App Store Optimization (ASO): The 2026 Guide

App Store Optimization is the work of getting your app found in the App Store and Google Play, and getting the people who find it to install it. It is the app store equivalent of SEO: you optimise your title, keywords, screenshots and reviews so the store’s algorithm ranks you higher and shoppers tap Install rather than scrolling past.

If you have shipped an app and nobody is downloading it, ASO is usually where the problem is. And in 2026 it changed more than it has in years, because Apple now generates some of your discovery terms for you.

What changed in 2026

Most ASO advice online predates these. If you learned ASO before iOS 27, four things are new.

App Store Tags: discovery terms you no longer write

Apple now attaches glanceable tags to apps, such as “Sleep” or “Budgeting”. They show up in search results and as tappable entities on the Search Landing Page and on product pages, and tapping one surfaces related apps.

The part that matters: you do not write them. Apple generates tags from the metadata you supply in App Store Connect, using a mix of AI and human curation, based on your en_US listing. What you get is a veto. In App Store Connect, go to App Information → App Store Tags → Edit and deselect any that do not fit. You need the Account Holder, Admin, App Manager or Marketing role.

Two caveats worth knowing. Tags are currently supported and displayed in the United States only. And deselecting all of them can reduce your discoverability, so prune rather than clear.

This makes your en_US metadata more consequential than it used to be. It is no longer just what shoppers read, it is the input Apple’s system reads to decide what your app is about.

Apple does not publish how many tags an app can carry, or the model behind the assignments. Anyone telling you they have reverse-engineered it is guessing.

Creative assets: a new slot in search results

From iOS 27 and iPadOS 27, a new asset type appears across the App Store, in search results and as a header on your product page.

You do not get to opt out of the layout. If you supply nothing, Apple falls back to your In-App Events, app previews and screenshots, so your app appears in the new slot using whatever it can find. Supplying a deliberate asset is the difference between choosing your first impression and having one chosen for you.

Apple’s content rules for these are stricter than most teams expect:

• No pricing, no discounts, no website URLs, no copyright symbols
• No logos or references to other platforms
• No Apple recognitions, so no “Editor’s Choice” or “App of the Day” badges
• No awards you have not actually received
• Everything must meet a 4+ age rating, even if your app is rated higher
• Assume audio is muted, because it is by default

Apple publishes templates for Figma, Photoshop, Pixelmator and Sketch.

Custom product pages: up to 70 versions of your listing

You can publish up to 70 additional versions of your product page, each with its own URL. Per localisation, you can vary the screenshots, promotional text and app previews.

Two things make this more than a vanity feature. You can assign keywords to a custom page so it surfaces in search for those terms instead of your default page, and you can attach a deep link so a tap opens a specific screen in your app, on iOS 18 and later.

Metadata on custom pages goes through App Review, but it does not require a new app version, so you can iterate on your listing without shipping a build.

Apple reports that developers see an average 2.5 percentage point conversion increase when linking to a custom page, which it frames as a 156% improvement over the 1.6% average on default pages. That is Apple’s own figure, measured by Apple, so treat it as a direction rather than a promise.

Product page optimization: A/B testing your listing

You can run a live A/B test on real App Store traffic, comparing up to three treatments against your current page. You can test app icons, screenshots and app preview videos, localised per language.

The constraints matter more than the feature:

• A test runs for a maximum of 90 days, or until you stop it
• You can only run one test at a time, and you cannot edit a test after it starts
• You choose what share of traffic sees a treatment, split evenly between them. Allocate 40% across two treatments and each gets 20%, with 60% staying on your original
• Apple recommends waiting for 90% confidence before applying a winner. Applying a treatment while the test is running ends the test
• Testing alternate icons requires every icon variant to ship inside your binary. Tests without icon changes need no new app version

One date to put in your calendar

From April 2027, iPhone Duo screenshots become required for apps built with the iOS 27.1 SDK or later. If you are shipping to the App Store next year, that is a new asset size you will need. Our iPhone screen sizes guide has the exact dimensions.

What is App Store Optimization?

ASO is the process of improving your app’s visibility in app store search and increasing the share of people who install it once they find it. It has two halves, and they are worth separating because they are fixed in different ways.

Discoverability: getting found

• Keyword search is how most people find new apps. They type something into the search field, and if your metadata matches, you appear.
• Featuring is when the store’s editorial team highlights your app. It drives real downloads, but you do not control it.
• Top charts surface apps by category, and reaching them needs volume you probably do not have yet.
• Tags, as of 2026, are a fourth route, and they are generated from your metadata.

Keywords are the lever you actually control. Start there.

Conversion: getting installed

Once someone lands on your listing, these decide whether they tap Install:

• Icon, screenshots and video. The visual first impression, and the thing most teams under-invest in.
• Title. Carries keyword weight and has to read well to a human.
• Ratings and reviews. The single strongest trust signal on the page.
• Description. Few people read it in full, but it matters for Google Play, which indexes it.

A note that undercuts everything else here: app quality drives ratings, and ratings drive both ranking and conversion. ASO improves how a good app performs. It does not rescue a bad one.

Why ASO matters

Both major stores hold roughly 2.3 to 2.4 million apps each as of mid-2026, according to app market trackers such as 42matters, though counts vary by source and methodology. Whatever the exact figure, the point stands: nobody stumbles across your app by accident.

There are two ways to get installs. Pay for them, at a cost per install that does not stop when you stop paying. Or earn them organically, which is slower, compounds, and is what ASO is.

Most teams need both. ASO is the one that still works when the ad budget runs out.

The basics: what to actually optimise

Keyword research

Research your users and competitors before you write a single piece of metadata. You are looking for the words people type when they want what your app does, which are often not the words you use internally.

Three ways to find them:

1. App store autocomplete. Type a keyword you know fits and read the suggestions. That is real search behaviour, straight from the store.
2. Competitor reviews. Read what users say about apps like yours. The phrases they repeat are the phrases they search.
3. ASO tools. Paid tools aggregate keyword data across sources and track rankings over time. Worth it once you are past the basics.

Pick keywords that describe your main feature, define your category, and have meaningful volume without impossible competition.

In 2026, this work does double duty, because your en_US metadata is also what Apple reads when it assigns your tags.

App name

Your name has to be memorable, descriptive, and carry a keyword or two without reading like keyword soup.

Mind the limits: 30 characters on the Apple App Store, 30 characters on Google Play. Both stores have tightened these over the years, so check the current limit in App Store Connect and Play Console before you commit.

Description

The two stores treat descriptions completely differently, and this is where most cross-platform advice goes wrong.

Google Play indexes your description. You get an 80-character short description and a 4,000-character full one, and both feed ranking. Use your keywords naturally, a handful of times.

Apple does not index your description. Keyword stuffing there does nothing for ranking and makes the copy worse. Apple gives you a separate 100-character keyword field, which is where that work belongs. Use the description to convince a human.

App icon

Keep it simple. One clear idea, legible at the smallest size the store renders it, which is smaller than it looks in your design tool.

Avoid forcing a long company name or a detailed logo into the square. A weather app’s icon should communicate weather at a glance, not spell out your brand.

Apple’s newer layered icon format uses Icon Composer, which lets the system render your icon with its own lighting and material treatments.

Screenshots and app previews

Screenshots do more conversion work than your description ever will. Most people decide from the first two.

Lead with your core value, show real interface rather than marketing abstraction, and keep any text short enough to read on a phone. For video, pick a poster frame that works standing still, and remember it autoplays muted and loops.

You can supply up to 10 screenshots on your product page and up to 3 in search results. We cover this properly in designing App Store screenshots that drive downloads.

Ratings and reviews

You cannot control these, but you can influence them. Ask at the right moment, which is after someone has had a good experience, not on first launch. Respond to negative reviews, because other shoppers read your replies.

Never buy reviews. Both stores detect it, and the penalty is worse than the problem.

A realistic sequence

If you are starting from nothing, do it in this order:

1. Fix your keywords and title first. Cheapest change, biggest discoverability effect.
2. Replace your first two screenshots. Biggest conversion effect.
3. Review your en_US metadata, now that it feeds Apple’s tags, and prune any tags that misrepresent you.
4. Add a creative asset, so your search listing is deliberate rather than inherited.
5. Once you have steady traffic, start a product page optimization test. Before that you will not reach significance.
6. Add custom product pages if you run paid campaigns or serve distinct audiences.

Where this stops working

ASO is a merchant’s job, and no platform can do it for you. Nobody can pick your keywords without knowing your customers, and no tool writes screenshots that convert.

What ASO will not fix: an app people do not want, an app that crashes, or an app that is slower than your website. Ratings reflect the product, and ratings drive ranking. Fixing the listing on a product with a two-star average is the wrong order of operations.

What we can take off your plate is the production side. Twinr turns your existing website into native iOS and Android apps, and we handle App Store submission ourselves, which means the asset specifications above, including the April 2027 iPhone Duo requirement, are our problem rather than yours. If Apple pushes back on a submission, that is ours to fix too.

You still write the keywords. You still choose the screenshots. But you do not have to learn Xcode to ship, and you do not have to maintain two codebases to stay in both stores.