# How ASO tools estimate downloads

Source: https://asostudio.app/aso-download-estimates

How every ASO tool builds a download and revenue figure for an app it has no sales data for, and how real installs correct it.

## How the figure is built

### 1. Apple publishes no download counts for any app but your own
Sales reports exist only for the developer who owns the app. Everything a tool prints about someone else's app is inferred from what Apple does publish: ratings, reviews, price, chart positions and the listing itself.
**What to check:** Ask any vendor which public signal their figure is built from. If the answer is a panel or a proprietary model, ask how large the panel is.

### 2. The figure is written reviews times a rate
The most common method counts how fast an app gains written reviews and multiplies by an installs-per-review rate for its category. Ratings arrive twenty to fifty times more often than written reviews, so the rate is large and the estimate swings with it.
**What to check:** Two tools disagreeing by three times are usually using two different rates on the same reviews.

### 3. The rate is a category average, and nobody publishes it
No source publishes installs per written review by category, so vendors compose it from two halves: installs per rating, a trade assumption near one in eighty, and ratings per review, which can be measured from public data. Both halves are the same for every app in a genre.
**What to check:** A fitness app that prompts after a workout and a banking app nobody rates cannot plausibly share a rate, and in every tool they do.

### 4. Small apps cannot be estimated at all
Under about a hundred ratings there is no sample to read a rate from. A launch week where sixty friends each leave a review looks like a thousand installs a day to a naive model.
**What to check:** A tool that prints a confident figure for an app with forty ratings is printing noise.

### 5. Revenue stacks a second guess on the first
Revenue is the download estimate times a conversion rate times a price. The conversion rate is a category median, and the price has to be read off the product page and blended across weekly, monthly and yearly plans. Each step multiplies the error of the one before it.
**What to check:** Ask which conversion rate and which price a revenue figure used. A tool that will not say is stacking two averages and showing you one number.

### 6. What changes when real installs enter the model
A user who connects their own App Store Connect account has real installs for their apps. Beside the written reviews the same app gained in the same month, that is the rate measured, for that category, in that storefront. Pooled anonymously across users, the measured rate replaces the average for everyone.
**What to check:** Look for a figure that says which of the two rates it used. ASO Studio marks a calibrated figure with a seal and says how many anchors stand behind it.

## Frequently asked questions

### How accurate are ASO tool download estimates?
For apps other than your own, every figure is a model, because Apple publishes download counts only to the developer who owns the app. The common method multiplies how fast an app gains written reviews by a category-average installs-per-review rate, so two tools can disagree by several times on the same app and both be following their model. Treat the figures as a way to rank apps against each other, not as counts, and expect nothing usable for apps under about a hundred ratings.

### Where do ASO tools get download numbers from?
From public signals: how quickly an app gathers ratings and written reviews, its chart positions, its price and its in-app purchase list. Larger vendors add panels of devices that report what is installed on them. None of them have Apple's own count for an app they do not own.

### Why do different tools show different download numbers for the same app?
Because the rate they multiply reviews by is different, and no source publishes the right one. Installs per written review is composed from a trade assumption about installs per rating and a measured ratings-per-review ratio, and each vendor composes it differently. The reviews are the same; the multiplier is not.

### Can download estimates be made more accurate?
Yes, with real installs from the apps whose owners are willing to contribute them. ASO Studio reads a connected user's own monthly sales report, counts the written reviews the same app gained in the same month, and sends the pair as one anonymous anchor per app per storefront. The pool fits a measured installs-per-review rate per category from those anchors, and every figure built on a measured rate carries a seal that says so. The more people contribute, the closer everyone's figures get to real.

### Is my sales data shared when I contribute?
No. An anchor is the category, the storefront, the month's first-time installs, the month's written reviews and the window length. It carries no app id, no name and no account, and nothing from it can be traced back to an app.

ASO Studio uses its own algorithms to model keyword demand, competition, and market size from public App Store signals. The models draw on search results, autocomplete, ratings, review activity, and pricing to help you compare opportunities. Keyword scores are relative indexes, not search-volume counts. An optional Apple Search Ads connection supplies Apple's own search popularity index, which is distinct from the modeled scores. Other apps' download and revenue estimates cover the stated sample; they are not Apple-reported totals. Your own downloads and revenue come from App Store Connect when connected. Keep measured observations, modeled estimates, and missing data distinct.