Judged, not guessed
Every candidate gets four calibrated judgments from a decision model: does the searcher want an app like yours, does the store already rank your kind of app for it, is it a phrase people really type, and what kind of query is it.
Native macOS · iOS App Store keywords
Keyword research judged for your app.
Sextant mines what the App Store already says about your competitors, then asks a decision model whether each keyword's searcher actually wants an app like yours. Not popularity guesses. Judgments, with the odds shown.
Lovit · Wishlists for CouplesUS storefront · 7 of 30 judged keywords
| Keyword | Intent | SERP fit | Kind | Score | Placement |
|---|---|---|---|---|---|
| gift wish list | 72% | 69% | Generic category | 70 | Title or subtitle |
| wishlists | 61% | 63% | Generic category | 67 | Title or subtitle |
| duplicate gifts | 66% | 51% | Specific need | 64 | Keyword field |
| gift ideas | 46% | 57% | Generic category | 57 | Review |
| couples app | 39% | 17% | Generic category | 38 | LowThe store ranks dating apps here, not wishlists. |
| wish | 22% | 14% | Generic category | 29 | LowSearchers mean the marketplace. |
| secret santa | 9% | 17% | Specific need | 25 | SkipNot our searcher. |
Mined from the store, not invented
Candidates come from competitor titles and descriptions and from App Store autocomplete. Nothing is generated. Competitors that are not really your kind of app are filtered out before their vocabulary counts.
The keyword field, packed
Generic terms go to your title and subtitle, needs to the 100-character keyword field, brands are skipped. Sextant packs the field for you and shows the preview before you paste it into App Store Connect.
Native on your Mac
Rank tracking across 60+ storefronts, Apple Ads popularity, difficulty, translations and reviews, in a fast native macOS app. One flat price, unlimited apps and keywords.
Popularity tells you how often a phrase is searched. It does not tell you whether those searchers want you.
- Apple's popularity index has a floor of 5. In September 2025 most of the long tail dropped to that floor within days, in every ASO tool at once.
- Apple's top-terms dataset covers roughly 500 terms per genre, and about four in five of them are brand names.
- Sextant keeps popularity and difficulty as supporting columns. The ranking you act on comes from what the searcher means and what the store already rewards.
Questions
Is it Mac only?
Yes. Sextant is a native macOS app for macOS 14 and later, and it covers the iOS App Store at launch.
What leaves my Mac?
Your app's public listing, the candidate keywords and the top results for each query. They go to Sextant's judging service, which is how the decision model runs without shipping a key inside the app. Nothing about your Apple account or your sales.
What does the judging?
Jev, TypeSafe's System One model. It answers typed questions with calibrated probabilities instead of generating text, so it cannot invent a keyword. It only judges the ones the store already suggests, and you see the odds behind every score.
Why not rank by popularity like everyone else?
Popularity tells you how often a phrase is searched, not whether those searchers want you. Sextant shows popularity and difficulty too. It just refuses to rank your keywords by them alone.
How much will it cost?
One flat tier with unlimited apps and keywords, around $9 a month billed yearly. People on the waitlist get the founding-member price.
When?
Paid beta this winter. The waitlist hears first, and early testers shape the placement rules.
Be first to the beta.
Indie iOS developers on a Mac. One email when it opens, founding-member price for the list.