it is one, and it signals low production values in a category where production values are the product.
Localisation is where this bites twice. Long captions that fit in English overflow in German and Russian, and a game caption that was already decorative becomes actively ugly. If you're shipping to multiple markets, design the caption box for the longest language you support rather than the shortest.
Building the variants for that is tedious by hand, which is exactly the case for doing it in the mockup studio where the frame and layout stay fixed and only the text and screen swap.
What the store itself does differently
Games get merchandised differently. Editorial collections, themed rows, seasonal features. A lot of game discovery happens inside curated surfaces where your icon and first screenshot appear alongside competitors in a grid. That grid context is brutal, because your asset is judged against five others simultaneously rather than on its own.
Utilities get found through search far more often, where your listing is usually evaluated alone after a click. Different competitive context, different design pressure.
The practical consequence: for a game, check how your first screenshot looks tiled next to four competitors in your genre. Screenshot their listings, put yours in the middle, and look at the row. If yours disappears, the problem is contrast and composition rather than content.
For a utility, that exercise matters less. What matters more is whether your caption answers the exact query people used to find you. If you rank for "pdf to word" and your first caption says "Document management, simplified", you're losing people who were one word away from converting.
I've done both exercises on real listings and the game one is consistently more uncomfortable. Studios tend to have strong art and weak composition at small sizes, so the asset that looks gorgeous full screen turns into visual mush in a row of six. Cropping tighter on a single character usually fixes it faster than redesigning anything.
What about hybrid apps?
The test I'd apply: does someone search for this by name of the problem, or do they discover it by browsing?
Search-driven discovery means utility rules. Even a beautifully designed photo editor is usually found by someone searching "background remover", and that person wants to know you remove backgrounds.
Browse-driven discovery means game rules. If people find you in a chart or a themed collection, they're in an evaluative visual mode.
If it's genuinely both, lead utility and reward browsers later. Frames one and two state the benefit clearly, frames three onward can show personality, character and delight. That ordering protects your search traffic, which is usually the more reliable channel, while still giving browsers something to like.
One more consideration that catches indie teams. A game with a strong art style can afford to be mysterious. A game with generic art cannot, and should lean on showing recognisable mechanics instead, because "looks fun" has to come from somewhere. Be honest about which one you have. Testing that honestly rather than guessing is what the A/B testing walkthrough is for, and Apple's own product page guidance is worth reading before you commit a set.
Frequently asked questions
Should game screenshots have text captions?
Usually not, or at most two to four words treated as a design element. Players are judging whether the world looks fun, and captions cover the art that communicates that. If you do use text, match the game's own typography and colour so it reads as art direction rather than a template overlay.
What should the first screenshot be for a utility app?
The primary benefit stated as a short caption over a clean interface shot. The user arrived with a specific problem and is checking whether you solve it, so the caption answers their question and the screenshot proves it's real. Three to six words is the practical limit.
Why do game listings need a different strategy?
Because the user has no task. Utility users search with intent and read screenshots as evidence, while game browsers have no problem to solve and scan visually for appeal. A game listing full of captioned interface shots reads as a game with weak visuals, which is fatal regardless of quality.
How do I handle screenshots for a hybrid app?
Ask whether people find you by searching a problem or by browsing. Search-driven discovery means follow utility rules, browse-driven means game rules. If it's genuinely both, lead with a clear benefit in frames one and two to protect search traffic, then show personality from frame three onward.
How many words fit in a screenshot caption?
Three to six for a utility app, which reads clearly at thumbnail size. Design the caption box for the longest language you ship in rather than English, since German and Russian commonly overflow layouts built for English. Building those variants is much faster in the [mockup studio](/create) than by hand.
Does screenshot order matter more than screenshot quality?
Order matters enormously because most users only see the first two frames without scrolling. A brilliant fifth screenshot is close to invisible. Get the strongest asset into frame one for games, or the clearest benefit for utilities, and treat everything after frame two as being for people already interested.
Build the mockup in your browser.
Drop a screenshot into a real device frame and export at the exact store size — free, no signup.