How App Store "estimated rank" is actually measured

One request returns at most 200 results

The limit parameter on the iTunes Search API (itunes.apple.com/search) only accepts a range of 1–200, with a default of 50 — this is documented in Apple’s own reference (see Sources). ASO Lens requests with limit=200 and converts an app’s position in that list into a 1-based “estimated search rank.”

That means an app sitting at position 201 or below can’t be measured this way at all — not approximately, not roughly, not at all. When that happens, ASO Lens doesn’t invent a number; it shows “Out of range (>200)” instead. That’s not a rounded estimate — it’s an honest statement of what the API structurally cannot tell you beyond its own window.

“X search results” means different things to different tools

The iTunes Search API’s response includes a resultCount field. It is not “the total number of apps that exist for this keyword” — it’s “the number of items this particular request returned.”

This is something we verified directly against the live API, not something we’re inferring. Searching the same keyword with limit=5 returns a resultCount of 5; with limit=200, it returns up to 200. In other words, resultCount isn’t a market-size metric — it moves with whatever limit the request asked for.

ASO Lens’s keyword difficulty score (0–100) does not use that raw count as-is. It’s a heuristic blend of three signals from the top 10 results: the median rating count, the share of titles that match the keyword, and the average rating — the raw result count is shown separately as reference, not folded into the score. If another tool shows “X search results” as the basis for a difficulty figure, there’s no way to know from the outside whether that’s the API’s raw resultCount or something computed differently. Since the methodology can differ, comparing the raw numbers side by side across tools isn’t comparing like with like.

A rank is a snapshot, taken under one specific set of conditions

The rank ASO Lens records comes from a signed-out, non-personalized search request. What an actual App Store visitor sees can differ by storefront (country/region), by time of day, and by that visitor’s own personalization based on their usage history.

Apple itself states that search results are shaped by a combination of factors — text relevance (matches against the app name, subtitle, keywords, and category) together with signals like downloads, ratings, reviews, and engagement (see Sources). ASO Lens’s estimated rank does not reproduce that internal algorithm. It’s a record of “what order the public API returned results in, at one specific moment.”

What ASO Lens does — and doesn’t — do

Does:

• Queries the iTunes Search API (or an equivalent path) for the app’s storefront and records the tracked app’s position as a 1-based rank
• Caches each (storefront, keyword) result so the same pair isn’t re-fetched more than roughly once every 24 hours
• Stores that rank daily, so you can see a keyword’s trend over time, not just one reading

Doesn’t / can’t:

• Reproduce or reverse-engineer Apple’s actual search-ranking algorithm
• Retrieve personalized results shaped by a visitor’s sign-in state or usage history
• Determine an exact rank below position 200 — it can only report “out of range”
• Count the total number of apps that exist for a given keyword

Summary

We call it an “estimated rank” not to hedge, but because that’s literally what the measurement method allows. ASO Lens is built to disclose that constraint openly, and to track rank consistently over time within it — a trend you can act on, not a number pretending to be more precise than the API that produced it.