Observed, not modelled
Where does my app actually come up?
BirikiAso asks Apple's public App Store search the same question a person would type, then reports the position your app came back at — or that it did not come back at all. Every rank on the screen was returned by Apple for that phrase, on that day. The two scores beside it are worked out from those readings and print the inputs they came from. Nothing is turned into a search volume or a download count.
Steps
Four steps, about a minute.
01
Name the app
Paste an App Store link, an Apple ID or the app's name. The app does not have to be yours — any public listing in the market you picked works, including a rival's.
02
Enter up to twenty keywords
One per line, or comma separated. Phrases this service has already seen the app come up for are offered under "already seen", and phrases Apple Ads suggests appear under the table. Those are leads to check, not proof the app ranks.
03
Your browser asks Apple
Each search leaves from your own connection, so nothing queues behind other visitors. Results stream in as they land; a phrase Apple answers with 200 results is read to that depth.
04
Read the evidence
Rank, the depth Apple actually returned, the rivals that came back with you, their public review counts, and one plain next move per keyword. A saved keyword is checked again on the server every day, so its history grows on the days nobody opens the page.
Context
What ASO is, briefly.
App Store Optimization is the work of being findable in the store's own search. Most installs still begin with someone typing a phrase, so the question that matters is narrow and practical: for the phrases that describe your app, does it appear at all, and who is standing above it?
That question has an observable answer. Apple returns an ordered list for every phrase, identical no matter who is asking. Your rank is simply your position in that list — and if the list ends before your app appears, the honest reading is "not in the 200 results returned", not a guess at what position you might hold beyond them.
AI agents
Ask Claude instead of reading the table.
BirikiAso runs a read-only MCP server, so an AI agent can answer from your workspace: which apps you track, where each keyword stands, its rank history, Apple's popularity, and the Field and Opportunity scores. It reads the same stored readings as the board. Each estimate is marked as an estimate, and a keyword Apple did not return reads "not in the top N" for the N results Apple returned (at most 200), never a rank.
It cannot start a scan, call Apple, or change anything in the workspace. Your notes and tags stay out of it. Each key is limited to 60 calls a minute and 2,000 a day.
Setup for Claude Code, Cursor and Claude Desktop is on the MCP page.
Limits
What this will never tell you.
Apple's public search API publishes no search volume, no download counts and no revenue. Tools that show those numbers are modelling them from other sources. BirikiAso does not.
Popularity is Apple's own 0–100 figure from Apple Ads, relative and not a count of searches. Apple returns the same value in every market. At Apple's lowest value it reads "Low", never a number.
Field (how hard the top results are to get past) and Opportunity (our estimate of which saved keyword is worth working on) are derived scores. The weights are ours. Every row shows the inputs behind them, and a score with thin evidence is marked "low". Neither predicts a rank or a download.
Review counts appear beside rivals because Apple returns them, and they are useful context for how established a competitor is. They are not a target and not a cause of ranking.
A reading belongs to one App Store market. The United States and Türkiye return different apps in a different order for the same phrase, so an app saved in one is watched, scored and kept apart from the same app saved in the other — and nothing here compares the two or averages them into one number.
The rule the product is built on: if Apple did not return it, you will not read it here.
Now and next
What is running, and what is only intended.
Without an account, a workspace lives in your browser, and clearing site data starts a fresh one. Signing in with Google keeps the workspace you already have and does not start a second one.
Saved keywords are checked on the server every day. The Market tab shows the apps these scans have returned, with the measured change in each one's public review count and the time span it was measured over.
Being considered: an optional email when a saved keyword drops.
That is an intention, not a promise with a date. It is not built until it appears on the screen.
Feedback
Tell us what is wrong with it.
This is an early open trial and the fastest way to change it is to say what got in your way — a phrase that returned something odd, a number that read wrong, a step that felt pointless.
Written by @buraknevruzoglu on X — replies and DMs are open, and that is also where changes get posted.
Free · 20 keywords per batch · no account needed