App Store Screenshot Refresh Cadence by Category [2026]

Mobile games refresh App Store screenshots roughly twice as often as productivity apps, and casino games refresh roughly 6 to 8 times as often as the median app, because three things vary by category: how often a signature event hits the calendar, how volatile the category's keyword set is, and how long an AB test needs to reach 90% confidence at the app's install volume [1][2]. A single universal cadence ignores all three. The right number depends on the category you're in.

The cadence guide covers the universal trigger checklist and the 4-week keyword stabilization window. This piece splits that framework by category and gives you a defensible per-category number, not a generic "every 3 months."

TL;DR:

• Apptweak's 2024 data: 49% of top App Store apps updated screenshots at least twice in the year; 57% of top Google Play games AB tested screenshots at least twice; casino games on Google Play updated screenshots 13 times in a year [1]. Most non-game App Store categories average under 4 updates per year.
• Three axes split cadence by category: signature-event frequency (games have weekly events; utilities have none), keyword volatility (gaming keyword churn is multiple times higher than productivity), and AB-test runway needed to reach Apple's 90% Bayesian confidence threshold at the app's install volume [2].
• Apple PPO test duration is not fixed. Apple uses Bayesian analysis with a 90% confidence cap; the time to significance "depends on your traffic volume and the magnitude of improvement" [2]. Tests need at least 5 first-time downloads to appear in Analytics [2].
• Per-category cadence matrix: Mobile games (4-12+ per year), Subscription apps (4-6), Health and fitness (4 seasonal anchors), Finance and fintech (3-4 tied to regulatory calendar), Productivity (2-3), Utility (1-2).
• Solo indie default is 4 per year if your category isn't surveyed; adjust up for games and seasonal categories, adjust down for utility apps with stable feature sets.

Table of Contents

• How often do top apps actually refresh their screenshots by category?
• What three axes split refresh cadence by category?
• How often should mobile games refresh App Store screenshots?
• How often should subscription, health, and seasonal apps refresh their screenshots?
• How often should productivity, utility, and finance apps refresh their screenshots?
• How long does an App Store screenshot test need to run by category?
• What's the right per-category cadence for a solo indie developer?
• Ship by category, not by calendar

How often do top apps actually refresh their screenshots by category?

Apptweak's 2024 benchmark study found 49% of top App Store apps updated their screenshots at least twice in the year, and 57% of top Google Play games AB tested screenshots at least twice [1]. Update frequency was highest in Books, Sports, and Graphics & Design on the App Store, and in Photography, Art & Design, and Education on Google Play. Casino games on Google Play updated screenshots 13 times in a single year, the highest individual category number Apptweak surfaced [1].

The aggregate hides a wider spread. Apptweak notes that most non-game App Store categories still average under 4 updates per year, while gaming sub-genres routinely exceed that [1]. Update frequency correlates with category-level conversion competitiveness: the more apps in a category are running AB tests, the more pressure on every other app to keep up or lose visible rank position. ASO World's April 2026 volatility analysis recorded 16,170 keyword ranking drops in a single day in the US App Store, evidence that the floor under any category's rankings can shift faster than a quarterly refresh cadence keeps up with.

This isn't a "games refresh more, apps refresh less" story. It's a "your category's signature-event calendar and keyword volatility set your floor" story. Games sit at the high end because their floor is high. Productivity sits at the low end because its floor is low. Indie apps in any category should be at least at the category median if they want their conversion rate to keep up with the cohort they're being benchmarked against.

What three axes split refresh cadence by category?

Three things vary by category and together determine the right refresh number: signature-event frequency, keyword volatility, and AB-test runway. Get all three on the table and the cadence falls out of the math, not out of a calendar. None of them is universal; treating refresh as "every N months" without checking these three is how indie apps end up under-shipping in a high-cadence category or over-shipping in a stable one.

Signature-event frequency is how often something happens in your category that your audience expects to see reflected on your store page. For mobile games, this is weekly to monthly: seasonal events, live ops drops, new characters, new game modes, expansion content. Top games refresh metadata at least monthly, timed around content updates and seasonal events. For a finance app, the signature events are slower: tax season, end of fiscal year, Apple's annual Review Guidelines refresh, occasionally a major regulatory change. For a utility app like a flashlight or unit converter, signature events are essentially zero. Match cadence to the actual event frequency, not to a generic calendar.

Keyword volatility is how often the keywords your screenshots reinforce shift their search volume or competitive set. Gaming categories have the highest keyword churn because new game launches, seasonal trending titles, and platform feature releases reshuffle the SERP weekly. Productivity, finance, and utility have far more stable keyword sets, which means screenshots that reinforced the right keywords two quarters ago are likely still reinforcing the right keywords today. The keyword research workflow covers how to audit which keywords your screenshots currently carry.

AB-test runway is the calendar time you need to reach Apple's 90% Bayesian confidence cap at your app's install volume [2]. Apple's docs are explicit: "Test duration will depend on the level of improvement detected" and "time to significance depends on your traffic volume and the magnitude of improvement" [2]. A 5,000-installs-per-day app reaches 90% confidence on a 10% lift in under a week; a 50-installs-per-day app needs many weeks for the same lift to show up as significant. The PPO testing guide covers how to design tests against this constraint. Your refresh cadence is capped above by your AB-test runway: ship a new variant before the previous test reaches significance and you erase your own measurement.

How often should mobile games refresh App Store screenshots?

Mobile games should refresh App Store screenshots 4 to 12+ times per year, scaling with sub-genre. Live-ops-driven games (RPGs, match-3, casino slots) sit at the high end (monthly or more frequent), tied to in-game events and live content drops. Premium and indie games without live ops sit at the low end (3 to 4 per year), tied to expansion releases and seasonal moments. Casino games on Google Play hit 13 updates per year in Apptweak's 2024 data, the highest published category number [1].

The signature-event axis explains most of the variance. A live-ops game gets a new event every 2 to 6 weeks, and each event is worth a frame swap (often frame 1 only, swapped to feature the current event's hero character or theme). Mobile gaming industry guidance suggests rotating screenshots to feature seasonal events and new content updates every 4 to 6 weeks for top performers. A premium game with no live ops has nothing comparable on the calendar; refreshing every month would be churn for its own sake.

Keyword volatility pushes games higher too. Gaming SERPs reshuffle weekly because new game launches enter the top charts constantly. A puzzle game whose first screenshot reinforced "match-3 puzzle" in January 2026 might find that "match-3 puzzle" is now competing with three new high-velocity entries by April, and a refresh that re-anchors the screenshot copy to a more current sub-genre term (e.g., "cozy puzzle game") buys back keyword reinforcement that the original screenshot was losing.

AB-test runway is rarely the bottleneck for games at scale. A top-100 game typically has the install volume to reach 90% confidence within 1 to 2 weeks for a 5%+ lift [2]. An indie game with 20 installs per day has the same constraint as any indie app: tests take many weeks, and the cadence is capped by that runway, not by ambition. The audit framework helps you check which frames in your current set are losing keyword coverage before you commit to a refresh.

How often should subscription, health, and seasonal apps refresh their screenshots?

Subscription apps and health and fitness apps should refresh 4 to 6 times per year, anchored to two predictable beats: paywall-pre-sell iteration cycles and seasonal demand spikes. Subscription apps run quarterly paywall AB tests as the default (frame 1 is the pre-sell surface; iterating it against new conversion benchmarks every quarter is standard). Health and fitness apps anchor cadence to seasonal demand: January (New Year resolutions), May to June (summer body), September (back-to-routine), November (pre-holiday).

Subscription apps have the strongest paywall-pre-sell signal in the frame 1 surface. The subscription app frame 1 playbook covers how the first frame functions as a paywall preview, and why iterating it on a quarterly cycle (tied to paywall AB test outcomes) outperforms calendar-based refresh. Across the indie subscription apps that came through our builder in 2026, the pattern we observed is the same one Apptweak surfaces at category aggregate: frame 1 swaps happen 3 to 4 times more frequently than full-set redesigns.

Health and fitness apps have the cleanest seasonal calendar of any category outside gaming. Fitness category iteration patterns in our /templates/category/fitness usage data spike around January and again around late May, with smaller bumps in September and November. The seasonal screenshot guide covers how to plan visual rotation 4 to 6 weeks ahead of each seasonal peak. The miss most indie health apps make is treating "summer body" as a single moment instead of a 6-week window, and refreshing too late in the window to capture the demand.

Wellness, meditation, and habit-tracker apps share the same seasonal calendar as fitness but with softer peaks. A 4-per-year cadence covers them comfortably. The wellness category page catalogs the design patterns that hold up across this seasonal surface, and the fitness category page covers the fitness-specific frame patterns.

How often should productivity, utility, and finance apps refresh their screenshots?

Productivity apps should refresh 2 to 3 times per year, utility apps 1 to 2 times per year, and finance and fintech apps 3 to 4 times per year, anchored to feature-launch cadence and a regulatory calendar that's slower and more predictable than gaming or subscription pacing. None of these categories benefits from monthly churn. All of them benefit from precision: refresh when the underlying product or rule actually changes, not on a generic timer.

Productivity apps have a moderate signature-event surface (feature launches, integration releases, occasional UI overhauls) but low keyword volatility. A note-taking app's keyword set has been stable for years; the screenshots that reinforce "markdown notes" or "second brain" continue to reinforce the same intent quarter over quarter. Across the productivity projects we see in /templates/category/productivity, the common refresh trigger is a major feature launch (AI integration, new collaboration mode, mobile-first redesign), not a calendar tick.

Utility apps (calculators, converters, weather, file managers, flashlights) have near-zero signature events and near-zero keyword volatility. Their cadence floor is whatever Apple's annual Review Guidelines refresh prompts (compliance check) plus one optional yearly visual modernization. Indie utility apps that refresh quarterly with no triggering event are wasting their AB-test budget on noise; better to save the runway for the moments when a real change ships.

Finance and fintech apps have a unique calendar: tax season (January to April for US tax prep, April to June for crypto and investing apps positioning for year-end statements), end-of-fiscal-year compliance refresh, and Apple's annual Review Guidelines refresh (which historically tightens financial claim language). The finance frame 1 playbook covers how trust signals carry the conversion weight in this category, and why finance screenshots survive longer between refreshes than other categories: an audited "FDIC-insured" or "SOC 2 Type II" trust signal doesn't go stale every quarter the way a game's seasonal event banner does.

How long does an App Store screenshot test need to run by category?

Apple's PPO test duration is not fixed. It depends on traffic volume and the magnitude of the lift Apple's Bayesian analysis is trying to confirm at 90% confidence [2]. The implication is that AB-test runway varies sharply by category not because Apple treats categories differently, but because traffic volume varies sharply by category. Gaming apps at scale reach significance in days. Indie apps in any category often need many weeks.

Apple's own documentation states three things explicitly: tests need at least 5 first-time downloads attributed to the test to appear in Analytics, the confidence threshold is 90% (lower than the 95% common in web AB testing), and "it requires fewer weeks of data to determine significance on higher improvement values (e.g. 30%) as opposed to lower improvement values (e.g. 5%)" [2]. Apple uses Bayesian techniques with prior information about the app's product page performance to update results daily [2]. The 4-week keyword stabilization window the cadence guide references is a separate constraint: even after an AB test completes, the keyword reshuffle following any indexed-field change takes about 4 weeks to settle [4]. Bundling a screenshot test with a subtitle change means waiting through both windows.

Practically, this means a category like casino games (where top apps see thousands of installs per day per locale) can finish a 90%-confidence test on a 10% lift inside 1 to 2 weeks and ship the next refresh within a month. An indie productivity app at 30 to 50 installs per day might wait 6 to 10 weeks for the same test to clear, which caps its annual refresh cadence at 3 to 4 tests per year regardless of how often the team wants to ship. The PPO testing guide covers test design within this constraint. The metadata-only update guide covers how to ship a refresh without burning a binary submission.

What's the right per-category cadence for a solo indie developer?

For a solo indie developer, the right number is the category median floor, not the category top-end ceiling. Top apps in your category have install volume to support fast AB-test cycles you don't. The category median number from Apptweak is a defensible target: about 2 refreshes per year for general App Store apps, 4 for fitness and subscription, 6+ for live-ops games. Stay at the median and your conversion rate keeps pace with the cohort you're being compared against in category rank.

Three pragmatic rules keep an indie cadence sustainable across categories:

• Audit before refreshing. Run your current set through an ASO audit before committing to a refresh date. Frame 1 carries most of the conversion weight in every category we've looked at, but the failure mode differs: fitness apps lose on the seasonal anchor, finance apps lose on the trust signal, productivity apps lose when a new feature outdates the screenshot's depicted UI. The audit tells you which frame is actually leaking conversion in your category, which is what triggers a refresh, not the calendar.
• Match refresh size to test runway. A solo indie with 30 installs per day shouldn't run 4 simultaneous PPO tests per year; pick the 2 highest-leverage swaps (almost always frame 1 plus one supporting frame) and run them sequentially. The metadata-only guide covers how to bundle a screenshot swap with a subtitle change without burning two submissions.
• Anchor to the category's signature events, not to a clock. Fitness anchors are January and June. Finance anchors are tax season and Apple's annual Review Guidelines refresh. Subscription anchors are quarterly paywall reviews. Game anchors are content drops. Pick the 2 to 4 events that actually matter in your category and refresh against those, not against "every 3 months."

The solo indie developer who nails per-category cadence ends up looking less like a designer running a redesign cycle and more like a product manager running a category-calibrated ASO program. The cadence isn't fast for fast's sake. It's calibrated to the category's event surface, keyword volatility, and the test runway the app's install volume actually supports.

Ship by category, not by calendar

The universal "refresh every 3 months" advice is a useful default when you don't know your category. Once you do know it, the right number is whatever sits at your category's median, scaled down if your install volume can't support the AB-test runway top apps in your category use. Games sit at the high end because their signature events and keyword volatility set the floor high. Utility apps sit at the low end because their floor is genuinely low. Finance, productivity, and subscription apps sit between, anchored to their category's specific event calendar.

The mistake to avoid is refreshing on a calendar you can't measure against. A test you shipped before the previous test reached significance erases its own signal; a refresh you shipped without a category-triggered event teaches the algorithm nothing. Ship by category, measure against the category median, and let your install volume tell you how fast you can actually move.