free private beta · macOS · local-first
App Store keywords, measured.
Feed it a plain-text description of your app. It returns ranked keywords plus a title, subtitle and keyword field — and every number expands to the raw Apple response it was computed from.
Free during beta. No card. Runs on your Mac — your data stays local.
new apps shipped to the App Store in 2025.
source: Sensor Towergrowth in app submissions, quarter over quarter.
source: Sensor Towerof installs start with an App Store search.
source: AppleThe store is flooding. Search is still how users find you. Metadata is the only distribution channel you fully control.
One pipeline, five steps —
each signed by who does it.
The AI's role is deliberately narrow: it reads your brief and judges relevance. Everything countable is counted by code, from live Apple data.
Brief in
youA plain-text description of your app: what it does, who it's for, what it is not. Two minutes. That's the only input.
Context extraction
AI · judged by youJobs-to-be-done, user vocabulary, and anti-semantics: what your app must never rank for. You review and correct before anything runs — the pipeline's only checkpoint.
Crawl of Apple's suggest graph
codeCompletions, continuations, alphabet-soup expansion. Every candidate is a phrase Apple itself suggests to users — none are generated by a model.
Measurement
code + AI judgeP from probing live autocomplete, D from the live top-10. R — the only AI-scored value — follows a fixed 0–3 rubric with a written reason, full prompt and response in the log. The AI never produces a number for P or D.
Set-cover layout
codeWords distributed across title, subtitle and the 100-char keyword field with zero repeats. The second indexed localization (es-MX for a US listing) is filled automatically.
Ship-ready metadata
Title · subtitle · 100-char keyword field · second localization · full research trail.
Audit every number.
The industry sells "popularity scores" without saying where they come from. Below is every formula this tool ships with. Each parameter is measured, logged, and traceable to the request that produced it.
PDemand, from autocomplete probing
The phrase is typed against Apple's live autocomplete character by character. Recorded: the prefix depth where it first appears (earlier = stronger demand) and its rank in the list. It's Apple's own demand signal, read directly.
DDifficulty, from the live top-10
For every keyword the actual search results are pulled, and each of the ten apps you'd fight is scored for strength — weighted by its position on the page. A top-10 full of dead apps is an open door, and you will see it.
+ 0.15·freshness + 0.25·exact-match)
volume — review count; quality — rating; freshness — time since last update; exact-match — the keyword verbatim in a competitor's title.
RRelevance — the only AI-scored value
A language model scores each keyword against your confirmed brief, on this rubric and no other. A written reason is mandatory for every score; the full prompt and raw response are kept in the run log.
ΣOne score per keyword
Demand and ease are combined as powers, not a sum, on purpose: a keyword nobody searches is worth zero no matter how easy it is, and a keyword you can't crack is worth zero no matter how popular. Relevance scales the result linearly.
If a number can't be traced to a raw Apple response, it doesn't ship.
We measured what others assume.
The engine's rules come from live App Store data, not habit. Three findings that shaped it — each one checkable from your own keyboard.
Share of keyword candidates that turn out to be real user queries: phrases invented by an LLM versus phrases harvested from Apple's suggest graph. Measured on 828 probed keywords across two niches. This is why the crawler supplies candidates and the AI only judges them.
probe any LLM-written keyword list against Apple autocomplete and count the hitsApple seeds app names into autocomplete — including apps with zero ratings. So the name of a dead app looks exactly like a high-demand keyword. Every phrase is cross-checked against its own top-10: an exact name match on a weak app gets its score zeroed, with the evidence attached to the row.
type the name of any abandoned app into store search — it still autocompletesQuestion-style and four-plus-word phrases from our probes that turned out to be real queries: zero out of 102. People do not type questions into App Store search. Your character budget goes only to queries that exist.
type any "how do i…" phrase into store search and watch the suggestions stay emptyShip-ready metadata, with receipts.
Title and subtitle from real queries
Assembled from whole top-scoring phrases, so they read like people talk. Your best phrase goes into the title verbatim.
A packed keyword field
Set-cover optimized: no word repeated across any field, coverage report showing which queries each field wins.
The second indexed localization
Apple indexes one extra locale per storefront. Most listings leave it empty; a run fills it, guaranteed non-overlapping.
Full research trail
Every probed keyword with P, D, R and score — sortable, filterable, exportable, re-runnable from cache.
The deal, plainly.
You get
- Full runs, free, for the whole beta period
- Local-first: briefs and research never leave your Mac
- Founding price locked when paid launch happens
- A direct line to the author — fixes ship in days
We ask
- Run it on a real app you ship
- 15 minutes of honest feedback after the first run
- Optional: a before/after case with numbers
devs with a live app in the store get onboarded first
Is it actually free?
During the beta — yes, fully. After launch it becomes pay-per-run credits, not a subscription: ASO is episodic work, and a subscription would mostly bill you for the months you don't touch it. Beta testers keep founding pricing.
Where does my data live?
On your Mac. Your brief, your runs and your keyword research never reach our servers — there are no servers in the loop. Requests to Apple go directly from your machine.
What exactly does the AI do — and not do?
It extracts product context from your brief (which you confirm) and scores keyword relevance on a fixed 0–3 rubric, with a written reason each time, logged in full. It never produces demand or difficulty numbers — those are computed by deterministic code from Apple responses you can open.
Why trust your scores over the big tools?
Don't trust them — check them. The formulas are printed on this page, and every number in a run expands to the raw Apple response it came from: the exact prefix and rank for P, the ten apps behind D, the full model call behind R.
Which storefronts?
20 storefronts at beta start, US included, each with the correct extra-locale pair. The semantic language is configured separately — for example Spanish-language semantics for the US store.
When do I get in?
Onboarding goes in small batches so every tester gets real attention. Earlier signups go first; devs with a live app in the store skip the queue.
Metadata you can audit.
Free during the beta. Small batches, live apps first.
Free during beta. No card required.