App Store keyword research: a practical 2026 guide

Most advice says to find high volume, low competition keywords. Neither store publishes volume, so that instruction cannot actually be followed. Here is what to do instead: what gets indexed, how to build a term list, and how to tell if a keyword is winnable before you spend a month chasing it.

What actually gets indexed

On the App Store, three fields feed search: the app name (30 characters), the subtitle (30 characters) and a keyword field (100 characters) that users never see. Terms in it are comma separated with no spaces, and Apple’s own guidance says not to repeat a word already used in your name, subtitle or category, since a repeated word gains nothing.

Google Play has no hidden field at all. The title, short description and full description are all indexed as written, which means a keyword stuffed into the Play description is a keyword your actual users have to read past. The two stores reward different writing for the same reason: one hides the keyword list, the other does not have one.

One detail that costs people characters

Apple counts the keyword field in bytes rather than in characters. Plain English letters cost one byte each, so for most listings the two are the same number. Accented and non-Latin characters cost more, which matters the moment you localise into German, Greek, Japanese or Arabic.

How to build the term list

Why nobody can sell you App Store search volume

Neither store publishes search volume for any term. Apple exposes it only through the Search Ads API, to advertisers, for their own account, and Google publishes nothing comparable for Play. Any tool showing a precise number next to a keyword is showing you a model trained on indirect signals, not a count. That can still be useful, but only if you know which one you are looking at.

The question worth asking instead: is this term reachable

A more answerable question than "how many people search this" is "who currently holds the top results, and could my app realistically outrank them." Open the top five results for a candidate term and look at their rating counts and how established they are.

None of that requires a volume estimate, and all of it is visible in the store itself. It is also the comparison that actually decides whether a term is worth a month of your attention.

After you pick the terms

Rankings move on their own schedule. Ship the metadata change, note the date, and leave it alone long enough to produce a readable result. Checking once, two days later, tells you almost nothing either way, and the temptation to rewrite again immediately is how listings end up churning without ever learning anything.

Blog · Free ASO tools · Products · AsoTheory