Browse by category
All categoriesBrowse by platform
All platformsResources
Latest postsFree tool
Paste an App Store or Google Play link. You get a score against the norms that actually move installs, a ranked list of what to fix first, and every raw field we could read.
What it scores
Not a checklist of every field in the console. The seven that move the number, weighted by how much they move it.
Both stores cut the title around 30 characters, the App Store subtitle is an indexed field most listings waste, and roughly three lines of the description show before the more link. That is the whole pitch.
Five to ten, with the strongest first, and more than one device set where the app runs on more than one device. A phone-only set on an app that runs on tablets reads as unfinished.
Rating volume against credibility thresholds, the last update against a 90 day window, and how many languages the metadata is in. Metadata localization is the cheapest reach in the whole listing.
Method
For an App Store link we take the numeric id out of the URL and ask Apple’s public lookup endpoint for the listing. Title, seller, description, screenshot sets, rating, rating count, last release date, languages, size, genre, price and content rating all come back structured, so those numbers are exact rather than scraped.
The subtitle is the one App Store field the lookup does not return, so it is read from the store page itself and reported as unavailable when the page will not give it up. That is deliberate: a checker that guesses at a field is worse than one that says it does not know.
For a Google Play link there is no public API, so the store page is fetched and read with deliberately loose patterns. Play changes its markup without notice, so every field is marked parsed or unavailable, and an unavailable field is dropped from the score rather than counted as a failure. Expect a Play report to be thinner than an App Store one.
Scoring is weighted against norms that move installs: title inside 30 characters, a subtitle that adds something rather than repeating the title, five to ten screenshots covering more than one device set, an opening 300 characters that states the benefit before the more link cuts it off, enough ratings to be credible, an update inside 90 days, and localization past a single language. The fix list is ranked by how many points each item is actually costing you.
