See where your app ranks — and why.

No — they are estimates, and the panel labels them as such. Google Play publishes no official search volume. The popularity score is derived from signals like whether a keyword appears in Play suggestions, how high it appears, how many different searches surface it, and how many apps rank for it. The values are relative to each other, not absolute. Difficulty goes one step further: it is shown only once enough of the apps ranking for the keyword have been measured, and stays blank otherwise — a gap scored “easy” on missing data would send you after the hardest term.

Most ASO tools are built around English: they lowercase with ASCII rules, match whole words only and filter with an English stop-word list. Applied to an accented or suffixed language that folds letters together, treats a word and its inflected form as unrelated, and drops real search terms as noise. ersASO takes lowercasing, stemming, stop-words and the harvest alphabet from the app’s own language. Each language gets exactly what is known about it, and where nothing is known the filter stays empty rather than guessing.

No. The system works only with publicly available store data; a package name is enough to start. Nothing is read from your Play Console account, and no Console data is stored.

Not for now. The scope is deliberately limited to Google Play. Supporting both stores means a second data source and a second scoring model; we chose depth in one store instead.

Once a day, spread across a six-hour overnight window; a keyword you start tracking during the day is measured within the hour instead of waiting for the night. Keywords that have never ranked in the top 50 and show very low popularity are re-checked once a week instead, so a term that starts to move is still noticed. Measurement runs through a queue: requests stay under one fixed per-second ceiling shared by everyone and are spaced with variable delays. The slowness is deliberate — crawling a store quickly is the shortest path to being blocked.

No, and it does not keep a second copy either. The metadata page shows the live Play listing read-only and analyses coverage and density on that text. AI alternatives are generated in a workspace beside it and are never saved — you copy the one you like into Play Console yourself. A savable draft would create a second source of truth, and nobody could tell which one the store actually shows.

28 stores and 20 languages, chosen as a pair when you add an app — the same package in two markets is two apps, each with its own keyword pool, competitors and rankings. Lowercasing, the harvesting alphabet and the stopword list follow the app's language, not a server default, and accents are always preserved: “über” and “uber” are different keywords. Where no stopword list exists for a language the filter stays empty rather than inventing one and dropping real search terms.

Yours. Pick Claude, Gemini or ChatGPT for your account and add that provider's key; it is stored encrypted and used only for your own requests. All three receive the same instruction, which spells out Play's rules — no exaggerated claims, no competitor names, character limits respected. Without a key the rest of the panel works exactly as before and the AI surfaces simply report themselves unavailable. Switch AI off for the workspace and those surfaces — the drafting card on Metadata, the AI ideas tab in Discover, the “Why this ranking?” card on a keyword — are not drawn at all, so no dead buttons are left behind. Keyword discovery, scoring, rank tracking, the coverage matrix, the ASO score and the AI-free “Try your copy” card never needed a key.

Yes. Create an account and confirm your e-mail; a Free account opens as soon as the address is confirmed. Only during a wave of automated sign-ups may new accounts be reviewed by hand first, and then you get an e-mail when yours is ready. The Free plan needs no card and has no time limit; paid plans are set up by e-mail for now, so there is no checkout to go through and no trial countdown.