Localization tactics

Create posting accounts per country (Clover supports 20; set half TikTok/half Instagram, age, gender, bio, search terms per account). Relaunch proven videos there: replace on-screen text with translated text in the native TikTok font, or sub, dub and lip-sync the creator into the local language/dialect. Best countries: Spain, Germany, Australia; secondary Italy, Brazil. Can be automated: pull niche videos over 1M views, translate, remove captions, add translated overlay, publish, with optional human approval.

Test ~500 ads per week; run the tests in countries with similar conversion rates but low CPMs (e.g. Indonesia) at ~$10/day instead of ~$1,000/day at US CPMs, then scale winners in the US.

On the Japanese country manager's advice, replaced the minimal Western best-practice paywall with a long, colorful, scrolling one-pager: the company story, founders, values, and many user quotes/social proof.

Previously paid an expensive localization SaaS plus an agency. Team kept finding ChatGPT phrasing better, so they ripped out that infrastructure and built their own GitHub Action plus a thin library that extracts/wraps all strings. Every PR sends new strings to an LLM with a per-language localization-guide prompt. Localized apps then enable localized ads.

After hearing Cal AI's founder describe scaling to $1M+/month, the founder rebuilt the same core experience (photo a meal, get calories and macros) for Israel: Hebrew language plus small changes to fit how Israelis eat and talk about food. Rule used: in a small market the niche must be huge, so he picked fitness/nutrition, a niche he knew personally. He treats brand and distribution, not product, as the real differentiator. Built with Cursor (working version in 2 weeks, live in ~2 months).

Instead of the App Store's default exchange-rate conversion of US prices, ran price tests in the top LatAm markets (Brazil, Mexico) with lower local prices.

Created an Apple Ads max-conversion campaign targeting all countries except the US, with a few top-of-funnel keywords and a $2 target cost per install. Checked impression share by country and which countries produced paying users (India: impressions, no purchases; Netherlands, Canada, UK: impressions and sales). Cross-checked with Mo's Astro-based Claude skill listing keyword volume by country: Netherlands had the same popularity (~55) as the US but difficulty ~17 vs 60-something. Then stop the worldwide campaign and create a Netherlands-only campaign with more budget.

Years ago started with the Big Mac index itself, then evolved a custom index combining purchasing power with category-specific factors (what people pay to learn an instrument or drawing). Modeled one pricing strategy for all geos, rolled it out everywhere as a test, then optimized per geo over time and revisits as the app matures and reaches new audiences.

Treat creator recruiting like talent scouting for undiscovered people who see themselves as marketers, not influencers. Sources: Handshake (student job board), Instagram and TikTok DMs, email outreach, referrals, and chance encounters: the founder asked his online Spanish tutor to make content 15 minutes into a lesson; she now makes native Spanish content for LATAM (and some English).

Split App Store ratings by country; where a country rated notably lower than the US, investigated why and fixed the underlying issues (often service quality).

Budget is mainly bottom-up: each channel is its own growth engine. For Google search (biggest channel): strong ad copy, localized to all geos, high-converting landing pages, then scale 10-20% week over week. At each step ask what the next bottleneck is: first funnel optimization, then localization, and so on. Meta handled similarly with quantity plus quality of creative.

Once apps are localized, localize all ads: dub videos with ElevenLabs' dubbing model and translate ad copy by passing it to an LLM with good prompts.