How to Measure ASO Metadata Changes

Yusuf Yılmaz · November 16, 2026 · 8 min read

Short answer

Record your rankings for the keywords you are targeting before you change anything, note the day the new version went live, and compare the same keywords in the same countries over equal periods before and after. Then check App Store Connect for impressions, product page views and downloads over the same dates. A before/after difference shows what moved around your update; it does not prove your update caused it.

Why metadata changes need measurement

Apple says App Store search results are based on several factors, including text relevance (matches for your app’s title, subtitle, keywords and primary category) and user behavior such as downloads, ratings and reviews. Your metadata is the part of that list you control directly.

Without measurement, metadata edits turn into guesswork. Say you add “contraction timer” to the keyword field of Meyu: Pregnancy Tracker, and a week later downloads feel higher. Maybe it was the keyword, or maybe a competitor dropped out of the results. If you don’t record what changed and when, you can’t tell.

Measuring also protects what already works. The app name and subtitle are each limited to 30 characters and the keyword field to 100, so adding a word usually means removing another. A careful comparison shows what you lost as well as what you gained.

Establishing a baseline

A baseline is how your keywords ranked before you touched anything. Without one, an “after” number has nothing to be compared against.

  • Track your keywords for at least two weeks before the change ships, so you know their normal day-to-day range.
  • Record the rank for each keyword in each country you care about, once a day, at roughly the same time.
  • Save the exact old and new text of the name, subtitle and keyword field for every localization you edit.
  • Write down anything else happening around the same dates: a feature release, an ad campaign, a price change, a mention in the press.

If a keyword is outside the results you can see, record it as “not found” rather than inventing a number. “Not found” is a valid baseline.

Tracking target keywords

Before the change goes live, split your keywords into two groups:

  • Target keywords are the ones you expect to move. If you add “baby bump” to the keyword field, “baby bump” and phrases that combine it with your existing words are targets.
  • Control keywords are keywords you already rank for and didn’t touch. They show whether the whole results list moved for reasons that have nothing to do with you.

Track the words you removed as well. If you dropped “due date” to make room, a decline there is part of the price of the change. For help choosing what goes into the 100 characters, see our guide to the App Store keyword field.

Comparing before and after

Compare equal periods on either side of the day the new version went live, not the day you submitted it. For each keyword and country, look at:

  • The typical (median) rank in each period, not the single best day.
  • How many days the keyword appeared in the results at all. For keywords that were often not found, this says more than any average.
  • What your control keywords did over the same dates.
KeywordCountryRoleMedian rank, 14 days beforeMedian rank, 14 days after
affirmationen GermanyTarget#38#21
positive gedanken GermanyTarget200+ (0 of 14 days)#64 (14 of 14 days)
selbstliebe GermanyControl#17#19
daily affirmations United StatesControl#45#44
Illustrative example with made-up numbers, not real data. Somedaily: Affirmations updates only its German keyword field; “selbstliebe” was already there and wasn’t touched, and the US store doesn’t use the German localization.

Read the example like this: both target keywords improved in Germany, the German control keyword stayed roughly flat, and the US keyword, which the change couldn’t reach, didn’t move either. That pattern is consistent with the change helping. If the control rows had improved just as much, the gain would more likely come from something else.

Country-level differences

Metadata is localized, so an edit to one localization only reaches the storefronts and users who see that localization. Apple’s own example: if your primary language is English and you add a French localization, the French version is shown to users whose language setting is French and in countries or regions where the App Store supports French but not English, and people can find your app with your French keywords in every country or region where the App Store supports French. Where no localization matches, Apple uses the next most relevant one, and elsewhere your primary language is shown.

  • Before comparing, write down which storefronts actually show the localization you edited. Those are the only places you should expect a direct effect.
  • Compare each country on its own. Averaging Germany and the United States together hides exactly the difference you’re looking for.
  • Use storefronts whose localization you didn’t touch as a control, as in the example above.
  • Expect different results even in countries that share a language; each store has its own competitors.

Separating keyword movements from other factors

Rankings move for many reasons. Before you give your metadata the credit (or the blame), check these:

  • Competitor moves. A new app enters the results, a competitor ships an update, or an app disappears from the store. You can see who ranks above you on a given day, but not what is in their keyword field.
  • Seasonality. Some searches rise and fall at predictable times of year. An affirmations app may behave differently around the new year than in midsummer.
  • App Store changes. Search results can shift across the board for reasons outside your app. If many unrelated keywords and countries move on the same day, including ones you didn’t edit, suspect the store rather than your change.
  • Ratings and reviews. Apple says ratings and reviews can influence how your app ranks in App Store search. A wave of new reviews can move rankings on its own.
  • Your other releases and promotion. A new feature, a price change, ads, a press mention or a social post can all change downloads, and downloads are part of the user behavior Apple lists among search factors.

Your control keywords and untouched countries are the main defence here.

Using a 14-day window carefully

Fourteen days on each side is a practical default. It is long enough to smooth out daily noise, and two full weeks means both periods contain the same mix of weekdays and weekends. It still has limits:

  • Rankings can be unsettled for the first few days after a version goes live, so don’t judge the result on those days alone.
  • If you ship another change inside the window, the two overlap and you can’t separate their effects. Where you can, leave at least one full window between metadata changes.
  • A holiday, a competitor launch or a featured placement that falls inside one window but not the other will skew the result.
  • For keywords where you rank low or are often not found, 14 days may not be enough to show a clear direction.

If you use AsoGlass, this is what its metadata change tracking does: you set a change marker manually, and it compares the 14 days before and after that marker for each country and target keyword, next to your other tracked keywords. Because markers are manual, put yours on the day the version went live. The comparison is a measurement aid, not proof of causation; it shows what moved, and the checks in this guide are how you decide why. The underlying data comes from daily keyword rank tracking, and with the optional App Store Connect connection the same markers appear on your download charts.

Combining rankings with App Store Connect analytics

A better rank only matters if more people see and download your app. App Store Connect gives you the other half of the picture. Apple defines the main metrics like this:

  • Impressions: the number of times your app was viewed on the Today, Games, Apps and Search tabs of the App Store for more than one second, including product page views.
  • Product page views: the number of times your product page was viewed, including when other apps load it with StoreKit.
  • Total downloads: first-time downloads plus redownloads.

Look at the same countries and the same dates as your ranking comparison, and focus on downloads that came from App Store search, since that is the traffic a keyword change can affect. Then read the signals together:

  • Target rankings up, plus more impressions and search downloads in the same country: the story is consistent.
  • Rankings up, impressions flat: the keywords you gained may bring little traffic.
  • Impressions up, product page views flat: people see the new name or subtitle in results but aren’t tapping through.

Mind the timing. Sales and Trends daily reports are available the following day, generally by 8 a.m. Pacific Time, so the last day of your “after” window may not be in yet. For small apps, note that Apple only shows App Store metrics once an app has at least five first-time downloads or pre-orders. With numbers that small, a handful of downloads swings any percentage a lot.

Common mistakes

  • Changing the name, subtitle, keyword field, screenshots and icon in one release, then crediting one of them.
  • Using the submission date instead of the date the version went live.
  • Tracking only the new keywords and missing the ranks you lost by removing others.
  • Expecting movement in countries that don’t show the localization you edited.
  • Averaging ranks that include “not found” days as if they were a number.
  • Calling a rank gain a win without checking impressions and downloads.
  • Checking rankings by searching on your own phone. You only see your own country’s store, and your results can be slightly personalized.

FAQ

Do I need a new app version to change my metadata?

For the app name, Apple’s App Store Connect Help says you can change it when you create a new version or when the status of the app version permits editing. Check in App Store Connect which fields are editable for your current version. In practice, name, subtitle and keyword field changes usually ship together with a version update, which is why the live date of that version is the date to measure from.

How long should I wait before judging a change?

At least two full weeks after the version goes live, and longer for keywords where you rank low or are often not found.

My target keywords improved but downloads didn’t. Was the change wasted?

Not necessarily. You may have gained ranks on keywords that few people search for, or moved from a position nobody scrolls to into another one nobody scrolls to.

Does a before/after comparison prove that my change worked?

No. It shows what moved around the date of your change. Control keywords, untouched countries and App Store Connect data make the case stronger or weaker, but other factors can always play a part.

Sources Apple Developer, “App Store search” App Store Connect Help, “App information” App Store Connect Help, “Localize app information” App Store Connect Analytics Help, “Metric definitions” App Store Connect Help, “Sales and Trends reports availability”
Metadata· 9 min read
How to Optimize the App Store Keyword Field Learn how Apple's 100-character App Store keyword field works, avoid wasted characters and duplicate words, and organize keywords more effectively for ASO.
App Store Connect· 10 min read
Understanding App Store Connect Analytics for ASO Understand App Store downloads, impressions, product page views and App Store search acquisition using Apple's official analytics data and report timing.
Keyword rankings· 3 min read
How to track App Store keyword rankings What an App Store keyword ranking is, why it differs from country to country, the limits of checking by hand, and how to track your rankings every day.