Skip to main content

​
title: Localization ROI Calculator

description: Surfaces countries where you’re getting downloads without local metadata — your highest-leverage translation opportunities. The agent reads ASC metrics for the last 30 days, computes per-country download share, and scores each country by localization ROI.

​
How the score works

For each country with ≥ 3% download share:
A country with 25% share and no localization → score 100. A country with 25% share with localization but flat conversion → flagged separately as a lagging market. We map ISO-2 country codes → expected App Store locale (MX → es-MX, JP → ja, BR → pt-BR, etc.) to decide whether you’re “localized” for that market.

​
What you see

​
Pending localization drafts (My Apps)

When Co-founder (or you) saves metadata that did not reach the live store, drafts appear in the collector as:
The reasoner should reference these under ⏰ Still waiting on you — they are real proposed work, not suggestions. Chat can inspect them with list_localizations / get_localization and publish with publish_localization (write mode). See My Apps workspace for the full bridge.

​
In the structured report

The localization array of DailyReport contains:

​
Why this matters

Indie apps default to English-only and leave money on the table when traffic naturally flows from other countries. The ROI score answers two questions: 1. Where should I localize next? → top of topOpportunities 2. Where is my localization not working? → laggingMarkets (often a sign that the translation was machine-translated badly)

​
What’s “localized”

The agent reads asc_metadata_locales (when present) to determine which locales you have App Store Connect entries for. If that table is empty, it falls back to assuming you have only en-US — which means everything else looks like an opportunity. Once you start localizing properly, the agent will recognise it on the next run.

​
Edge cases

- Country ↔ locale mapping — the mapping table covers the top 25 countries. For long-tail countries, the agent assumes “localization possible” and adds a generic note rather than computing a precise score. - Spillover — Mexican users of a Spain-localized app are counted as “localized” since es-ES content is broadly readable. This may overstate localization coverage; the lagging-markets list catches the cases where it actually hurts conversion.