App Store Optimization

What is app store optimisation?

App store optimisation (ASO) is the work of getting an app found and installed in Apple’s App Store and Google Play. When someone searches “expense tracker”, “yoga for beginners” or “train ticket booking”, the store decides which apps to show and in what order. Then the app’s icon, screenshots, ratings and description decide whether people tap install. ASO improves both: being found, and being chosen.

It works a little like SEO for apps, with important differences. The stores have their own search systems, fields with strict character limits, ratings that heavily influence downloads. And growing use of on-device intelligence to suggest apps without any search at all. And unlike a website, an app’s success doesn’t end at install. Stores more and more reward apps that people keep using.

Metadata: titles, subtitles and keywords

Store text fields are tiny, so every character earns its place.

(Store limits and rules change from time to time. Check current guidelines before submitting.)

Choosing keywords

The best keywords sit where three things overlap: people search them, they describe your app honestly, and you can realistically rank. “Finance” is too broad and too competitive for a new budgeting app; “monthly budget planner” or “expense tracker with UPI” may be realistic. Keyword research leans on the stores’ own search suggestions, ASO tools that estimate volume, and competitors’ listings.

Localisation multiplies reach. Both stores let you write listings for many languages and countries. An Indian app can carry Hindi, Tamil, Telugu and other listings, each with its own keywords and screenshots. Local-language listings often rank for searches the English one never shows up for.

Writing the Google Play description

Google Play reads the whole description for search. So it needs more care than on iOS. The first few lines matter most, since they appear before users tap “more”. They should say what the app does and for whom, plainly, with the main keywords in naturally. The rest can expand on features, use cases and benefits, using short paragraphs and lists. Repeating a keyword over and over doesn’t help and looks spammy. Covering related ideas does. Google’s policies also prohibit misleading claims, irrelevant keywords and fake testimonials in store listings.

Promotional text and what’s new

On iOS, promo text sits at the top of the description. You can change it without a new app release. That makes it perfect for timely news, like a festival offer. “What’s new” notes, on both stores, are read by more people than developers expect, especially after an update that fixed a widely reported bug. Plain, human release notes saying what got fixed help ratings recover after a bad update.

One more rule: on iOS, don’t repeat words across the name, subtitle and keyword field; Apple combines them. So repetition wastes precious characters. Separate keywords with commas and no spaces to fit more in.

Ratings and reviews

Ratings change both where you rank and how many people install. An app rated 4.6 usually converts far better than a similar one at 3.9. Recent reviews matter too. A run of negative reviews after a buggy update can hurt quickly.

• Use the official in-app review prompts (Apple’s SKStoreReviewController, Google’s In-App Review API).
• Ask at a happy moment: after completing a task, reaching a milestone or getting value.
• Never ask right after a crash, an error or on first launch.
• Respond to reviews, especially negative ones, with help and fixes.
• Use review themes to put first bug fixes and features.
• Never buy reviews or incentivise ratings. Both stores prohibit it.

Both stores cap how often you can show the review prompt, so timing matters. Ask right after a good moment, never mid-task. The surest way to better ratings is boring: fix what people complain about.

Designing an icon that earns taps

The icon appears everywhere: search results, top charts, the home screen, notifications. It has to be recognisable at a tiny size and stand out against competitors’ icons in the same search. Good icons usually use one simple shape, a bold colour and no tiny text. Drop a draft icon into a mock search results screen next to the top rivals and see if it stands out. Small tweaks, like a new background colour, can shift tap-through rates noticeably.

Responding to negative reviews

A good reply to a bad review names the problem and says what you’re doing about it. Offer a way to get help. Skip defensive or copy-paste lines. “Sorry about the login issue on Android 14. We fixed it in version 3.2.1, out today. If it still happens, please write to support@ with your phone model and we’ll sort it out.” Replies are public. So they speak to every future reader as much as to the reviewer. Users can also update their rating after a helpful reply, which often happens when the fix genuinely works.

Screenshots for different devices

Both stores ask for screenshots in specific sizes for phones and, if supported, tablets and other devices. Many apps upload phone screenshots only and leave tablet listings looking stretched or bare. If the app supports tablets, foldables or other form factors, dedicated screenshots for those devices help users on them understand the experience. And keep the listing looking professional wherever it appears.

Finally, keep screenshots honest. Showing features that don’t exist or exaggerating the interface leads to installs followed by disappointment, uninstalls and low ratings.

Testing store pages

Both stores now let developers test variations of their listings. App Store Connect lets you test different icons, screenshots and preview videos on real visitors, a feature Apple calls product page optimisation. It also offers custom product pages: other versions of your store page that you can link from specific ads. Google Play has store listing experiments for testing graphics and text.

Testing rules from the web apply here. Test one big idea at a time, let tests run long enough to mean something. And don’t trust a 2% “win” on small numbers. Custom pages matched to ad campaigns are especially useful: someone clicking an ad about “budget tracking for couples” should land on a page showing exactly that.

Apple and Google app ads

Paid app promotion works alongside ASO. Apple’s search ads programme (rebranded as Apple Ads in 2025) places apps at the top of App Store search results and in other App Store placements. Google promotes apps through App campaigns in Google Ads, which run across Search, Play, YouTube, Display and Discover.

The trap is paying for installs that never turn into active users. An install that opens once and is deleted is worth almost nothing. Measure cost per active user, per subscriber or per paying customer, with proper tracking. Then you’ll see which campaigns really grow the business. Our Apple search ads service manages this with retention in mind.

Three worked examples

These are composite scenarios from common patterns, not one client’s results. The apps are invented.

Example 1: a fitness app invisible in search

The situation. A home workout app has good reviews from existing users but gets few organic installs. Most growth comes from paid ads.

What we’d find. The app name is a brand word only. The subtitle is a slogan. The iOS keyword field repeats words already in the title. Screenshots show raw app screens with no captions. There’s only an English listing, though much of the audience is in India.

The fix. The name becomes “Brand: Home Workouts, No Gym”, the subtitle covers “yoga and weight loss plans”. And the keyword field is rebuilt with distinct, relevant terms. Screenshots get benefit captions (“20-minute workouts, no equipment”). Hindi and other Indian language listings are added with localised keywords and captions.

What changes. The app starts ranking for dozens of relevant searches, organic installs rise. And the cost of paid installs falls because store conversion improves.

Example 2: a fintech app with a ratings problem

The situation. A payments app’s rating dropped from 4.4 to 3.6 after a buggy update. Installs fell sharply.

What we’d find. Most recent one-star reviews mention the same login loop on certain Android devices. The team hasn’t replied to any of them. The in-app review prompt appears on first launch, before users have done anything.

The fix. Engineering prioritises the login bug. Support replies to affected reviewers once the fix ships, explaining what was fixed. The review prompt moves to after a successful payment. Release notes clearly describe the fix.

What changes. Many affected users update their reviews, new reviews skew positive. And the rating recovers over the following weeks, with installs following. See finance and fintech.

Example 3: an edtech app paying for the wrong installs

The situation. An exam-prep app pours money into App campaigns optimised for installs. Installs are cheap. Paid subscriptions barely move.

What we’d do. In-app events (trial start, first lesson completed, subscription) are tracked properly. Campaigns switch to chasing trial starts and subscriptions. Custom store pages match ads about specific exams. Onboarding is shortened so new users reach a first lesson quickly.

What changes. Installs cost more, subscribers cost a lot less, and revenue finally covers ad spend. See education and training.

What the examples share

None of these came from one clever trick. Each paired better text or creative with an honest look at the product itself. Stores reward apps people find useful and keep using. So the product and the listing have to improve together.

Every time, results came over weeks, not days. Store algorithms need time to notice changes in conversion, ratings and retention.

They also show why ASO people and product teams need to talk often.

What to do first

Start with the fields that count most and the creative people see first. Small tweaks come later.

Monthly reporting

Our monthly reports show keyword rankings in each store and country. They also show page views, the share who install, and installs by source. They end with what we’re changing next and why. All along, we care about installs that turn into active, paying or engaged users, not raw downloads.

Developer accounts, creative files and data stay with you throughout. We work inside your App Store Connect and Play Console, with access you control.

We keep a log of every text and creative change with its date. So results can be matched to what changed.

ASO myths worth dropping

Signs your ASO needs attention

• Most installs come from paid campaigns, with little organic growth.
• Your app doesn’t appear for obvious searches describing what it does.
• Store page conversion is falling or below your historical level.
• Ratings have slipped after recent updates.
• Competitors with weaker apps outrank you.
• You only have an English listing despite a multilingual audience.

If two or three of these sound familiar, a focused audit is probably worth it before spending more on ads.

Cost and timelines

Cost depends on how many stores and languages, how much creative we produce, and whether paid campaigns are included. The first audit and text overhaul is a project. Testing, localisation and review management are ongoing. Our pricing page explains how we scope.

Metadata changes need an app update on iOS (except promotional text) and take time to be reflected in rankings. Creative tests need enough traffic to reach a conclusion. Pulling a rating back up takes fixes, and time. Most apps improve noticeably within two or three months of steady work.

Games

Games follow ASO principles with extra emphasis on creative. Preview videos showing real gameplay, eye-catching icons and screenshots that capture the feel of the game drive most installs. Ratings and reviews matter hugely in games, as do regular updates and live events. Game ads wear out fast, so new creative needs testing all the time.

Business and enterprise apps

Apps for businesses, such as CRMs, field service tools and B2B ordering apps, often aren’t discovered through store search at all. Users install them because their company told them to. For these, ASO is less about search ranking and more about a clear, trustworthy listing, good ratings from real users and a smooth first experience. When they do compete in search, feature-specific keywords (“offline field service app”, “GST billing app for distributors”) tend to work better than broad category terms.

Health and finance listings

Health and finance apps face stricter store policies and, often, regulatory rules about claims. Listings should avoid promising outcomes (“lose 10 kg”, “guaranteed returns”), clearly explain data handling, and show licences or registrations where relevant. Stores may ask for extra paperwork, like licences or registrations, before approving these apps or their updates. Get compliance right in the listing from day one and you’ll dodge rejections at bad moments.

Whatever the category, the basics don’t change: clear text, honest creative, a stable app, and attention to what users say.

Category leaders also tend to refresh their listings often. So watching them closely shows what’s working right now.

Launching a new app

New apps start with no ratings, no download history and no ranking. The first weeks shape a lot of what follows. A planned launch gives the app the best chance.

1. Before launch

   Research and metadata

   Pick keywords the app can realistically rank for early on, and write text for both stores before launch.

2. Before launch

   Creative

   Icon and screenshots designed to say value instantly, ideally tested with a small audience first.

3. Before launch

   Stability

   Beta testing through TestFlight and Play’s testing tracks to catch crashes before public reviews do.

4. Launch weeks

   Early users and reviews

   Invite existing customers, email lists and communities. Time review prompts after early successes.

5. Weeks 1 to 8

   Paid support and iteration

   Targeted search ads on relevant terms, close monitoring of reviews and retention, quick fixes and metadata refinements.

A launch on a stable app, clear store pages and a few hundred engaged early users beats a big ad splash on a buggy one.

Learning from competitors

Competitor listings are a rich source of insight. Which keywords do they target in titles and subtitles? Which benefits lead their screenshots? What do their one-star reviews complain about? Where do they rank, and how has that changed? Regularly reviewing the top apps in your category reveals gaps (keywords nobody targets well, features users want but competitors lack) and helps avoid copying weaknesses. Competitor reviews in particular show exactly what users are frustrated by, which can shape both product decisions and how you position your app.

Measuring ASO in a privacy-first world

Measuring app marketing has become harder. Apple’s App Tracking Transparency, introduced in 2021, requires apps to ask permission before tracking users across other companies’ apps and websites. And many users decline. Apple’s SKAdNetwork (and its successor, AdAttributionKit) provides privacy-preserving attribution for ads with limited detail. Android has its own privacy changes under way.

In practice, combine them: store analytics for organic, a measurement partner for paid, and cohort data for quality. Our analytics team helps set this up.

Getting featured by the stores

Both Apple and Google have editorial teams that feature apps in the Today tab, collections and Play’s editorial sections. Being featured can bring a large burst of installs and lasting credibility. There’s no guaranteed route. Some things help, though. A stable, well-built app that uses new platform features. Good accessibility. Careful local versions. Fresh in-app events. Apple offers a form for developers to tell its editorial team about upcoming launches and updates, and Google offers similar channels for some developers. Submitting well ahead of launches gives editors time to consider them.

Who should be involved

ASO sits in the middle of marketing, product and engineering. Marketing owns keywords, creative and campaigns. Product owns onboarding, features and what gets promoted in store listings. Engineering owns stability, performance, app size and frameworks like App Intents. Support owns review responses. When they work in silos, the store page promises one thing and the app delivers another. A shared monthly look at rankings, conversion, ratings, crashes and reviews keeps everyone pulling the same way.

ASO for the Indian market

India is among the world’s biggest app markets, and it has its own quirks. Android rules here by a mile, so Google Play usually gets the most attention. Lots of people have little storage and patchy data. So size and speed matter a great deal. People search in English, Hindi and regional languages, often mixed together. Trust signals (ratings, security messaging, recognisable brand names) carry extra weight for finance and shopping apps. And seasonal events like festivals, exam calendars and cricket tournaments shape search patterns. An ASO plan built for the US rarely works unchanged in India.

Accessibility in apps

Accessible apps reach more people and tend to get better reviews. Supporting screen readers (VoiceOver and TalkBack), dynamic text sizes, sufficient contrast and clear labels on buttons helps users with disabilities and makes apps easier for everyone. Apple and Google both highlight accessibility in their design guidance. And apps that do it well are more likely to catch editorial attention.

The common thread is simple: an app people find easily, understand instantly and keep using will rank better over time than one built around tricks.

For most Indian apps, starting with Google Play, Hindi plus one or two regional languages. And a lean, fast app usually brings the biggest early gains.

What we look at in the first hour

When a new app lands on our desk, the first hour follows roughly the same path. Take a made-up Bengaluru expense-tracking app with about 40,000 installs and a 3.8 rating.

We open the Play Store page on a cheap Android phone, the way most Indian users will see it. The icon is a generic wallet that looks like six others in the results. The first screenshot shows a login screen. Nobody installs an app because of its login screen.

Next, the reviews, sorted by most recent. Three of the last ten mention the app logging people out after an update. That one bug is quietly dragging the rating down, and no amount of keyword work will fix it.

Then the text. The title is just the brand name. The short description talks about “seamless financial management”. Meanwhile, people actually search for “expense tracker”, “daily kharcha app” and “split bills with friends”, none of which appear anywhere.

Finally, Play Console. Store visitors are decent. But only about one in six installs, and a big chunk of installs uninstall within a day.

So the plan almost writes itself. Get the logout bug fixed first and say so in the release notes. Redesign the icon to stand out in search. Replace the screenshots with a short story: snap a bill, see where the money went, settle up with friends. Rewrite the title and short description around the words people really type, in English and Hindi. Then run a store listing experiment and give it a few weeks.

None of this is glamorous. It’s just looking at the page the way a stranger would. And fixing whatever puts them off, one thing at a time.