Which languages to translate your Steam page into first

A translation is worth most where players already want the game but cannot read about it. The obvious way to choose is a ranking of the world's biggest gaming languages. The better way is to start from your own data: Steamworks shows store page traffic and wishlists by country, and a translation aimed at a region already sending you visitors pays back faster than one aimed at a region that has never heard of the game.

Then check that against two outside sources: what Steam's players speak, and what the games most like yours support. Where all three agree, translate first.

Valve's monthly Steam Hardware & Software Survey includes the language each user runs Steam in. For September 2026, the languages above 1%:

Valve's localization guidance puts it in one line: the less a language looks like English, the higher the expected return. A player who reads Spanish or German may manage an English store page; a player who reads Simplified Chinese, Japanese or Korean is much less likely to. That, and the size of the Simplified Chinese audience, is why it so often comes first for indie games.

Valve also suggests prioritising by traffic, sales and community feedback, and using a translated store page as a test before translating the game: put the page up in a language, then watch regional wishlists to see whether players there respond.

1. Pull your own numbers: store page visits and wishlists by country for the last three months, from Steamworks.
2. List what your comparable games support: the languages table on the store pages of five to ten games like yours.
3. Score each candidate language: high if your traffic, the comparables and the survey all point to it; lower if only one does.
4. Weight by distance from English: a language with a different script moves up.
5. Weight by cost: a text-heavy game (narrative, strategy, RPG) costs more per language to translate than an action game with little text; a store page costs the same for both.
6. Translate the store page in the top two or three, measure regional wishlists for a few weeks, then decide which earn the game itself.

For a typical indie game with no strong regional signal of its own, a common sequence is:

• First: Simplified Chinese, store page then interface. The largest non-English audience and the furthest from English.
• Next: the two or three of Russian, Spanish, Brazilian Portuguese, German, Japanese and French that your data favours.
• Then: Korean, Polish, Traditional Chinese and Turkish, as traffic justifies.

Within each language: the store page first, then the game's interface, then subtitles, then full audio if the game has voice. Release each as an update with an announcement in that language, so the players who were waiting find out.

Your Steamworks reports answer the question better than any survey once the page has been up for a few weeks:

• Wishlists by country show where interest already exists. A country sending more wishlists than its share of Steam suggests is a market that likes your genre.
• Traffic by country shows where players arrive but do not wishlist. High visits with a low wishlist rate from a non-English region often means the page is not being understood.
• Community posts and reviews in other languages, on your demo or playtest, are players telling you directly.

Compare a few weeks before and after a page translation, with no sale or festival in either window, and the effect of the translation is usually clear.

If the store page test says a language deserves the game too, the work is cheaper when the project is ready for it. Godot's localization system reads translations from CSV or gettext PO files through its TranslationServer; text shown through tr(), and Control nodes with auto-translate on, switch language at runtime. Three habits save rework:

• Keep every player-facing string out of scripts and scenes as literals: use translation keys from the start, even before you translate anything.
• Pick fonts with the glyphs you need. Simplified Chinese, Japanese and Korean need a font that covers them; a missing glyph renders as a box. Godot lets a font fall back to another font for missing characters.
• Leave room in the UI. German and Russian strings run longer than English, and CJK text needs a larger size to stay legible.

A big language is not automatically your market. Genres travel unevenly: some sell disproportionately in particular regions, and a game's theme can matter more than population. Price matters too, since regions with lower regional prices bring more players per translation but less revenue per player. And a translated store page for a game that is English-only should say so honestly in that language; otherwise the page that brought players in becomes the reason they refund. Treat the survey as a starting point and your own wishlists as the verdict.

Flockbay reads your Steam followers and an estimated wishlist range for free, and your exact wishlists from your own Steamworks key, which never leaves your machine. Because the game it builds is an ordinary Godot project, you can ask it to keep player-facing text in translation keys from the start, so later languages cost less. Its Steam page drafts the store page in English from your game, with textless images that serve every language, and exports it as a .zip you upload yourself. Preparing the page by hand and its .zip are free; drafting is free in beta until 1 December 2026.