App Store keyword research, step by step

01Step 1 · Write a seed list from the job your app does

Start with the words a stranger would type to find what your app does, not with your brand. Three places give you candidates without guessing:

• Your own reviews. The words users choose to describe the app are often not the ones in your description.
• Competitors' titles and subtitles. Those two fields are public and indexed, so they show which terms other apps are targeting on purpose. Read them for concepts; Apple's guidelines forbid using other apps' names as keywords.
• Your category's vocabulary. Tasks, objects and outcomes ("split bill", "receipt", "expense"), not adjectives.

Keep single words and short phrases in the list. The keyword combiner turns a word list into two- and three-word candidates; it only generates combinations, it does not tell you which ones anyone searches.

02Step 2 · Check demand with App Store autocomplete

Type each candidate into the App Store search box and read the suggestions. A term that autocomplete suggests is a term people type; a term that never appears, even after its first letters, is one almost nobody searches. This is the free autocomplete demand signal.

It is a presence signal, not a volume number. We compared it with Apple Ads search popularity on 1,000 popular terms across five storefronts and the rank correlation was ρ = 0.30 — it separates searched from unsearched, it does not order the biggest terms. The details are in autocomplete demand vs search popularity, and why no store gives you a real count is covered in App Store keyword search volume.

03Step 3 · Read the top-10 wall to judge difficulty

Search the term and look at the ten apps that rank for it. Two numbers decide most of the difficulty:

1. The wall — the highest rating count among those ten. One app with hundreds of thousands of ratings holds the term even if the other nine are small, so use the maximum, not an average.
2. Title matches — how many of the ten carry the term in their app name. That shows competitors targeting it deliberately.

Our keyword difficulty score is built from exactly these two inputs — 70 points from the wall on a logarithmic scale, 30 from title matches — and the full arithmetic is printed in keyword difficulty, with the formula printed. With no title matches, a term reads medium once the strongest app reaches 10,000 ratings and hard at 100,000. The free keyword difficulty tool reads both numbers for one term in one storefront.

04Step 4 · Keep terms that are relevant, searched and winnable

Sort what survives into three groups. Targets now: autocomplete suggests them and the wall is low. Later: searched but held by large apps. Drop: not suggested anywhere, or not what your app does.

Relevance is not optional. Apple lists terms that are not relevant to the app, trademarked terms and competing app names among the keyword uses that lead to rejection, and a user who finds you through an unrelated term is unlikely to install.

05Step 5 · Place the words: title, subtitle, then 100 characters

On the App Store, text relevance comes from the app name, subtitle, keyword field and primary category. Put the strongest target in the name (30 characters) or subtitle (30 characters), where people also read it, and use the hidden keyword field for the rest:

• 100 characters, commas between terms, no spaces after commas.
• Don't repeat words already in your name, subtitle or category, and skip plurals of words you already have — Apple treats them as duplicates.
• No filler such as "the", "app" or "game".

The keyword field tool packs a list into 100 characters and flags duplicates with your title; the rules are explained in the 100-character keyword field, and every field's limit is in App Store and Google Play character limits.

Google Play has no keyword field. Play reads the title (30), short description (80) and full description (4,000), and Google's own help asks for SEO best practices in the description within its spam policy.

06Step 6 · Repeat per storefront, then measure

Each localization you add in App Store Connect has its own name, subtitle and keyword field, and some storefronts index a second locale next to their default — see ASO localization. Translating an English keyword list is not research: demand and the wall have to be read again in each country.

After the release, the only proof is rank. Check a single term with the free rank check; Storelift reads rank daily to the top 200 on iOS and about 30 on Google Play across 175 storefronts, and its keyword discovery surfaces terms you already rank for but never tracked.

Read next