ASO keyword research: how to find keywords from the stores' own signals

ASO keyword research means finding the words people type in the App Store and Google Play, then keeping the ones your app can win. The stores hand you the raw material themselves: the suggestions that appear as people type, Apple Ads’ search popularity (1 to 5 in the Apple Ads interface, 1 to 100 in its API), the search terms Play Console reports for your own listing, and the results of each search, country by country. Neither store publishes how many people search a word. So list what people type, check who already answers each search and where your app stands, give each word one place in your name, subtitle or keyword field, and measure the result.

Every store fact below comes from Apple’s or Google’s own pages, and every measurement from searches made for this guide on 25 and 26 October 2026.

What signals do the App Store and Google Play give you?

Some signals are public. Others show only your own app, inside the stores’ consoles.

Signal	Where	What it tells you	What it does not
Search suggestions	App Store and Google Play	What people start to type, in one country	How many people type it
Discover section	App Store, Search tab	Popular or trending searches, before anyone types	How they are picked
Search popularity	Apple Ads	A relative score per search term and country	A number of searches
Search Terms tab	Apple Ads, when you run ads	The searches that showed your ad	Searches where you had no ad
Search terms	Play Console	What people searched before visiting your listing	Terms that never brought you a visitor
Source type	App Store Connect Analytics	How many views and downloads came from App Store search	Which words people typed
The results	App Store and Google Play	Who already answers a search, and where you stand	Why the store ordered them so

The two consoles differ here. Play Console has a search term dimension, “the term that the user searched for before navigating to your Store Listing”, and Google recommends using it:

Review Google Play search terms driving the most visits to see what users are most interested in.

Terms with too few visitors are grouped under “Other”. App Store Connect has no equivalent: its Analytics count views and downloads from App Store search, search ads included, and its filters and dimensions include no search term. On the App Store, the words come from the store itself and from Apple Ads.

What do search suggestions tell you?

Both stores suggest searches while people type. Apple describes them on its search page:

Search hints appear once someone inputs text into the search bar and can help inspire people before they search or select a result.

On Google Play, Play Console counts the visitors “who clicked on search autocomplete suggestions”. And before anyone types, the App Store’s Search tab shows “popular or trending queries” in its Discover section, which Apple says may be personalized in some cases. Neither store says how suggestions are chosen or ordered. What they show is plain: phrases people in that country type.

How to read them:

1. Type slowly. Start with the word your app is about, then add one letter at a time. Note which phrases come up early. That is an observation, not a rule either store publishes.
2. Search country by country. Suggestions belong to a storefront: repeat in each country you sell in, in its language.
3. Skip the brand names. They are people looking for one app, and Apple does not allow other apps’ names in your keywords.
4. Keep the modifiers. “Free”, “widget” or “with friends” say what people expect.
5. Write down the exact form. Singular or plural, with or without accents. The section on plurals shows why.

A suggestion tells you that people type a phrase, not how many do. For that, the App Store has a score.

What is Apple Ads search popularity, and what scale does it use?

Apple Ads, formerly Apple Search Ads, gives search terms a popularity. Its glossary defines it:

A relative indicator of an Apple Ads keyword’s popularity. The search popularity indicator is designed to help you learn which keywords in search results campaigns might be most important for promoting your app. Search popularity is displayed as numbers from 1 to 5, with 5 being the most popular.

Apple shows it in three places:

• The Recommendations page of an Apple Ads account, next to each recommended keyword: “the popularity of a keyword, based on App Store searches”, from 1 to 5 (Review recommendations).
• The keyword suggestion tool of your account, which “displays keyword popularity based on App Store searches” (Keywords best practices).
• The Apple Ads Platform API. Its search term popularity query returns, per country, genre and week (Sunday to Saturday) or month, a popularity from 1 to 100 across all genres of the country, a popularity from 1 to 100 within the genre, the term’s rank by volume within the genre, and a tier from 1 to 5. It returns “the most popular search terms for a given App Store genre and country”, so a rare term is simply missing. Its keyword suggestions carry their own popularity, “on a 0–100 scale”.

One idea, three scales. What Apple leaves out matters as much:

• It is not a volume. Apple does not say how many searches a 5, or a 72, stands for, or whether the steps are even.
• It belongs to a country. A term can be popular in one storefront and missing in another.
• It covers the App Store only. Google publishes no equivalent; Play Console’s search terms cover your own listing.
• It does not run from 5 to 100. ASO guides often say it does. Apple’s pages describe 1 to 5 and 1 to 100, and we found none that starts at 5.

Use it to rank candidates against each other, in one country and one week, not to forecast installs.

Popular or reachable: how do you weigh relevance and competition?

Apple describes the trade-off on its search page:

Consider the trade-off between ranking well for less common terms versus ranking lower for popular terms. Popular, functional terms, such as “jobs” or “social” and trending or seasonal terms, such as “Christmas” or “summer” may drive a lot of traffic, but are highly competitive – especially if you are a newer or smaller app. Less common terms drive lower traffic, but are less competitive.

Relevance comes first. Apple asks you to “choose keywords based on words you think people may use to find an app like yours”, and lists terms not relevant to the app among the causes of rejection. Google warns that repetitive or irrelevant keywords in the name or description can get an app suspended.

Competition you can read in the store itself. Search the phrase in the country, signed out, and look at the first 10 results:

1. Do they do what your app does? If not, the store reads the phrase differently from you.
2. How many have the phrase in their name? That is the most visible match there is.
3. How established are they? Apple writes that ratings and reviews “can influence how your app ranks in App Store search”, and that the summary rating “is specific to each territory on the App Store”. Their number of ratings in that country is a rough sign.
4. Where is your app? 5th, 150th, or absent from the 250 results the App Store returns. The guide to measuring keyword rankings shows how to get a reading you can trust.

Where should each keyword go?

On the App Store, three fields are matched as text: the name (30 characters), the subtitle (30) and the keyword field (100). Apple asks you not to repeat, in the keyword field, a word already in your name, subtitle or category. Each word gets one place.

• The phrase that matters most goes where people read it. The name and subtitle appear in search results; the keyword field never does. Apple does not say that a word counts more in the name, so choose the words that make people tap.
• The rest goes in the keyword field: one form of each word, commas, no spaces. The keyword field guide goes through it line by line, and the name and subtitle guide covers the two visible fields.
• Split phrases are something to measure. Apple does not document whether words from different fields combine, such as “habit” in the name and “tracker” in the keyword field, for a search of “habit tracker”.

Google Play has no keyword field. Google writes that once it knows what someone wants, “metadata (for example, title, description, category) and other signals are used to determine which apps best address the user’s query”. The title holds 30 characters, the short description 80 and the full description 4,000. Write them for people, with the words they type; the Google Play search guide covers the rest.

Singular or plural, and which spelling?

Apple treats plurals of words you already included as duplicates in the keyword field. That rule is about what you write. It does not mean the two searches return the same list.

Here the suggestions settle it: people are offered the singular, and the app ranks for the singular. When people type both forms, measure both.

A misspelling can go two ways. The store can correct it and say so: on 25 October, “morning rutine” came back as a search for “morning routine”. Or it can say nothing and still answer for the word you meant: the same evening, “habbit tracker” came back with no correction declared, yet with habit trackers at the top, in an order of its own, and the app was 8th. The spelling correction guide covers both. On Google Play, Google says it may identify other words, “such as synonyms”, and advises against “subtle misspellings of general terms” in a title, because people correct them as they type.

How do you find keywords in other languages?

Each localization of your listing has its own name, subtitle and keyword field, and Apple searches a language in every country where the App Store supports it:

Users can search for your app using localized keywords in all countries or regions where the App Store supports French.

A translation is a guess. Run the research again in each language, with that country’s suggestions and results.

The localizations guide has Apple’s table of the languages searched in each country.

A worked example, measured

The whole method, on four candidate phrases for the same app in the United States:

Phrase	Suggested	First 10 with the phrase in their name	First 10 with over 10,000 US ratings	The app’s rank
habit tracker	1st after “habit t”	7	7	5
daily routine	5th after “daily r”	6	4	2
self care	2nd after “self c”	4	5	158
to do list	1st after “to do”	5	9	Not in the first 250

Row by row:

• “habit tracker” and “daily routine” already work. The app is 5th and 2nd, among apps that name the phrase. Keep these words, and measure them after every change.
• “self care” is a candidate. People are offered it early, 6 of its first 10 apps are listed in Health & Fitness, and the app sits around 160th: the store links the app to the phrase, weakly. Test it in one field, with a before and an after; the guide to judging a listing change shows how to read the result.
• “to do list” is not for this app. It is the most suggested of the four, 9 of its first 10 apps have over 10,000 ratings in the United States, and the app is absent. Spend the characters elsewhere.

With an Apple Ads account, add one column: each phrase’s search popularity that week. It sorts the candidates by demand; the rest of the table says whether you can win them.

Common mistakes

• Reading popularity as a number of searches. Apple calls it relative and publishes no volume.
• Copying one country’s list to another. Suggestions, results and ranks belong to a storefront.
• Taking brand names from the suggestions. Other apps’ names are not allowed in your keywords.
• Checking a rank once, on your own phone. One search is one sample, and your account may shape what you see.
• Translating instead of researching. People in Brazil, Germany or Japan type their own words.

Sources

• Apple, App Store search: search hints, choosing keywords, popular against less common terms, ratings.
• Apple, App Store, Learn more about search results and ranking: the Discover section.
• Apple, Apple Ads Glossary, Review recommendations and Keywords best practices: search popularity, the Search Terms tab, the keyword suggestion tool.
• Apple, Apple Ads Platform API, Insights endpoints, Search Term Popularity Query and Suggestions endpoints: the 1–100, 1–5 and 0–100 scales.
• Apple, App Store Connect Analytics, Acquisition and Filters and dimensions: the App Store search source.
• Apple, App Store Connect Help, Localize app information: localized keywords.
• Apple, Creating your product page: name and subtitle limits, ratings per territory.
• Google, Play Console Help, Understand and grow your app’s user base: search terms, autocomplete suggestions.
• Google, Play Console Help, App Discovery and Ranking, Get discovered on Google Play search and Create and set up your app: metadata, synonyms, misspellings, field limits.
• Measurements, Paris time: suggestions in the United States, one request per prefix, 00:07–00:08 on 26 October 2026; the App Store app’s iPhone search, one search per phrase, 23:22–23:23 on 25 October and 00:08–00:11 on 26 October; ranks, three searches per keyword, 20:39–20:52 on 25 October; rating counts from Apple’s lookup, 26 October.

Checked on 26 October 2026.