App Store Optimization (ASO): The Complete Guide for 2026

App Store Optimization (ASO) is the work of making an app easier to find in App Store search and likelier to be downloaded once someone sees it. In practice that means choosing the words in your app name, subtitle and keyword field, and improving the icon, screenshots, ratings and reviews that people judge in a few seconds. It costs nothing but time, it keeps working on every search for as long as the app is live, and most of it happens inside App Store Connect.

This app store optimization guide is written for indie developers and small teams. It covers every listing field with the limit Apple publishes, keyword research step by step, ratings and reviews, conversion, localization, Google Play, an ASO checklist and how to measure the result. Every Apple rule links to the Apple page it comes from, and every limit and price here was checked on October 13, 2026.

One rule runs through the whole guide. Apple documents part of what App Store search uses and keeps the rest private, so each ranking claim below is marked as either documented by Apple or inferred by the industry, with the name of who infers it. For the wider plan around the store (positioning, launch, paid ads and retention), read how to market an app.

What is app store optimization (ASO)?

ASO has two halves. The first is visibility: getting your app into the results for the searches your future users type. The second is conversion: getting the person who sees your app in those results to tap Get. A listing that ranks first and converts nobody is as useless as a perfect page that nobody finds, so the two are always worked on together.

Search is where most of this happens. Apple's advertising site says 70% of App Store visitors use search to discover apps and almost 65% of downloads happen directly after a search (Apple's worldwide figures for 2022). The same page puts the audience at over 850 million visitors a week in 2026. Your listing is the only thing those people see before they decide.

Visibility: showing up for the right searches

Visibility comes from text and from behavior. The text is the short set of fields search reads, covered one by one below. The behavior is what people do after they see you: download, rate, review. You control the text directly and the behavior only through the quality of the app and the page.

Conversion: turning an impression into a download

Conversion is decided mostly in the search results list, before anyone opens your product page. The card shows your icon, name, subtitle, rating and the first screenshots, and a person scrolling gives it a second or two. That is why the icon and the first three screenshots get as much attention in this guide as the keyword field.

What ASO cannot do

ASO cannot rescue an app people stop using. Ratings and reviews are part of what Apple says search looks at, and both come from the product. It also cannot create demand that does not exist: if nobody searches for what the app does, a better keyword field changes little, and you need channels outside the store. Our guide on how to get app downloads covers those.

ASO vs SEO: what carries over and what does not

If you have done search engine optimization for a website, the habit of researching what people type carries over to ASO. Almost nothing else does. A web page can be any length and gains from links on other sites. An App Store listing gets a few short fields, one of them hidden, and is judged on downloads and ratings.

The two meet in one place. Your product page is also a web page, and Apple's version information reference says the description is used for web search engine results once the app is released. So the description does SEO work on Google and conversion work in the store, even though Apple does not list it among the fields App Store search matches.

Google Play sits closer to SEO than the App Store does. It has no keyword field, and Google's own help page tells developers to use SEO practices in the description. The Google Play section below has the details.

App Store ranking factors: what Apple documents

If you want to know how to rank higher in the App Store, start with what Apple itself says. It does not publish the App Store ranking algorithm, but it does publish a short list of inputs. Apple's App Store search page says results are based on a number of factors, including text relevance (matches for your app's title, subtitle, keywords and primary category) and user behavior (downloads, ratings and reviews, plus other signals it does not name). That one sentence is the firmest ground in all of ASO.

A second Apple page words it a little differently. Apple's discovery overview lists text relevance as the title, keywords and primary category, and describes customer behavior as downloads and the number and quality of ratings and reviews. Read together, the two pages agree on the shape: a handful of text fields, then proof that people want the app.

Two details in that table are often gotten wrong. The promotional text is a marketing field only: Apple's search page says "promotional text doesn't affect your app's search ranking". And the description is where many guides tell you to repeat keywords, while Apple's product page guidance says the opposite and asks you not to add unnecessary keywords there in an attempt to improve search results.

App tags and natural language search

Search results can also show app tags. Apple's search page says these are generated with large language models from the metadata you provide in App Store Connect, and that people can use natural, everyday language in the search bar. The practical lesson is plain: describe what the app does accurately in every field, because a model is now reading the listing as a whole.

Developers now have some say over those tags. Apple's help page on managing app tags says tags appear in search results and on the product page, where people can tap one (Apple's example is Sleep) to see related apps and games. They are built from the metadata you provide, artificial intelligence and human curation, and applied by default from your US English metadata. Anyone with the Account Holder, Admin, App Manager or Marketing role can deselect tags that do not fit, under App Store Tags in the App Information section. Apple warns that deselecting every tag may affect discoverability, and for now tags are shown only in the United States.

Ranking factors the industry infers

Everything past Apple's list is inference. Some of it is well tested and some is guesswork repeated until it sounds like fact. Here is what gets claimed most often, who claims it and how far to trust it.

• Field weights. Apple publishes no weights. Its product page guidance does say the app name plays a critical role in how users discover the app, which is the only hint Apple gives about order. Treat any exact ranking of name over subtitle over keyword field as convention.
• Download velocity. Apple says downloads count. It does not say whether recent downloads count for more than old ones. "Velocity" is an industry word, so plan around steady downloads and do not expect a documented reward for a one day spike.
• Screenshot captions. In June 2025 the app intelligence firm Appfigures theorized that text in screenshot captions was affecting rank. TechCrunch reported that Apple had said at WWDC 25 it uses AI to extract information from the description, category and screenshots to assign tags, and concluded that developers should not need to add keywords to screenshots. At that date the tags were not yet shaping search on the public store. By October 2026 Apple's help pages describe tags in US search results and on product pages, built from your metadata with AI and human curation, so the advice holds: write captions for people and describe the app accurately everywhere.
• Cross localization. AppTweak says keywords from more than one language version are indexed in each country, based on tests it ran with made up keywords. The localization section below explains what Apple documents and what AppTweak adds.
• Update freshness. Apple does not document a ranking reward for recent updates. Google does list regular updates among the ways to improve visibility on Google Play.
• Retention and crashes. Apple's search page does not mention them. Google states that crash and freeze rates affect how discoverable an Android app is.

Every App Store listing field, its limit and what it is for

This is the reference part of the guide. The table gives each field's limit as Apple publishes it, whether search reads it and when you are allowed to change it. To check your own text against these limits as you type, use our free App Store character counter.

App name: 30 characters

The name carries the most visible weight on the page and Apple calls it critical to discovery. Apple's app information reference sets it at no fewer than two characters and no more than 30, and says you can change it when you create a new version. The pattern most small apps use is a brand word plus the plainest description of what the app does, such as "Driftwood: Sleep Sounds".

Section 2.3.7 of Apple's App Review Guidelines asks for a unique name and bars packing metadata with trademarked terms, popular app names, pricing information or other irrelevant phrases. A name that is only a list of keywords risks rejection and reads badly on a Home Screen.

Subtitle: 30 characters

The subtitle sits under the name everywhere on the App Store and is limited to 30 characters. Apple's product page guidance says to use it to explain the value of the app in more detail and to avoid generic boasts about being the best. The guidelines add that a subtitle must not reference other apps or make product claims you cannot verify.

Treat it as your second strongest search phrase and your first line of sales copy at once. Use different words from the name, since repeating a word spends characters and gains nothing. You can only change it with a new version, so each release is a chance to test a new one.

Keyword field: 100 characters or 100 bytes?

Apple's pages disagree on the unit. The App Store search page and the product page guidance both say keywords are limited to 100 characters in total. The version information reference in App Store Connect Help says you can provide up to 100 bytes of content. For plain English the two are the same number, because each unaccented letter and comma takes one byte.

The difference shows up in other languages. If the field is counted in bytes, an accented letter such as é takes two and a Japanese or Chinese character takes three, so a localized keyword list can fill up long before 100 visible characters. Apple's reference does not name the text encoding, so trust the counter in App Store Connect over any rule of thumb, and check localized lists there before you plan around them.

Apple's rules for the field are specific, and following them is the cheapest gain in ASO:

• Separate terms with commas and no spaces. A space after each comma wastes a character every time.
• Use a space only inside a phrase of two or more words.
• Do not repeat any word that is already in your app name, subtitle or category.
• Skip plurals of words you already included. Apple treats "climb" and "climbs" as duplicates.
• Skip "app", "game" and filler words such as "the" and "to".
• Leave out special characters such as # or @ unless they are part of your brand.
• Never include competing app names, trademarks or celebrity names. Apple calls improper keywords a common reason for rejection.
• Each keyword has to be longer than two characters, according to the version information reference.

The rule against repeating words tells you something useful about how search works. If a word in the name already counts, then the fields are read together, and the keyword field is best spent on single words that are not anywhere else on the listing.

Promotional text: 170 characters

Promotional text appears at the top of the description, holds up to 170 characters and can be changed at any time without submitting a new version. Apple suggests using it for news such as limited time sales or upcoming features. It is the one piece of the page you can edit on a Tuesday afternoon, so use it for whatever is current: a new feature, a seasonal use, an award. It does nothing for ranking.

Description: 4,000 characters

The description is limited to 4,000 characters of plain text. Apple's guidance is to open with a concise paragraph and follow with a short list of main features, and it points out that the first sentence matters most because it is all a visitor reads without tapping to see more. Apple also asks you to leave specific prices out, since they vary by region.

Write it for two readers: the person deciding whether to download, and the web search engine that will index the page. Clear sentences that say what the app does serve both. A block of repeated keywords serves neither and goes against Apple's advice.

What's New: 4,000 characters

What's New is the release note for each version after the first, limited to 4,000 characters. Guideline 2.3.12 requires it to clearly describe new features and product changes, and allows a generic line only for simple bug fixes, security updates and performance improvements. Apple's product page guidance adds a good habit: when you fix something users asked for, say so here.

Release notes are shown on the product page and in the Updates tab, so they reach people who already have the app. Our app release notes examples show formats that work, and the free app release notes generator turns a changelog into App Store wording.

Screenshots: up to 10 per device size

Apple's screenshot specifications require at least one screenshot and allow up to 10 per device size, in JPEG or PNG. If the interface is the same across sizes you can upload only the highest required resolution and App Store Connect scales it down. Guideline 2.3.3 says screenshots should show the app in use and not merely title art, a login page or a splash screen.

The exact pixel sizes for each iPhone and iPad are in our App Store screenshot sizes reference, and the free App Store screenshot size checker reads your files in the browser and tells you whether the store will accept them. How to design the set is covered in the conversion section below.

App previews: up to three videos of 30 seconds

Apple's app preview guidance allows up to three previews for each language, each up to 30 seconds, made from footage captured on the device. Previews autoplay on the product page and in search results with the sound muted, so Apple recommends on screen copy and a strong first few seconds. They must show only the app itself: no hands, no over the shoulder shots.

App icon

The icon appears in every search result, so it works as an ad for the listing. Apple's Human Interface Guidelines for app icons recommend a simple design built on one idea, text only when it is essential to the brand, and illustration over photos. Small details vanish at the size shown in search, so judge your icon at thumbnail size beside three competitors.

Categories

You assign a primary category and an optional secondary one, and Apple's search page says both are indexed by the search algorithm. Apple's category guidance says the primary category decides where the app appears when people browse or filter results. Pick it by the app's main purpose and by where similar apps already sit. An irrelevant category is grounds for rejection.

Because the category is indexed, Apple tells you not to spend keyword characters on its name. An app in Health and Fitness does not need "fitness" in the keyword field.

In app purchases on the product page

You can promote up to 20 in app purchases and subscriptions on the product page, and Apple's product page guidance says they can also appear in search results. Each has a display name and description of its own. Apple's pages give different limits for these (the product page guidance says 35 and 55 characters, the App Store Connect reference says 30 and 45), so go by the counter in App Store Connect.

In app events

Apple's in app events page describes them as timely events inside an app: a challenge, a competition, a live session, a premiere, a major update. Each gets an event card on your product page and in search results. An event can run for up to 31 days and be promoted for up to 14 days before it starts, with up to 10 published at a time. Our guide to App Store In App Events lists every limit, media size and badge, and which older figures changed.

The event name is limited to 30 characters, the short description to 50 and the long description to 120. In search, people who already downloaded the app see the event card beside it, while people who have not see your screenshots. That makes events a tool for bringing lapsed users back as much as for winning new ones. Apple lists daily tasks and plain price promotions as poor candidates.

Custom product pages

Custom product pages are extra versions of your page, each with its own URL and its own screenshots, previews and promotional text. Apple allows up to 70 per app. You send different audiences to different pages: runners to a page about running plans, cyclists to one about routes.

They matter for ASO because you can assign keywords from your approved version to a custom page, and that page then appears in search results for those keywords in place of the default one. Apple's custom product pages overview reports that developers see a 2.5 percentage point lift on average when they send people to a custom page, against an average conversion rate of 1.6% on default pages. Apple does not date that figure.

Keywords on custom pages are also quick to change. In Apple's WWDC25 session on App Store Connect, the team explains that when you only assign keywords to a custom product page, the change can go live without a new submission for review, and it recommends giving each custom page its own keywords instead of repeating the same ones across pages.

Product page optimization tests

Product page optimization is Apple's built in A/B test. You can test up to three alternate versions of the icon, screenshots and previews against the original, one test at a time, for up to 90 days. Results appear in App Analytics, and Apple recommends waiting for at least 90% confidence before applying a winner. A test cannot be changed once it starts.

App Store Creative Assets: Product Page Header and Search Results

The newest part of the listing arrived in October 2026. Apple's announcement of October 5 introduced creative assets: images and videos that sit in your product page header and in search results, next to your screenshots and app previews. Unlike screenshots, they do not have to show the app in use. Apple's asset best practices say they can highlight your brand, a seasonal offering or new content, and that they appear across the App Store on iOS 27 and iPadOS 27 and later, including the product page, search results and featuring.

For ASO the search result asset matters most, because it changes what a searcher sees before tapping. Apple's guidance says that if you do not add one, your in app events, app previews and screenshots appear in search as before. If you do, the asset takes that place, so it carries the same job the first three screenshots have carried until now: make the purpose of the app obvious at a glance.

The sizes come from Apple's creative assets specifications, which add that images cannot contain alpha channels or transparency and that videos loop continuously. Keep the focal point in the center, since Apple's templates define a safe area for the different devices and orientations.

The rules for creative assets

Guideline 2.3.14 was added to the App Review Guidelines on October 16, 2026 to cover them. Creative assets must relate to the app's content and functionality or its overall brand, follow the standard metadata rules and suit all audiences, and they must not include references to other apps meant to mislead, claims you cannot verify, or specific or discounted prices. Apple's best practices go further: no website addresses or copyright symbols, no logos of other platforms or stores, imagery that suits a 4+ age rating even when the app is rated higher, and no Apple recognitions such as Editors' Choice or App of the Day, because many of those already appear beside your app.

How to set them up in App Store Connect

1. Design one clear idea per asset. Apple advises a single focal point for the header and, for search, showing the interface or content so people can decide quickly. A universal asset can serve both placements, which keeps the brand consistent.
2. Upload it to the Asset Library or with a new version. Apple's help page on managing assets says creative assets can be submitted on their own, independent of an app version, without saying where you will use them, so they can be approved in advance. The Account Holder, Admin, App Manager or Marketing role is required.
3. Preview the page before you submit. App Store Connect now has a preview tool that shows the header, app name, description, screenshots and search result asset on iPhone and iPad, in both orientations and in Dark Mode.
4. Swap approved assets when the season changes. In Apple's WWDC26 session, the App Store Connect team shows an approved winter header replacing a summer one with no new submission.
5. Test it. Apple's best practices suggest product page optimization to compare header visuals, and custom product pages can carry their own header and search asset for the keywords assigned to them.

One caution about ranking. Apple's App Store search page lists text relevance and user behavior as the inputs to search, and creative assets are not among the text fields it names. Treat them as a conversion tool: a clearer search asset can win more taps from the same impressions, and those downloads are the behavior Apple does count. Do not put keywords into the artwork expecting search to read them.

App Store keyword research, step by step

App Store keywords are chosen, not guessed. The process below takes an afternoon and uses free sources first. You will end up with one phrase for the name, one for the subtitle and a list of single words for the keyword field.

Free methods first

1. Write a seed list. Put down what the app does, the problem it solves and the words your testers use for it. Ten to fifteen plain phrases is enough to start.
2. Collect App Store search suggestions. Apple's search page says hints appear as soon as someone types in the search bar. On an iPhone set to your target country, type each seed and write down every suggestion, then type the seed followed by each letter of the alphabet. These are phrases people really enter.
3. Read the listings that rank. Search each seed and note the names and subtitles of the top five results. You cannot see anyone's keyword field, but the words they put in their most valuable 60 characters tell you what they are betting on.
4. Mine reviews. Read 50 reviews of your app and of two competitors, and copy the nouns and verbs people use. Users describe a "sleep timer" where a developer writes "audio scheduling".
5. Check popularity in Apple Ads (the product formerly called Apple Search Ads). Apple's keyword guidance for advertisers describes a keyword suggestion tool that shows popularity based on App Store searches, and the Apple Ads glossary says search popularity is shown from 1 to 5, with 5 the most popular. You need an Apple Ads account to see it.
6. Look at who you would have to beat. For each promising phrase, open the top ten results and note how many ratings they have. If every result has hundreds of thousands, that phrase is not yours yet.

How to choose between keywords

Score each phrase on three things: relevance, popularity and competition. Relevance comes first, because a download from a search that does not match the app turns into an uninstall and a poor review. Apple's own advice on the search page is to weigh ranking well for less common terms against ranking lower for popular ones, and it warns that popular and seasonal terms are highly competitive for a newer or smaller app.

For a new app that usually means long tail phrases: "rain sounds for sleep" before "sleep", "split bills with roommates" before "finance". Fewer people search them, but you can reach the first few results, and that is where downloads happen. Move up to broader terms as your rating count grows.

Where each keyword goes

Put the strongest phrase in the name, the next in the subtitle and single leftover words in the keyword field, with no word used twice. Here is a worked example for a made up sleep sounds app:

Notice what the keyword field leaves out. "Sleep", "sounds", "noise", "rain" and "timer" are already in the name and subtitle, so they are not repeated. There are no plurals, no spaces after commas and no competitor names. Fourteen new words fit in the space that four repeated phrases would have used.

What paid app store optimization tools add

Free methods tell you which phrases exist. Paid ASO tools add three things: daily rank tracking for your keywords, estimates of search volume and difficulty for each phrase, and the keywords your competitors rank for. Each vendor calculates volume and difficulty its own way, so treat the scores as estimates and compare them inside one tool only.

The annual figures are simple division: $108 over 12 months is $9, and $949 over 12 is about $79. A solo developer with one app can do real work with the free methods plus Astro or a 100 keyword plan. The larger suites earn their price when you track many apps and countries, or when you need to know which keywords bring competitors their downloads.

App Store Rank Tracking: How to Check Where Your App Ranks

App Store Connect does not report your position for each keyword. Apple's metric definitions cover impressions, product page views, downloads and conversion, and you can filter them to the App Store search source, but none of them is a rank. That leaves three ways to see where you stand, from free to paid.

1. Check by hand. On an iPhone whose App Store is set to the country you care about, search each target phrase and write down where the app appears. Do it on the same weekday each week for the same 10 to 20 phrases, and keep the numbers next to your change log.
2. Read the Apple Ads impression share reports. Apple Ads' reporting definitions describe impression share for a search term, a rank from 1 to 5 or more than 5 against other advertisers, and search popularity from 1 to 5. That rank is about your ad, not your organic position, but it shows which phrases have demand and who else is competing for them.
3. Use a rank tracker. The paid tools in the table above check positions daily for each keyword and country, which is the only practical way to follow more than a few dozen phrases.

Category charts are the other ranking people ask about. In Apple's tech talk on App Store Connect Analytics, Apple explains that an app appearing in the top charts on the Apps tab, or featured on the Today tab, shows up as App Store browse traffic. So a chart climb is visible in your browse numbers, not your search numbers.

Read every rank against impressions. If a phrase moves from position 30 to position 8 and impressions from App Store search barely change, few people search that phrase, and the characters it uses in your keyword field may serve you better elsewhere. A move that lifts impressions and downloads together is the one worth keeping.

App Store Optimization Services: Agencies, Freelancers or Doing It Yourself

An ASO agency or freelancer does the work this guide describes: keyword research, the name, subtitle and keyword field, screenshot and icon tests, localization and a monthly report. You pay for their time and for the tools they run. The subscription prices in the table above set a floor, because an agency working in AppTweak or App Radar pays for that software before any hours are billed. For one app in one language, the free methods plus one entry level tool usually cover it. Outside help earns its fee when you run many apps or countries, or when nobody on the team has the hours.

What to ask before you hire

• Which App Store Connect role they need. Apple's role permissions reference describes App Manager as managing an app's pricing, App Store information and delivery, Marketing as managing marketing materials and promotional artwork, and Sales as analyzing downloads and other analytics. Give the narrowest role that fits the job. The Account Holder is the only user who can sign legal agreements with Apple, so that login stays with you.
• How they will measure success. The answer should be impressions, downloads and conversion from the App Store search source in App Analytics, compared over the same weeks before and after a change, using Apple's definitions.
• What changes each release. One change at a time, logged with its date, is the only way anyone can tell what worked.
• Who owns the research. The keyword lists, the competitor notes and the test results should stay with you when the contract ends.

Red flags in an ASO offer

• A guaranteed ranking or a top ten in a week. Apple says search depends on downloads, ratings and reviews as well as text, and nobody can promise those legitimately.
• Keyword installs. Some services sell installs from people paid to search a phrase and download the app, so it climbs for that phrase. Guideline 5.6.3 says manipulating charts, search, reviews or referrals is not permitted, and the business section warns that inflating rankings with paid or incentivized activity, or hiring a third party to do it, can get you expelled from the Apple Developer Program. Apple reported terminating 193,000 developer accounts over fraud concerns in 2025.
• Review packages. Bought or rewarded reviews break the same rules. What is allowed instead is in our guide on how to get App Store reviews.
• Competitor names in your keyword field. Apple lists competing app names and trademarks as improper keywords and a common reason for rejection, so an agency that suggests them is putting your next release at risk.

Ratings and reviews: what Apple says they do for ranking

This is one of the few ranking effects Apple states outright. The App Store search page says ratings and reviews appear on your product page and in search results and can influence how your app ranks in search. Apple's ratings and reviews page adds that the summary rating is specific to each territory, so a 4.8 in the United States says nothing about your rating in Germany.

Ratings also work on conversion. The star average and the count sit beside your name in every search result, and people compare them down the list. On some storefronts the product page shows a review summary as well: Apple says these are generated with large language models from user reviews, in the United States and in English in eight other countries including the United Kingdom, Canada and Australia.

Apple's rules for asking

• Use Apple's prompt. Guideline 5.6.1 says to use the provided API and that custom review prompts are disallowed.
• Expect three prompts a year at most. Apple's ratings page says you can prompt up to three times in a 365 day period, and names the SKStoreReviewController API. That API is now deprecated in favor of RequestReviewAction, which carries the same limit.
• Do not count on a second ask. Apple's documentation says that once a person has rated the app on a device, the prompt appears again only for a new version and after more than 365 days.
• Do not tie the prompt to a button. The system decides whether to show it, so a "Rate us" button that calls it may do nothing. For a button, link to the product page with the write review action.
• Never require a rating. Guideline 3.2.2 says apps must not force users to rate or review in order to access features.
• Do not manipulate. Guideline 5.6.3 bars manipulating reviews, charts and search.

When to ask

Apple's design guidance on ratings and reviews says to ask only after people have shown engagement, such as finishing a level or a significant task, and never on first launch or during onboarding. It also suggests leaving at least a week or two between requests. In practice: find the moment a user has just succeeded, wait until they have done it a few times and call the prompt there.

Replying and resetting

You can reply to every review in App Store Connect. The reviewer is notified and can update the review, which is the only way a one star rating ever becomes a four. Apple suggests answering low ratings and technical complaints first, and returning to old reviews when an update fixes what they describe.

You may also reset the summary rating when you release a new version. Apple recommends doing this sparingly, because a page with few ratings can put people off, and written reviews stay either way. The full method, with wording for replies, is in our guide on how to get App Store reviews.

Conversion: screenshots, icon and preview video

Ranking gets you seen. Conversion decides whether being seen was worth anything, and it feeds back into ranking because downloads are one of the behaviors Apple names. Start by knowing how Apple counts it: Apple's App Analytics overview defines conversion rate as total downloads divided by unique impressions, so 100 impressions and 50 downloads is 50%.

The first three screenshots in search results

Apple's product page guidance says that, depending on orientation, the first one to three screenshots appear in search results when no app preview is available. Portrait screenshots show three across, and a landscape one fills the row alone. For most searchers those images are the whole pitch.

Few visitors go further. A SplitMetrics study from March 2018, based on 1,800 A/B tests run on its own test pages, found that only 11% of visitors scrolled through all five portrait screenshots and 15% through all five landscape ones. The data is old and comes from a testing vendor, but the direction has not been seriously disputed: the first frames do nearly all the work.

• Give each of the first three screenshots one benefit, written as a short caption above the screen.
• Make the caption readable at search result size. If you cannot read it on your own phone at arm's length, it is too small.
• Show the app in use, as guideline 2.3.3 requires. A real screen with real looking data beats an abstract illustration.
• Lead with the outcome the user wants, then the feature that delivers it.
• Use screenshots four to ten for depth: secondary features, settings people ask about, a Dark Mode view, which Apple suggests including if the app supports it.
• Localize the captions when you localize the listing.

The icon

Icons are easy to test and can move conversion on their own. Apple's product page optimization page describes a test by Peak Brain Training that ran four icons for 44 days: the winning brain design converted 8% better than the original, while two other designs did worse. The lesson is less about brains than about testing. The team's first guess, a robot, lost.

Preview video

A preview is not automatically an improvement. The same Apple page reports a 12 day test by Simply Piano that put a preview video first on the page, expecting a lift, and found that the original page without the video performed better. Video helps when the value of the app is motion, such as a game or an editor. For a utility, three clear screenshots often explain it faster than 30 seconds of footage.

If you do add one, remember where it plays. Apple's app preview specifications say previews always appear before screenshots, and they autoplay muted, so the first three seconds replace your first screenshot in search results. Apple also advises against seasonal or dated references in a preview, since they age the video.

What a good conversion rate looks like

Benchmarks disagree because they divide by different things. Apple's custom product pages overview cites a 1.6% average conversion rate on default product pages. AppTweak's benchmark for 2025, published in May 2026 for the United States, reports an average of 3.8% of impressions turning into first time downloads on the App Store, and 8.56% when it divides by product page views instead.

Category matters more than any average. AppTweak's figures by product page views run from about 5% for trivia games to over 50% for food and drink apps. The most useful comparison is the one Apple gives you for free: peer group benchmarks in App Store Connect compare your conversion rate with similar apps, using Apple's own definition. For the published averages by category and why they disagree, see our App Store conversion rate benchmarks.

How to run a product page test

1. Write one hypothesis, such as "a caption about falling asleep faster will beat a caption about 40 sounds".
2. Change one element per treatment. Apple notes that the more you change, the harder it is to say what caused the result.
3. Send enough traffic. If you give 40% of visitors to a test with two treatments, each treatment gets 20% and the original keeps 60%.
4. Use Apple's estimate before you start. App Store Connect shows how long the test should take to reach 90% confidence. If the estimate is longer than 90 days, test fewer treatments or send more traffic.
5. Leave it alone. A test cannot be edited once it starts, and applying a treatment early ends it.
6. Apply the winner, then test the next element. Icon tests need the alternate icons inside the app binary, so plan them with a release.

Updates, freshness and seasonality

Does updating often improve your App Store ranking? Apple does not say so. What Apple does document is the other end: under App Store Improvements, an app that has not been updated in three years and has had no downloads, or extremely few, over a rolling 12 months gets a notice of possible removal. Regular updates keep you far from that line.

Updates help ASO in ways that are easy to see. Each version is the moment you can change the name, subtitle, keywords and description. Each one puts fresh release notes on the page and in the Updates tab. And each fix for a complaint is a reason to reply to the review that raised it.

How often should you change keywords? AppTweak's guidance on metadata updates, last updated in October 2025, says ASO practitioners generally find keyword rankings settle about four weeks after a new build goes live. That is an industry observation, not an Apple statement, but it gives a sensible rhythm: change, wait a month, read the numbers, change again.

Seasonality

Search demand moves with the calendar: fitness in January, travel in early summer, gifts in December. Apple's search page names seasonal terms such as "Christmas" or "summer" as drivers of a lot of traffic that are also highly competitive. A small app does better with a seasonal angle on its own niche than with the bare seasonal word.

• Promotional text: change it the day the season starts. No review is needed.
• Screenshots: test a seasonal first screenshot with product page optimization, which Apple lists as a use for the feature.
• Custom product pages: build a seasonal page and link your campaigns to it.
• In app events: a challenge or special event with a start and end date gets its own card in search.
• Keywords: seasonal words need a version update, so submit two to three weeks ahead and swap them out afterward.

Localization: 50 languages and 175 regions

Apple's localization page says the App Store is available in 175 regions and 50 languages. The App Store localizations reference lists the 50 languages and locales you can add metadata for, from Arabic to Vietnamese, including four kinds of English, two of French, two of Spanish and two of Portuguese.

Here is how the store picks what to show. Apple's help page on localizing app information explains that if you provide only your primary language, that metadata appears in every country. When you add French, it is shown to users whose language setting is French, and users can search for your app with the localized keywords in all countries where the App Store supports French.

A new language starts with the primary language's values for every property except the description and keywords, which you must write. That is the main reason localization is worth it for ASO: every language you add is a new keyword field for the people who search in that language.

Extra languages in one country

Most countries support more than one language. Apple's reference table shows the United States with English (US) as the default and nine more: Arabic, Chinese (Simplified), Chinese (Traditional), French, Korean, Portuguese (Brazil), Russian, Spanish (Mexico) and Vietnamese. A Spanish (Mexico) listing can therefore reach people in the United States whose language is set to Spanish, as well as people in Mexico.

The industry takes this one step further. AppTweak's article on cross localization says keywords in those additional languages are indexed for everyone in the country, so an app aimed at the United States could put more English keywords in its Spanish (Mexico) keyword field. AppTweak bases this on tests with made up keywords run for a 2022 book. Apple does not document it, and filling a Spanish listing with English words makes a worse page for Spanish speakers, so weigh the gain against that.

Which languages to add first

1. Open App Analytics and view product page views and downloads by territory. Apple's localization page recommends this to spot countries where people already find the app.
2. Pick the two or three countries with interest but weak conversion.
3. Localize the listing before the app: name if needed, subtitle, keywords, description and the captions on the first three screenshots.
4. Do keyword research again in that language. Translated keywords are rarely the phrases people search.
5. Have a native speaker read it. Apple's localization page says to have native speakers test any machine translation.
6. Compare conversion in that territory after four weeks.

Google Play ASO: the short version

Everything above is ASO for iOS and the App Store. This section is about Android and Google Play only, where the principles are the same and the fields are not. Google's store listing help page gives the limits: 30 characters for the app name, 80 for the short description and 4,000 for the full description.

The biggest difference is the description. With no keyword field, Google reads the listing text. Google's page on getting discovered in Play search says to use SEO best practices in the description while staying inside its spam policies, and the listing help page warns that repetitive or irrelevant keywords can get an app suspended. So write natural sentences that use your main phrases, with the most important text near the top.

Google is also more open about behavior. The same search page says apps are ranked on a combination of ratings, reviews, downloads and other factors, with the weights kept private, and lists regular updates and replying to users among the ways to improve visibility. Google's Android vitals page states that crash rate and freeze rate affect how discoverable an app is.

The rest follows the same playbook. Google's preview assets page allows up to 8 screenshots per device type and requires a feature graphic. Store listing experiments let you run one default graphics experiment or up to five localized ones at a time, and custom store listings go up to 50 pages.

App store optimization checklist

Use this ASO checklist before launch and again at each major update. It puts the app store optimization tips from this guide in working order, so every row has a section above that explains it. For everything else a launch needs, see the app launch checklist.

Common ASO mistakes

• Repeating words. A word in the name does not need to be in the subtitle or keyword field. Each repeat throws away characters.
• Spaces after commas in the keyword field. Ten keywords with spaces waste nine characters, enough for one more word.
• Competitor names as keywords. Apple lists competing app names and trademarks as improper keywords and a common reason for rejection.
• Stuffing the description. Apple asks you not to, does not list the description as a search field, and visitors read the first sentence only.
• Keywords in the promotional text. It has no effect on ranking.
• Chasing the biggest term. A new app aiming at "photo editor" ends up on page five. A narrower phrase can reach the top three.
• A splash screen or login page as the first screenshot. Guideline 2.3.3 says screenshots should show the app in use.
• Asking for a rating at first launch. Apple's design guidance says people may even be more likely to leave negative feedback when asked too early.
• Building a custom rating popup. Guideline 5.6.1 disallows it.
• Changing everything at once. If the name, screenshots and keywords all change in one release, you cannot tell which change moved the number.
• Writing "bug fixes" for a feature release. Guideline 2.3.12 requires real notes for significant changes, and you lose a free announcement.
• Watching rank and ignoring downloads. A higher position that brings no extra downloads is not progress.
• Translating the description and forgetting the keywords. The keyword field is the part of a localized listing that search reads.
• Prices in the name, subtitle or screenshots. Guideline 2.3.7 says those fields should not include prices.

How to measure ASO in App Store Connect

Everything you need to judge ASO is free in App Analytics. Apple's metric definitions spell out what each number means, and the wording matters because third party tools often use the same names for different sums.

Read it by source

A single conversion rate hides more than it shows. Apple's acquisition help page breaks downloads into sources: App Store search, App Store browse, app referrer and web referrer, plus campaigns. Filter conversion rate by App Store search to see the effect of keyword and screenshot changes. Note that Apple counts downloads from search results ads inside the search source, so pause to account for any Apple Ads spend.

App referrer and web referrer are where traffic from outside the store shows up. To tell one outside channel from another, use campaign links. Apple's campaign links page explains the format, and our free App Store campaign link generator writes a tracked link for each channel in one go.

A simple measuring routine

1. Before a change, write down four weeks of impressions, downloads and conversion rate for the App Store search source.
2. Ship one change with a version or a product page test.
3. Wait four weeks. Shorter windows mostly measure noise and weekday patterns.
4. Compare the same four numbers, in the same territory.
5. Check peer group benchmarks to see whether your conversion rate is typical for apps like yours.
6. Keep a one line log of every change and its date. In six months it will be the most useful ASO document you own.

How long does app store optimization take?

The first pass takes one to two days of work: an afternoon of keyword research, a day on screenshots and text. After that, plan on an hour or two a month. Results arrive on a slower clock than the work.

• Review: Apple's App Review page says it typically reviews at least 50% of submissions in under 24 hours and 90% in under 48 hours. Your new metadata goes live with the version.
• First movement: rankings for new keywords begin to shift in the days after release. Do not judge anything yet.
• A reliable read: about four weeks, the point at which AppTweak says practitioners find keyword rankings settle.
• Screenshot and icon tests: up to 90 days, the longest a product page optimization test can run. Low traffic apps often need most of that.
• Compounding: ratings build over months, and since Apple says ratings can influence ranking, an app's second year of ASO usually goes better than its first.

So the honest answer to how long ASO takes is a month for the first clear signal and a quarter for the first tested improvement. Anyone promising a top ranking in a week is selling something Apple's own description of search does not support.

Where social proof outside the App Store fits

ASO ends at the edge of the store, but the store counts what arrives from outside. A person who taps a link in a post and downloads the app appears in App Analytics as an app referrer or web referrer, and that download is as real as one from search. Apple names downloads among the behaviors search looks at. It does not say whether the source of a download changes its weight, so do not expect a documented ranking boost from social traffic. Expect downloads, which is the point.

The material for those posts is already on your listing. A new version, a 5 star review and a rating milestone are the same things ASO works to produce, and each is proof a stranger can check. The trouble for a small team is that turning each one into a designed post, every time, is the job nobody has hours for.

That is the job Spreadcast does. Spreadcast is not an ASO tool: it does not research keywords, track rankings or edit your listing. You paste your App Store link once, it checks the public listing every day, and it turns each update, 5 star review and rating milestone into a designed post, then publishes it to Instagram, TikTok, LinkedIn, Facebook, Threads, Pinterest, Bluesky, Mastodon, Telegram and Discord, plus YouTube for video posts. A post that quotes a reviewer's words waits for your approval, because Apple's ratings page allows reviews in marketing only with the reviewer's permission.

The free plan drafts every post for one app to download. Pro is $29 a month flat, with the design, a caption for each network and the publishing included. It works for apps on the App Store and makes still image posts. More on the method is in how to promote an app on social media, and you can paste your App Store link to see the first drafts.