The problem
LoveMemo is an app for romantic surprises: a page with photos, a message, music and a live counter of time together, delivered by link or QR code. It always did well in Brazil, especially on Android. But iOS users pay better, and better still outside Brazil. He asked Claude Code to research how to grow organic reach in other countries and which languages were worth it, then applied the plan across the app, the website and both store listings.
The blueprint (as he describes it)
- Read the sales report before translating anything. Apple’s September financial report showed about three quarters of iOS sales already came from outside Brazil. Mexico, Spain, Chile, Peru and Honduras added up to “about 36% of revenue”, and there were English sales with zero ratings on the US store. His summary: “People just weren’t finding the app.”
- Size markets from public data. Store search volume is not public, so Claude checked who ranks at the top of a search and how many ratings they have, country by country, through the public iTunes search API. A single couple counter had 325,000 ratings in Germany alone.
- Fix the positioning, not just the language. Outside Brazil, people search for relationship counters and couple widgets, not surprises. The English name changed from “LoveMemo - Surprise your love” to “LoveMemo: Couple Days Counter”, and weak keywords were swapped for terms like together, anniversary and tracker.
- Add listings that carry extra keywords. He added es-MX because third-party tests by AppTweak and Mobile Action say the US store also indexes Mexican Spanish keywords, which adds another 100 characters of keywords. He notes this comes from third-party tests, not Apple’s documentation. An en-GB listing brings in a different keyword set for the UK and Australia.
- Let the agent ship the version. On iOS, names, keywords and new languages only change with a new version, so Claude created version 1.6.1 in App Store Connect with the asc CLI, including release notes in every language. “All I did was upload the build.”
- Keep listings as code. All listing copy lives in the repository as JSON and in fastlane’s format and is pushed from the command line, so it can be reviewed in a diff.
- Translate everything, then ship. German went into the app’s ~550 strings, the Android widget, dates, the website and the page the recipient opens from the QR code. The AI translation shipped without a native-speaker review, a deliberate call so he could fix it from real user feedback.
- Generate screenshots with a script. Claude took captures in the iOS simulator and a headless browser, and a script composes the final images per store, with each language’s captions and frame order from a JSON file. When the app was redesigned two days later, every screenshot in four languages and two stores was redone in one afternoon.
Stated results
- Before the work: about 36% of revenue from Spanish-speaking countries with a single listing and no ASO work; about three quarters of iOS sales from outside Brazil.
- After: the Play Store listings went live on the 19th, and Android, which used to sell almost only in Brazil, started selling in other countries. The iOS changes shipped in versions 1.6.1 and 1.7.0, and he says the effect is just getting started.
- Workload: keeping listings, keywords and screenshots as code turned redoing everything in four languages from a week of work into an afternoon, by his account.
- Downloads, revenue change or rankings: None stated.
Caveats
- Results are early and given without numbers. He says some of the bets probably won’t pay off and that he will report hits and misses later.
- Claude Code is central. This is today’s Claude-centric piece.
- The German translation was not reviewed by a native speaker.
- The es-MX indexing tactic rests on third-party tests, as he notes.
- He sells freelance development work from the same site. His contact details are on his page and are not reproduced here.
- Handles: X @jxnata, LinkedIn jxnata and GitHub jxnata, all linked from jxnata.dev.
FAQ
What did Claude Code do for LoveMemo?
It researched markets through the public iTunes search API, wrote the listing copy and translations, built the scripts, created version 1.6.1 in App Store Connect and took the screenshot captures. He kept the decisions.
Why didn’t he just translate the app?
The sales data showed the bigger problem was English, which already existed and did not rank. Abroad, people search for couple counters and widgets, not surprises.
What has changed so far?
Android started selling outside Brazil after the Play Store listings went live on the 19th. He gives no figures yet, and says the iOS effect is just starting.
Which decisions stayed with him?
Leaving Portuguese as it was, redoing the bland screenshots and shipping German without waiting for a review. He writes that the direction “was a conversation”.
Sources
- https://jxnata.dev/en/blog/taking-lovememo-international/ (primary: “Taking LoveMemo beyond Brazil: research, ASO and localization”, Oct 22, 2026)
- https://jxnata.dev/en/ (his site: background, projects, profile links)
Want agents like these working your distribution? Get It Built. Get new teardowns weekly: subscribe (double opt-in).
Get one teardown a week.
Same format, a new operator every issue. Double opt-in.