How to Measure ASO Results Beyond Downloads (2026)

To measure an ASO update, compare two equal periods, before and after the release, at three levels: how many people saw the app in search, how many of them the listing convinced, and how many of the new downloads reached an early step and a value step inside the app. Downloads alone show that more people came. They do not show that the right people came.

Why downloads are not the result

A metadata update changes who finds your app. New words in the title, subtitle or keyword field match different searches, and different searches bring people with different expectations. If the app gives them what the search promised, downloads rise and the new people stay. If it does not, downloads can rise just the same while more of the new installs open the app once and leave.

Both outcomes look identical in a downloads chart. So an ASO result has three parts: did more people see the app, did the listing convince them, and did the people it convinced find what they were looking for.

What each tool can and cannot tell you

Three kinds of tools cover the journey, and each one stops where the next begins.

Keyword research

Tells you: how popular a search is, how hard it is to rank for, which apps rank today and where yours stands.

Does not tell you: who downloads, or what they do next.

App Store Connect Analytics

Tells you: impressions, product page views, conversion rate and downloads, split by source type (App Store Search, App Store Browse, Web Referrer, App Referrer) and by territory.

Does not tell you: which search term led to an organic download, or which steps people take inside your app.

In-app analytics

Tells you: which events happen in sessions, where onboarding loses people, which events go with a conversion, by country and platform.

Does not tell you: which keyword or ad brought a session. Respectlytics keeps no persistent user ID, so it counts sessions and events, not people.

No tool closes the gap between a search term and what happens in the app for organic traffic. Apple does not report which search term led to a download, and Apple Ads reports search terms only for your own ads. The method that works is a comparison over time: the same measurements, before and after the update.

Step 1: Write down what changed and when

The version, the day it went live, and exactly which fields changed: title, subtitle, keyword field, screenshots, a new localization.
Change one thing per release where you can. If the title and the screenshots change together, you will not know which one moved the numbers.
Anything else that moves the same numbers: a price change, your Apple Ads budget, a feature on the App Store, a holiday, a press mention.

Step 2: Pick two equal windows

Use the same number of days on each side, covering the same weekdays. Two or four weeks each works for most apps.
Leave out the day the new version goes live. Rankings can keep moving for a few days after a release, so a longer window gives a steadier picture than the first days alone.
As a rule of thumb, if either window has fewer than 100 first-time downloads, lengthen both. Small numbers swing by chance.

Step 3: Pull the store numbers

In App Store Connect, open Analytics and set the date range for each window. Filter by the source type App Store Search, because that is where keyword changes act. If the update was one localization, filter by its territory as well. Write down three numbers per window:

Impressions (Unique Devices): devices that saw your app in search for more than a second.
Product Page Views (Unique Devices): devices that opened your product page.
First-Time Downloads: new installs, without reinstalls.

Two things to know.

App Store Search includes Apple Ads taps. If you run search ads, keep the budget the same across both windows, or the comparison mixes paid and organic changes.

People can tap Get in the search results without opening your product page, so downloads can be higher than page views. Measure both against impressions rather than one against the other.

Step 4: Pull the in-app numbers

Pick two events and count them over the same dates:

An early step that happens once per install, such as onboarding_completed or first_project_created. It shows whether new installs get started.
A value step, such as trial_started, purchase or the core action of your app. It shows whether they found what they came for.

In Respectlytics, the dashboard gives event counts for any date range. A funnel between the two events shows where sessions stop, conversion analysis shows which events go with your conversion event, and segment comparison splits the numbers by country and platform, which is what you want after a localization.

Respectlytics is session-based: session IDs live in memory and rotate every two hours or when the app restarts, so it counts sessions and events, never people. For this method that is enough, because you compare how often a step happens per download, before and after.

Step 5: Compare rates, not totals

Totals grow with any traffic. Rates show whether the update changed who arrives. Work out four for each window:

Page view rate: product page views ÷ impressions.
Store conversion rate: first-time downloads ÷ impressions.
Early step per 100 downloads: early step events ÷ first-time downloads × 100.
Value step per 100 downloads: value step events ÷ first-time downloads × 100.

The free ASO funnel calculator does this arithmetic in your browser, shows both windows side by side and lists what to check where a step moved.

Open the ASO funnel calculator

How to read the result

Each pattern points at a different part of the journey.

Impressions up, page view rate down

The app now appears in searches where its icon and title do not look relevant.

Try: check that the title answers the searches you added.

Page view rate steady, conversion rate down

The product page does not convince the people who now open it.

Try: make the first screenshots show what the new keywords promise.

Downloads up, early step per 100 downloads down

New installs expected something the app does not show them first.

Try: compare the title and subtitle with the first screens of the app.

Early step steady, value step down

The problem sits inside the app, after onboarding.

Try: build a funnel between the two events and find where sessions stop.

Every rate held or rose

A sign the update brought more people who find what they came for.

Try: keep the change and test the next one.

Nothing moved

The change was small, or the windows were too short to show it.

Try: lengthen both windows before deciding.

These are hints, not proof. A comparison over time cannot rule out everything else that changed in the same weeks, which is why Step 1 asks you to write those things down.

Limits worth knowing

No keyword-level answer for organic search. You learn what the update as a whole did, not what each keyword did.
Two systems, two ways of counting. App Store Connect counts devices and your analytics counts events, and the two may cut days at different hours. Ratios per download are good for comparing windows, not as exact figures.
Seasons and outside events. A holiday, a feature on the App Store or a competitor's launch can move every number at once.
Small apps need patience. With few downloads, use longer windows and look for large, steady moves.

Questions

Can App Store Connect tell me which keyword brought a download?

Not for organic search. App Store Connect splits impressions, page views and downloads by source type and territory, but not by search term. Apple Ads reports search terms only for your own ads, so the practical method is to compare equal periods before and after an update.

How long after a metadata update should I measure?

Leave out the day the new version goes live and compare two to four weeks on each side. Rankings can keep moving in the first days after a release, and a longer window gives a steadier picture.

Do I need user IDs to measure ASO results?

No. The method compares how often an event happens per first-time download in two periods, which needs counts, not identities. Session-based analytics such as Respectlytics is enough.

Which keyword research tool should I use before the update?

Any tool that shows popularity, difficulty and current ranks works. RespectASO is a free, open-source Mac app from the team behind Respectlytics that covers 175 App Store storefronts.

Before the update: choose the keywords

Measuring tells you whether an update worked. Deciding what goes into it is keyword research. RespectASO is a free, open-source Mac app (AGPL-3.0) from the team behind Respectlytics: it scores how popular a search is and how hard it is to rank for, shows the apps that rank today and tracks where yours stands, across 175 App Store storefronts.

How to Measure ASO Results
Beyond Downloads