We are building the app-intelligence tool we wanted to trust.
Developers make real decisions with app-market numbers — what to build, where to launch, how to price. Those numbers are estimates. We think you deserve to know how good each one is.
- 01
Show the work
Every number carries its range, its grade, its inputs and the method version.
- 02
Own the data
We collect public signals ourselves, without logging in, with an honest User-Agent.
- 03
Admit the limits
Thin data gets a wide band or no number at all. Errors get published.
- 04
Agents are users
The API and the MCP server are first-class products, not an afterthought.
Built like a cat: sees in the dark, stares back
Most app-intelligence tools hand you one confident number. We hand you the number, how sure we are, and where it came from.
- Night vision
The app market is dark. We read the faint signals — charts, ratings, prices, ads — and show you what moves.
- The stare
We look straight at you, including when we are wrong. Errors get published next to the method that made them.
- Grooming
Clean data, every day. Every field keeps its source, its kind and the time we saw it.
- The tail
Drawn as the long-tail curve we model: a few apps take most of the downloads, and we say how sure we are about the rest.
No logo wall. Just commitments you can check.
We are pre-launch, so you will not find made-up customers or star ratings here. These are the rules the product ships with — each one is enforced in code and tested.
P10, P50 and P90 — on the dashboard, the API, the MCP server, webhooks and exports. When we are unsure, the band says so.
Curated by hand, not mirrored from the API. Every response tells your agent what it cost and where to continue.
At launch: no card, no trial clock. Enough to answer real questions before you decide to pay.
Where we are, honestly
Targets, not promises. We update this page as each step lands — including when one slips.
Monorepo, design tokens, architecture decision records and a dated study of the category built only from public sources.
Color system checked for contrast and color blindness, the vector mascot and a full brandbook in Figma.
The marketing site you are reading, then the high-impact version: scroll story, WebGL night vision, EN/ES.
iOS catalog and charts first, then Google Play, reviews and keywords. Every source behind an adapter with a kill switch.
The waitlist opens with the first Data Drops. Your position is real; there is no fake counter.
Partner apps share real numbers to calibrate the model and get free access in return.
Dashboard, API, MCP and billing — after at least four weeks of beta and thirty days of on-time data in a row.
Questions a skeptic would ask
Good. Ask them. Here are the straight answers.
Still curious?
A person reads every email.
Stop guessing in the dark.
The free plan, the first Data Drops and the private alpha go to the waitlist first.