Localized Keywords Per Storefront in Apple Search Ads: Why a Translated Term Isn't the Same Query

If you run Apple Search Ads in more than one country, you have probably copied your English keyword list into a second campaign, maybe translated a few terms, and hit start. It feels efficient. It is also one of the most common reasons a second storefront looks broken when it is really just mismatched.

The issue is simple: a translation is not the same thing as a search. People in Germany, Brazil, or Japan do not type the literal equivalent of your English keyword. They use their own phrasing, their own abbreviations, their own competitor names, and their own assumptions about what an app should be called. Each storefront has its own query vocabulary, and your keyword list needs to reflect that.

What actually changes between storefronts

Remember the basic structure. One campaign targets one country or region. Ad groups hold keywords and bids. Each keyword is matched against queries typed in that specific App Store. So a keyword list is really a per-storefront asset, even if the words look similar on paper.

Three things tend to differ:

• Phrasing. A literal translation may be grammatically correct and still rare. The common term might be a loanword, a compound, or a category name you never considered.
• Competitor landscape. The apps people compare yours against change by market. A rival that dominates the US charts may barely exist in another storefront.
• Intent shape. The same category word can lean toward a tool ("converter") in one language and toward a brand or a sub-genre in another.

Build a local seed list from scratch

Don't start by translating your English list. Start with what local users would type.

• Open the App Store in that country and search for your core job-to-be-done. Note the autocomplete suggestions. These are close to real demand patterns.
• Look at the top results for those queries. Which apps show up, and what words appear in their titles and subtitles? Those words are signals of how the market names the problem.
• Read a few local reviews of apps in your category. Users describe what they wanted in their own words, and those words are often better keywords than anything you would brainstorm.
• Ask a native speaker you trust to write the first ten queries they would type. Not translations of your keywords, but the queries themselves.

You are building a list that comes from the market, then checking it against your product.

Keep a per-country keyword table

For each storefront, keep a simple record:

• Keyword and match type (exact or broad)
• Why you think it is relevant locally, in one line
• Source: autocomplete, native speaker, competitor review, or your own guess
• Max CPT and the date you set it

The source column matters more than it looks. When a keyword underperforms six weeks from now, you want to know whether it was a guess or something you verified, because that changes what you should do next.

Don't compare countries on one blended number

This is where most second-storefront confusion starts. You look at a dashboard showing total ROAS and assume the German campaign is doing what the US campaign does. Each country has its own auction density, its own CPT levels, its own price points, and its own conversion behavior. A keyword that converts well in one market can fail in another with identical wording because the surrounding competition is different.

Judge each storefront against its own baseline. For example, if your US campaign has a typical tap-through rate that sits in one range, a new market should be assessed against its own first few weeks of data, not the US figure. Illustrative only: if a German campaign's TTR is lower than the US one, that alone does not tell you the German keywords are wrong. It may just mean the results page looks different there, or your creative is less relevant to how that market scans.

Watch for translation traps

A few patterns show up again and again:

• Literal translations that nobody types. The phrase is correct but absent from autocomplete. Impressions stay near zero, and the campaign looks dead.
• Brand names that do not carry over. Your brand may be unknown in a new market. Bidding on it there is not the same as bidding on a known brand at home.
• Mixed-language queries. In some markets people mix English terms into their search. A purely local list can miss those.
• Overconfident broad match. Broad will find many variants, but if your seed terms are wrong, it will spend on mismatched queries. Check the search term report early.

Use the search term report to correct course

After the first couple of weeks, pull the search term report for each storefront. Look for:

• Queries you did not expect that are clearly relevant. Add them as exact keywords in that storefront with a sensible starting bid.
• Queries that are relevant to your category but describe a different job. Add them as negatives in that country.
• Keywords with healthy impressions and taps but no installs over a meaningful sample. Treat them as candidates to lower or pause, not as proof the market is weak.

Do not treat a single week as the verdict. Installs can lag, and small volumes are noisy.

A simple rule for localization

If you cannot explain, in one sentence, why a keyword is what a local user would type, it does not belong in that storefront yet. Localize with intent, not with a dictionary.

Takeaway

Localizing Apple Search Ads is less about translation and more about rebuilding your keyword list from how each market actually searches. Start from local autocomplete, native input, and competitor language. Keep a per-country table with sources. Judge each storefront against its own numbers, and use the search term report to fix mismatches early.

If you want a daily, prioritized list of which storefront keywords need attention and why, AdsBuddy can surface those items from your ASA and revenue data. You still review and approve every change yourself.