localization-strategy

# Localization Strategy End-to-end localization planning for Apple platform apps — from market prioritization through translation workflow to localized App Store optimization. ## When This Skill Activates Use this skill when the user: - Wants to localize their app for new markets - Asks "which languages should I support?" - Needs help deciding market expansion priorities - Wants to plan a translation workflow - Asks about internationalization (i18n) best practices - Mentions "localization", "translation", "multi-language", or "global markets" - Wants to understand localization ROI by language - Needs cost estimates for translation - Asks about right-to-left (RTL) layout support - Wants to localize App Store metadata (title, description, keywords, screenshots) ## Process ### Phase 1: Market Prioritization Framework Not all languages are equal. Prioritize by revenue potential, not just speaker count. #### Revenue Potential by Language Based on App Store revenue data and per-user spending patterns: | Language | Relative Revenue Potential | Why | |----------|---------------------------|-----| | English | Baseline (1.0x) | Largest single market, highest per-user spend | | Japanese | 0.8-1.2x | Extremely high willingness to pay, especially for subscriptions | | Chinese (Simplified) | 0.6-0.9x | Massive volume, growing per-user spend, competitive market | | German | 0.4-0.6x | High per-user spend, values quality and privacy | | French | 0.3-0.5x | Large market across France, Canada, Africa | | Korean | 0.3-0.5x | Tech-savvy users, high app spending per capita | | Spanish | 0.3-0.4x | Huge population, lower per-user spend, growing market | | Portuguese (Brazilian) | 0.2-0.4x | Large market, mobile-first users, growing economy | **Key insight:** Japanese and German users often pay 2-3x more per user than Spanish or Portuguese users. Volume vs. value is the fundamental tradeoff. #### Category-Specific Priorities App category changes which languages to prioritize first: | App Category | Prioritize First | Why | |--------------|-----------------|-----| | Productivity / Business | Japanese, German, French | High willingness to pay for tools | | Games | Japanese, Korean, Chinese Simplified | Gaming culture, high IAP spend | | Social / Communication | Spanish, Portuguese, French | Large connected populations | | Health / Fitness | German, Japanese, Korean | Health-conscious, high-spend markets | | Education / Kids | Chinese Simplified, Japanese, Spanish | Education investment culture | | Creative Tools | Japanese, German, French | Creative professional markets | | Finance / Budgeting | German, French, Japanese | Financial discipline culture | | Developer Tools | Japanese, Chinese Simplified, Korean | Large developer populations | #### Factors to Evaluate For each potential language/market, assess: | Factor | Questions to Ask | |--------|-----------------| | Market size | How many potential users in this language? | | Willingness to pay | Do users in this market pay for apps or expect free? | | Competition level | How many localized competitors exist? Less = more opportunity | | Localization cost | How expensive is professional translation for this language? | | Cultural distance | How much adaptation beyond translation is needed? | | Support burden | Can you handle support inquiries in this language? | | App Store presence | Is the App Store well-established in this market? | Use AskUserQuestion to gather the user's app category, current markets, and budget constraints before making recommendations. ### Phase 2: Language Tier Recommendations #### Tier 1 — Highest ROI (Localize First) | Language | Key Market | Estimated Revenue Lift | |----------|-----------|----------------------| | Japanese (ja) | Japan | +15-30% (high per-user value) | | Chinese Simplified (zh-Hans) | China, Singapore | +10-25% (massive volume) | | German (de) | Germany, Austria, Switzerland | +8-15% (high per-user spend) | | French (fr) | France, Canada, Belgium, Africa | +8-15% (wide geographic reach) | | Korean (ko) | South Korea | +5-12% (tech-savvy, high spend) | **Recommendation:** Localize into 2-3 Tier 1 languages first. Expected combined revenue lift: 20-50%. #### Tier 2 — Good ROI (Localize Second) | Language | Key Market | Estimated Revenue Lift | |----------|-----------|----------------------| | Spanish (es) | Spain, Latin America, US Hispanic | +5-10% | | Portuguese (pt-BR) | Brazil | +4-8% | | Italian (it) | Italy | +3-6% | | Dutch (nl) | Netherlands, Belgium | +2-4% | | Russian (ru) | Russia, CIS countries | +3-6% | **Recommendation:** Add Tier 2 after Tier 1 languages prove ROI. Expected per-language lift: 3-8%. #### Tier 3 — Volume Play (Localize for Scale) | Language | Key Market | Notes | |----------|-----------|-------| | Arabic (ar) | Middle East, North Africa | RTL support required — higher dev cost | | Turkish (tr) | Turkey | Growing mobile market | | Hindi (hi) | India | Massive volume, very low per-user spend | | Thai (th) | Thailand | Mobile-first market | | Vietnamese (vi) | Vietnam | Fast-growing app market | | Polish (pl) | Poland | Central European hub | **Recommendation:** Tier 3 is for apps that already have strong Tier 1 and Tier 2 performance and want maximum global reach. ### Phase 3: Minimum Viable Localization Levels Not every language needs full localization on day one. Start small, measure, then invest. #### Level 0: Metadata-Only **What you localize:** - App Store title and subtitle - App Store description - App Store keywords - Screenshots (text overlays and captions) - Promotional text **What stays in English:** - All in-app UI - Help/support content - Push notification text **Cost:** $100-300 per language (professional translation of ~500 words) **Effort:** 1-2 days per language **Best for:** Testing whether a market responds before investing further **Why this works:** Users search the App Store in their language. Localized metadata means your app appears in local searches. Many users worldwide are comfortable with English UI but discover apps through local search terms. #### Level 1: Metadata + UI Strings **What you localize:** - Everything in Level 0 - All user-facing strings (Xcode String Catalogs export) - Date/time/number format adaptation - System alerts and error messages **What stays in English:** - In-app content (help articles, onboarding guides) - Marketing materials - Support communication **Cost:** $500-3,000 per language (depending on string count) **Effort:** 1-2 weeks per language (including QA) **Best for:** Languages that showed strong Level 0 results #### Level 2: Full Localization **What you localize:** - Everything in Level 1 - In-app content (onboarding, tutorials, help text) - Marketing website and materials - Customer support in-language - Legal documents (privacy policy, terms of service) - Push notifications and emails **Cost:** $3,000-10,000+ per language **Effort:** 3-6 weeks per language **Best for:** Top 1-2 markets by revenue, when the language represents >15% of revenue #### Recommended Approach ``` Step 1: Ship Level 0 for 2-3 Tier 1 languages Step 2: Measure downloads and revenue lift per language (4-6 weeks) Step 3: Invest in Level 1 for languages that showed >5% download lift Step 4: Invest in Level 2 only for markets generating >15% of total revenue ``` This approach limits risk. If Japanese Level 0 doesn't move the needle, you saved yourself $2,000+ in translation costs. ### Phase 4: Translation Workflow #### Xcode String Catalogs String Catalogs (.xcstrings) are the standard localization format. **Export workflow:** 1. Ensure all user-facing strings use `String(localized:)` or `LocalizedStringResource` 2. Build the project — Xcode auto-discovers strings 3. Open the `.xcstrings` file to see all strings with translation status 4. Export: Product > Export Localizations (creates .xcloc bundles) 5. Send .xcloc files to translators 6. Import: Product > Import Localizations **String Catalog best practices:** - Add comments to strings that need context (e.g., "Title of settings screen" or "Button label, max 15 characters") - Use string interpolation with proper grammar agreement (Foundation's automatic grammar support) - Mark strings that should NOT be translated (e.g., brand names) with "Don't Translate" - Review the "Stale" state — it flags strings whose source changed after translation #### Translation Service Options | Service Type | Cost per Word | Quality | Turnaround | Best For | |-------------|---------------|---------|------------|----------| | Professional agency | $0.08-0.15 | High | 1-2 weeks | Level 1-2, critical apps | | Freelance translators | $0.05-0.12 | Variable | 3-7 days | Level 1, budget-conscious | | AI + human review | $0.03-0.06 | Good | 1-3 days | Level 0-1, fast iteration | | Community/crowdsource | Free-$0.03 | Variable | Unpredictable | Open source, community apps | **AI + human review workflow:** 1. Export strings from Xcode 2. Run through high-quality AI translation (provide app context and glossary) 3. Native speaker reviews and corrects (1-2 hours for 500 strings) 4. Import back into Xcode 5. Test in-context (some translations don't work in UI context) #### Cost Estimates by App Size | App Size | String Count | Level 0 Cost/Language | Level 1 Cost/Language | Level 2 Cost/Language | |----------|-------------|----------------------|----------------------|----------------------| | Small (utility) | 100-500 strings | $100-200 | $500-1,000 | $2,000-4,000 | | Medium (productivity) | 500-2,000 strings | $200-400 | $1,000-3,000 | $4,000-8,000 | | Large (complex app) | 2,000-5,000+ strings | $300-600 | $2,500-6,000 | $8,000-15,000 | **Budget planning:** For a medium app localizing into 3 Tier 1 languages at Level 1: ~$3,000-9,000 total. #### Review Process Native speaker review is mandatory. **Review checklist per language:** - [ ] Grammar and spelling correct - [ ] Tone matches the app's voice (casual, professional, playful) - [ ] Technical terms translated consistently (use a glossary) - [ ] No literal translations that sound unnatural - [ ] String length fits UI constraints (check for truncation) - [ ] Culturally appropriate (no unintended meanings) - [ ] Placeholder variables render correctly (%@, %d, etc.) ### Phase 5: Cultural Adaptation Beyond Translation Translation is necessary but not sufficient. Cultural adaptation catches issues that literal translation misses. #### Date, Time, and Number Formats **Rule:** Always use Foundation formatters. Never hardcode format strings. ```swift // CORRECT - adapts to locale automatically let formatter = Date.FormatStyle() .year().month().day() Text(date.formatted(formatter)) // CORRECT - number formatting Text(price.formatted(.currency(code: "JPY"))) // WRONG - hardcoded US format Text(String(format: "%m/%d/%Y", date)) ``` | Format | US | Germany | Japan | |--------|-----|---------|-------| | Date | 02/27/2026 | 27.02.2026 | 2026/02/27 | | Time | 2:30 PM | 14:30 | 14:30 | | Number | 1,234.56 | 1.234,56 | 1,234.56 | | Currency | $9.99 | 9,99 EUR | 1,500 JPY | #### Currency Display - Always use the user's locale for display, even if your product is priced in USD - App Store handles currency conversion for IAP — but display in marketing materials should match locale - Some markets are sensitive to seeing foreign currencies (Japan especially) #### Color Meanings | Color | Western Meaning | East Asian Meaning | Middle Eastern | |-------|----------------|-------------------|----------------| | Red | Danger, error, stop | Luck, prosperity, celebration | Danger (similar to Western) | | White | Purity, clean | Mourning, death (in some contexts) | Purity, peace | | Green | Success, go, nature | Nature, health | Islam, paradise, nature | | Yellow | Caution, warning | Royalty, courage (China) | Happiness, prosperity | **Recommendation:** Use semantic system colors (`Color.red` for errors) and let the system handle meaning. Avoid using color as the ONLY way to convey information (accessibility requirement too). #### Layout Direction **RTL languages** (require layout mirroring): Arabic, Hebrew, Urdu, Persian **What to mirror:** - Navigation direction (back button on right) - Text alignment (leading becomes trailing) - Horizontal scroll direction - Progress indicators (fill from right) - Icons with directional meaning (arrows) **What NOT to mirror:** - Media playback controls (play, pause, skip) - Phone number input