ASO keywords — research and analysis

What this module is for

Tracking rankings is useless if you track the wrong keywords. This page answers the question that comes first: what do people actually type into the App Store? It queries the official App Store autocomplete — the same one as the search bar on your iPhone — then helps you build and verify your keywords field. All of it is free: none of these sources is billed.

1. Search suggestions

• Type a term, pick the store's country and language, then Search.
• The App Store column lists queries actually suggested by store search, in decreasing popularity order: the first one is the most typed.
• The Google search column is a deliberate fallback — these are web suggestions, not Play Store ones (Google closed public access to its store suggestions). Useful to understand how the general public phrases things.
• Alphabetical expansion: tick this box to query “term + a”, “term + b”… through z. Slower (27 requests) but it surfaces many more variants.
• Click a suggestion to add it to your selection, or Add all.

2. My selection

• The selection is your working list: it becomes the target keywords for the listing analyzer below.
• Check queries iTunes search to see whether your app surfaces for that query. The rank shown is approximate — it answers “indexed / not indexed”, it does not replace the official rank tracking on the App Stores page.
• Copy puts the whole list on the clipboard (one query per line) to paste into the App Stores configuration.

3. Popularity, difficulty and opportunity

• Click Popularity & difficulty above your selection (10 keywords per batch).
• Popularity: is the keyword proposed by App Store autocomplete for its own prefixes, and at what position? The store ranks suggestions by decreasing popularity: being suggested first is worth 100, never being suggested is 0 — that keyword is barely typed.
• Difficulty: the strength of the real top 10 for that query — relevance of the apps found, weight of their reviews, ratings, keyword targeting in their titles, freshness, saturation. Click a row to see each factor with its weight.
• Opportunity: the two combined. Popular and lightly contested = worth targeting. Never suggested = don't spend your 100 characters on it.
• ⚠️ These are computed PivotSEO scores, not Apple Search Ads figures. Every factor is shown so you can judge for yourself — we prefer a verifiable score to a black box.

Measured reliability of these scores

• The difficulty score's weights are choices, not measurements. So they were tested against a real sample (20 queries actually suggested by the Canadian App Store versus 8 fabricated queries nobody types), and here are the results — published as they came.
• Separation power: 22.9 points. Average difficulty of 54.0 for genuinely typed queries versus 31.1 for fabricated ones. The score does tell the two populations apart. On popularity, separation is total: 93.6 versus 0.
• Correlation with reality: 0.45 (Spearman, score stripped of its “reviews” factor, against the real strength of the top 10). That is a moderate correlation: the score captures a real signal, but an imperfect one. The full score correlates at 0.70, but that figure is partly circular — review volume feeds the score, so it flatters itself.
• Dispersion: standard deviation of 16.7, from 14 to 88 across the sample. The score spreads enough to let you choose between two keywords.
• In other words: use it to rank your keywords against each other, not as absolute truth. An administrator can re-run this measurement at any time (POST /api/aso-tools/calibration) on their own seed terms.

4. Competitor keywords

• Paste up to 3 competitor apps (App Store ID or URL) and run the analysis — allow 20 to 40 seconds.
• The method: candidates are derived from their listing (title, salient description terms) and from store autocomplete, then each query is verified one by one to see who surfaces and at what rank.
• A keyword only appears in the table if the app was actually found in the results. No estimate, no extrapolation.
• Queries they all surface for are highlighted: that is the contested core of the market.
• Everything can be added to your selection in one click, then run through the difficulty check.

5. Keywords a listing uses

• Paste the app to dissect — yours or a competitor's — on either store or both.
• What is extracted: every 1-to-3-word term and phrase, weighted by the field it appears in. A title word counts six times a description word, a subtitle word four times. A “title” or “subtitle” tag shows next to the terms concerned.
• A major difference between the stores: Google Play indexes the description, so terms repeated there genuinely count. The App Store does not index it at all — with Apple, only the title, subtitle and hidden keywords field matter. The interface states this under each column.
• Cross-referencing both stores shows the shared terms, those specific to the App Store and those specific to Google Play: an app not targeting the same thing on both sides has usually forgotten to adapt one of the two listings.

Inferring the hidden iOS keywords field

• The App Store's 100-character field is public to no one: Apple exposes it in no API, and no tool can read it.
• It can, however, be inferred. Apple does not index the description: if an app surfaces in results for a term absent from its title and subtitle, that term almost certainly comes from its keywords field.
• The tool tests the 15 heaviest terms of the listing that are not in the title, keeps those the app surfaces for, and shows its approximate rank next to each.
• The other possible sources, rarer, are the publisher name, the category and in-app purchases. So this is a reasoned deduction, presented as such — never as a reading of the field.

Search volumes: two measures not to be confused

• App Store popularity (0-100): a PivotSEO score derived from store suggestions. Apple only publishes search volume through Apple Search Ads — nobody else has access without an ads account.
• Google web volume (searches/month): the real monthly search count on Google, via Google Ads. That is web search, not the Play Store — which publishes no volume, to anyone.
• These two columns neither add up nor compare. The first says what people type in the App Store, the second what they type in Google. A term strong in one can be weak in the other.

8. Query combiner

Stores index word combinations, not only exact phrases. Enter one group of words per column (one word per line): the combiner crosses the three groups to produce every possible query. This is how you cover a full semantic field with the 100 characters of the iOS keywords field.

9. Listing analyzer

• Fields pre-fill from the listing already analyzed on the App Stores page, when one exists.
• Lengths: title and subtitle at 30 characters, keywords field at 100 — the counter turns red when exceeded.
• Wasted words: Apple indexes each word only once. Repeating in the keywords field a word already in the title or subtitle burns characters for nothing — the tool lists them.
• Coverage: for each keyword in your selection, where it appears (title, subtitle, keywords field) and whether it is indexable — meaning present in at least one of those three fields.
• Density: the most repeated terms of the description, with their percentage.