Localization Is ASO Most Devs Miss

Why does the same indie app rank top-3 in Germany and rank 117 in the US?

Localization.

Apple ranks the App Store per locale. The title, subtitle, and 100-character keyword field are separate ranking fields in each of the 27 storefronts Apple supports. Translating the app code is not required. Localising the metadata is. Most indie devs learn this and still skip it.

Why Apple treats localization as a ranking signal

Two structural facts. First, Apple wants to send a Spanish- speaking buyer in Mexico a result that the buyer can read. Second, Apple is willing to surface localised metadata even when the app itself remains English. The product page in Mexico shows the buyer a Spanish title, Spanish subtitle, and Spanish keyword field; the buyer installs; the app launches in English. Apple is fine with that. Buyers in Spanish-speaking storefronts who install English-only apps install them all the time.

What Apple does not tolerate is metadata in the wrong language for the storefront. A Spanish storefront product page with English metadata reads as broken. Apple either downranks the listing or routes around it.

The leverage is on the indie dev. Localised metadata moves you up rankings in 25 of the 27 storefronts because most of your category competitors have not bothered. The work is the cheapest sustained ranking lift an indie launch can buy.

The 5-locale strategy for English-native devs

Twenty-seven storefronts at once is too many. The right first pass is five. The five highest-revenue non-English storefronts for indie B2C apps in 2026 are typically Germany, France, Japan, South Korea, and Brazil. Together they account for a substantial share of Apple App Store revenue outside the US, and the keyword competition in those storefronts is consistently lighter than in the English-language storefronts.

For each of the five, localise four fields:

1. Title (30 characters).
2. Subtitle (30 characters).
3. 100-character keyword field.
4. First 170 characters of the description.

Skip the rest of the long description for the first pass. It does not rank on iOS, and the localisation of the rest can wait until the metadata pass proves out.

Time to first ship of the 5-locale pass with the workflow: about one weekend. The first time you do this, expect six hours per locale because you are pulling competitor keywords per storefront. Every locale after is faster.

Why literal translation is the most expensive mistake

Google Translate of your English keyword is almost always wrong.

"Habit tracker" in German is not "Gewohnheitstracker." What German users actually type into App Store search is something shorter, uglier, and rooted in the verb (`Gewohnheit aufbauen`, `Streak`, or category- specific verbs that no dictionary translation surfaces). "Sleep tracker" in Japanese is not the kanji-compound noun a dictionary returns. It is the romaji or katakana that Japanese buyers actually type, which depends on whether the iOS keyboard suggests it.

The cheap fix is a one-hour conversation with a native speaker who uses the App Store in that locale. Three rules for the conversation:

1. Hand them the App Store, not a CSV. Have them search the outcome keyword on the local storefront. Note what they type before they tap.
2. Pull the top ten organic results in that storefront. Have the speaker scan the first five characters of each title and tell you which words feel native and which feel machine-translated.
3. Ask them what they would search for if they wanted your app. Their answer is your keyword anchor. Not your dictionary.

Cost: one hour per locale at a normal freelance rate. Time saved: weeks of bad rankings.

Localising the screenshot captions too

Apple shows the localised metadata to buyers in those storefronts. Apple also shows the screenshot captions to buyers in those storefronts. If the captions stay in English while the metadata is localised, the listing reads as half-finished and conversion drops.

The screenshot caption pass is small. Four words per caption, five captions per app, five locales: a hundred words of native translation. Pay a native speaker for an hour or run the captions through a tool that does keyword-mapped translation. Either way, ship the captions in the same language as the metadata.

The full screenshot framework, including the localised caption pass, is in App Store Screenshots That Convert.

Maintenance cadence

Localised metadata is not write-once. Apple Search Ads popularity scores shift in every locale on a different cadence. Competitor rotations in Japan are not the same as competitor rotations in Germany. Refresh per locale every 90 days.

Three triggers for an off-cycle locale refresh:

• A new top-three competitor enters the local search results with a sharper anchor keyword.
• Your product-page conversion in that locale drops more than four percentage points week-over-week.
• You ship a real new feature. Update the localised subtitle first, then the caption row.

Most indie devs ship the 5-locale pass once and never refresh. Six months later the metadata is stale and the rankings have slipped. The cadence prevents that.

Why we built AsoGrove around this

AsoGrove's keyword research tool pulls competitor keywords per storefront, scores them on the local Apple Search Ads popularity scale, and outputs the four localised fields ready to paste into App Store Connect. The keyword-mapped translation pass handles the literal-translation trap by anchoring every translated keyword to a real search from a real local user, not to a dictionary.

For the screenshot side, the screenshot studio runs the caption localisation in the same workflow. One pass, five locales, ready to ship.

The localisation pass, summarised

1. Pick five locales. DE, FR, JP, KR, BR is a strong default for indie B2C apps in 2026.
2. For each, localise title (30), subtitle (30), keyword field (100), and description first 170 characters.
3. Skip dictionary translation. One hour with a native speaker per locale.
4. Localise the screenshot captions in the same pass.
5. Submit metadata-only updates per locale.
6. Refresh per locale every 90 days.

90 percent of indie devs leave 90 percent of the App Store on the table by skipping this. The 5-locale pass is the cheapest sustained ranking lift an indie launch can buy.

50 founding seats at €49/mo for life are open while Cohort 1 is filling. Claim a founding seat 🌱.