App Store Optimization: Improving Your Listing and Measuring Results
App store optimization (ASO) is the work of making your app easier to find in the App Store and Google Play, and more convincing once people see it. Being found depends on the words in your listing and on how users behave – Apple names downloads, ratings and reviews among its search factors. Being chosen depends on the icon, the first screenshots, the first line of text and your ratings. Both stores offer free reports for each step and tools to test changes. Nobody can promise a position in store search, but you can make the listing accurate and clear, measure it, and improve it one change at a time.
How store search works – according to the stores
Apple says search results are based on text relevance – matches for your app's title, subtitle, keywords and primary category – and on user behaviour, including downloads, ratings and reviews. It doesn't publish the weighting. Google doesn't publish a formula for Google Play search, but its listing guidance says what it expects: a unique title of 30 characters or less, a short description of up to 80 characters that sums up the main benefits, and a full description of up to 4,000 characters that doesn't repeat the short one and avoids "excessive length, detail, improper formatting, and repetition of words".
Keyword stuffing breaks both stores' rules. Apple's App Review Guidelines forbid packing metadata with trademarked terms, popular app names or irrelevant phrases "just to game the system"; Google's listing graphics and text requirements say not to add unnecessary keywords to improve search.
Choose the words people actually search for
Start from the task, not your brand. People who don't know you yet search for what they want to do: "track fuel consumption", "book a table", "invoice on the go". List such phrases for each language you support – the same approach as keyword research for a website, with much less space.
On the App Store, words go into the name (up to 30 characters), the subtitle (up to 30) and the keywords field (100 characters). Apple's rules for the keywords field:
- separate terms with commas, without spaces;
- don't repeat words already in the app name, subtitle or category, and don't add plurals of words you already use;
- skip generic words such as "app", filler words, competitors' app names and trademarks you have no right to use.
Apple also notes the trade-off: popular terms bring traffic but are highly competitive, especially for newer or smaller apps; less common terms bring less traffic with less competition. Promotional text doesn't affect search ranking, so don't use it for keywords. Google Play has no keywords field – the title and descriptions carry the words, and they should read naturally.
Convince in the first seconds
- Icon. Simple and legible at small sizes. Google doesn't allow badges or text in the icon that suggest ranking, price or Play categories.
- First screenshots. Apple says that, depending on orientation, the first one to three screenshots appear in search results when there is no preview video. Show the main task, not a login or splash screen. The App Store allows up to 10 screenshots; Google Play requires at least two, allows up to 8 per device type and asks that text taglines take up no more than 20% of the image.
- Video. Apple's app previews last up to 30 seconds and autoplay muted, so the opening must work without sound. On Google Play only the first 30 seconds of the YouTube video autoplay.
- The first line of the description. Apple calls the first sentence the most important, because it's what people read without tapping "more".
- Each language. Google asks for separate screenshots and videos per language if they contain text.
Avoid claims you can't back up: Google's rules forbid words such as "Best", "#1", "Top", "New" or "Discount", and calls to action such as "download now".
Ratings and reviews
Apple states that ratings and reviews appear in search results and can influence how an app ranks. Both stores let you ask for a rating inside the app, with limits:
- On iPhone, the system rating prompt can appear up to three times in 365 days, and the system decides whether to show it. Apple advises asking after people complete an action or task, without interrupting them.
- On Google Play, the in-app review dialog has an undisclosed quota. Google says to ask only after users have experienced enough of the app, not to ask "Do you like the app?" first, and not to open the dialog from a button, because it may not appear.
Answer reviews. If you can't answer all of them, Apple suggests starting with the lowest ratings and reports of technical problems, and mentioning fixes in release notes.
Test changes instead of guessing
| App Store: product page optimization | Google Play: store listing experiments | |
|---|---|---|
| What you can test | Icon, screenshots, app previews | Icon, feature graphic, screenshots; descriptions in localized experiments |
| Variants | Up to 3 against the original | Up to 2 against the current listing |
| Duration | Up to 90 days | Stops automatically after 6 months |
| Result | Conversion rate with a confidence level; Apple recommends applying a variant only at 90% confidence or more | Install clicks with a confidence interval; outcomes include a better variant, "Draw" or "More data needed" |
Tests need traffic. With a few dozen visitors a week, a test can run for months without a clear result; fixing obvious weaknesses first – unclear screenshots, a vague first line – is then a better use of time. When different campaigns bring different audiences, Apple's custom product pages (up to 70) and Google's custom store listings can show each audience the screenshots that match the ad or email they came from.
Measure the whole path, not just downloads
| Question | App Store Connect | Google Play Console |
|---|---|---|
| Do people see the app? | Impressions – times the app was viewed for more than one second on the Today, Games, Apps and Search tabs | Store listing visitors – users who visited the listing and didn't already have the app on any of their devices |
| Does the listing convince? | Conversion rate – downloads and pre-orders divided by unique device impressions | Click-through rate – the share of visitors who tapped Install, Open or Pre-register; it replaced the old conversion rate |
| Where do visitors come from? | Which sources drive the most traffic to the product page | Google Play search, Google Play explore, ads and referrals; the search terms used; UTM source and campaign for tagged links |
Since July 2026, Google Play's listing reports focus on clicks rather than completed installs, so take care when comparing with older data. Visits from Google Ads count under "ads and referrals", and Play Console doesn't report Google Ads impressions or conversions. Apple's metric definitions are worth reading once, so that everyone means the same thing by "conversion".
- Log the date of every change – icon, screenshots, release, campaign. Without a log, you can't tell what moved the numbers.
- Change one thing at a time where you can, and compare equal periods; seasons and campaigns move numbers too.
- Look beyond installs. A listing that promises more than the app delivers raises downloads and deletions together. App Store Connect reports deletions and usage from users who agreed to share data with developers – treat them as estimates.
What ASO can't do
ASO improves how an app is found and presented; it can't fix an app people don't keep. If users install and leave, the work is in the app: the first minutes (onboarding to the first useful action), reasons to return (useful push notifications) and stability. And the store isn't the only channel: customers who already know you arrive from your website, emails and premises, and a tagged link to the app from those places is often worth more than another keyword. If you're still deciding whether you need an app at all, read app or better mobile website.
A simple monthly routine
- Compare visibility, conversion or click-through rate and the main sources with the previous month.
- Check the search terms Google Play reports against the words in your listing.
- Read and answer new reviews; note repeated problems for the next release.
- Improve one thing – a screenshot, the first line, a keyword – and log the date. Run a store test only if there's enough traffic.
If your listing hasn't been revisited since launch, start with a review of its texts, screenshots and console data. We can help as part of our mobile app development work – get in touch.