Steam Store Page Localization: Which Languages Are Worth It

Valve’s own documentation says it plainly: over 60% of Steam users run the platform in a language other than English, and “Steam favors products that are localized.” Yet most indie store pages I audit are English-only, which means most of the platform either can’t read them or never sees them. Translating your store page (the page, not the game) costs a few hundred dollars per language, and dollar for dollar it’s the cheapest visibility win I know of on Steam.

Most developers searching for this topic conflate two very different projects. Full game localization means translating every line of dialogue, UI string, and menu in your game; for a text-heavy RPG that runs $5,000-20,000+ per language. Store page localization means translating roughly 500-1,000 words of marketing copy: your short description, long description, and feature bullets. That typically costs $100-300 per language.

The two also do different jobs. Translated page text improves conversion once someone lands on your page. The supported-languages checkboxes in Steamworks (Interface / Subtitles / Full Audio) are a separate system that controls whether Steam shows your game to a language cohort at all. More on that in the visibility section, because mixing these two up is where most wasted budget comes from.

The case for translating the page even if your game stays English-only is strong. Chris Zukowski surveyed 100+ developers, sampled quiet non-promotional “resting” weeks, and found that games with store pages localized beyond English earned a median of 62 wishlists per week against 18 for English-only pages, about 3.4x. The averages were 529 versus 124. Zukowski is upfront that this is correlational (teams with budget to translate also have budget for everything else), and I’d repeat that caution. Still, the direction matches everything Valve and publishers have shared. Treat localization as a multiplier on the wishlist traffic you already generate, not as a traffic source by itself.

At GDC 2025, Valve presented full-year 2024 account-language data, and it contained a first: Simplified Chinese passed English as the most-used language on Steam, 33.7% to 33.5%. Russian was third at 8.2%, and the top four languages cover roughly 80% of Steam accounts.

The monthly Steam Hardware & Software Survey tells a slightly different story. The January 2026 snapshot put English at 37.01% and Simplified Chinese at 23.86%, with Russian at 10.12%. The survey samples whoever opts in during a given month, so it swings hard; Valve’s full-year account data is the more reliable basis for budget decisions. Here are both, side by side:

Whichever dataset you trust more, the conclusion is identical: an English-only page is unreadable or invisible to roughly two thirds of Steam.

Valve’s guidance, stated in the Steamworks docs and paraphrased by Zukowski: translate Japanese, Korean, and Chinese first, because French, Italian, German, and Spanish speakers are far more likely to also read English. The docs phrase it memorably: “the less a language looks like English, the higher the expected return.”

The audience data backs that up. A pre-release survey for Slay the Princess, a text-heavy visual novel, asked players in each country whether they’d play in English only:

Two thirds of Brazilian players and 61% of German players will tolerate English. In China, Japan, and Korea, an untranslated game loses two thirds to three quarters of its potential audience before they’ve read a word of your description.

Publisher numbers point the same way. Hooded Horse attributed wishlist lifts of roughly 35-40% to French/German localization announcements, over 50% for Spanish, and over 90% for East Asian languages. Fellow Traveller publishing director Chris Wright runs this priority order: Simplified Chinese first, then Japanese, Traditional Chinese (if you’re also shipping on Switch), Russian, and Brazilian Portuguese. One publisher’s catalog went from 5% or less of units sold in China to 20% or more after adding Chinese localization plus local PR.

My recommended order for a typical indie page budget:

Two notes on that table. Japanese looks small on paper (1.7% in Valve’s 2024 data, 2.86% in the January 2026 survey) but has disproportionately high spending power per player, so for premium-priced games its ROI rivals much larger markets. And if you can stretch to ten languages, GameDiscoverCo called Cobalt Core’s set “pretty much perfect” for an indie: English, French, German, Spanish (Spain), Japanese, Korean, Brazilian Portuguese, Russian, Simplified Chinese, and Traditional Chinese.

Steam supports 110 languages on paper, but only 31 of them have full platform support: a client UI and a storefront users can browse in. The other 79 are game-support-only. For store page translation, only those 31 matter, because they’re the only ones with shoppers behind them.

Two separate mechanisms decide what a user browsing in Chinese or Korean actually sees, and you need both working. (This is one narrow corner of how Steam’s algorithm works; I’m only covering language matching here.)

1. 01The supported-language checkboxes drive eligibility. If a language isn’t ticked for your game, users browsing in that language see far less of it across discovery surfaces like the Discovery Queue and tag browsing. GameDiscoverCo estimated in 2021 that a Chinese translation effectively cut the competition a game faced in the Chinese-browsing storefront from ~34,000 Steam games to ~6,000; the rest weren’t available to that cohort.
2. 02Translated page text drives conversion. Once your page is shown, a native-language description does the selling. A localized application name also overrides your default name in search suggestions for that language, free search placement most developers never set.

Eligibility gets you seen; translation gets you wishlisted. You want both.

If your game is text-light (an action roguelike, a puzzle game), full localization can cost as little as $500-2,000 per language; at that price, just do it. The discovery evidence is lopsided. Against The Storm shipped with 17 supported languages and Asian countries (China, Japan, Korea, Taiwan) accounted for at least 32% of its sales. Nebulous: Fleet Command launched English-only and got 2% of sales from China, despite fan translations and real interest from Asian streamers.

If your game is text-heavy and full localization is out of reach right now, Zukowski’s recommendation stands: translate the store page into every full-support language you can afford anyway. An untranslated page misses views and wishlists from shoppers who browse exclusively in their native language, and every regional wishlist is an argument for funding full localization later. Just be honest on the page about what the game itself supports.

Localized text and localized prices are separate levers; pairing them compounds the effect, and the regional pricing guide covers the second one.

The mechanics are simpler than most developers expect. Steamworks’ localization documentation covers the details, but here’s the workflow:

1. 01Export your source text. Short description, long description, and feature bullets. If your description is in good shape, this is 500-1,000 words. Translate the final version, not a draft you plan to rewrite.
2. 02Hire a human translator per language. Ideally a native speaker who plays games. Marketing tone survives machine translation badly.
3. 03Paste translations via the language dropdown. Every text editor in the Steamworks partner portal has a language dropdown in the top right. Switch it to the target language, paste the translated text, repeat per field.
4. 04Set a localized application name where it helps. A localized name overrides the default and is used in search suggestions for that language.
5. 05Upload localized screenshots if your game UI is translated. These use filename suffixes, e.g. foo_japanese.jpg. Skip this if your game is English-only; don’t show UI the player won’t get.
6. 06Tick the supported-language checkboxes that are true. Only the true ones. More on that below.
7. 07Publish and spot-check. Switch your own store language preview and confirm every field renders, including the short description that shows in search results.

Do this when your Coming Soon page goes live, not the week before launch. Wishlists from regional storefronts accumulate for months, and a Coming Soon page that’s readable in ten languages collects from all of them the entire time.

For the page: $100-300 per language, so a ten-language set lands around $1,000-3,000. For the game: roughly $500-2,000 per language for text-light games, $5,000-20,000+ for text-heavy games like RPGs, and professional multi-language localization of very text-heavy games can reach six figures.

On returns, be realistic. The 3.4x median from Zukowski’s survey is a correlation across many games, not a promise for yours. Localization multiplies the traffic you already have; if your page gets thirty visits a week, a translated page gets you a bigger slice of a small pie. The honest framing: for one to three thousand dollars you remove a language barrier between your page and roughly two thirds of the platform, and every documented dataset says that barrier is expensive to keep.

Ticking language checkboxes the game doesn’t support. Players who buy expecting Korean subtitles and find none will refund and review accordingly. The checkboxes are a promise about the game, not the page.

Shipping raw machine translation for Chinese, Japanese, or Korean. Players spot it instantly, and a sloppy translation reads as a sloppy game. Machine-assisted with native review is fine; machine-only for CJK languages is a false economy.

Translating into languages without a storefront. Only the 31 full-support languages have browsing shoppers. Paying to translate your page into a game-support-only language buys you nothing on the store.

Starting with French, Italian, German, and Spanish because that’s the traditional localization set. On Steam that order is backwards. Those audiences mostly read English; Chinese, Japanese, and Korean audiences mostly don’t.

Letting translations go stale. You’ll rewrite your English description three times before launch. Budget for one final retranslation pass after the English copy is locked, or your Japanese page will describe a game that no longer exists.

Deciding off one month of survey data. See the February anomaly above. Use Valve’s full-year account data.

Steamworks already gives you the instruments. Before you publish the translations, record two or three quiet weeks of baseline data: store page visits and wishlist additions by country or region (both are in your Steamworks reports). Then publish all your translations at once and compare the same regional numbers over the following non-promotional weeks. Quiet weeks matter — a festival or a streamer spike will swamp the signal, which is exactly why Zukowski sampled resting weeks in his survey.

What you’re looking for is regional mix shift. If Chinese and Korean page views and wishlist additions grow while your English-speaking regions stay flat, the translation is doing its job. After launch, the regional sales report answers the only question that matters: what share of revenue now comes from the languages you paid to add? Against The Storm’s 32%+ from Asia is the ceiling, not the median, but if localized regions are still rounding errors after a few months, check that the localized name, text, and language ticks are actually live before concluding the market isn’t there.

Should I translate my Steam page if my game is English-only?

Yes. This is Zukowski’s explicit recommendation: translate the page into every full-support language you can afford even if the game itself stays English-only, because untranslated pages miss views and wishlists from shoppers who browse only in their native language. Keep the supported-language checkboxes honest.

If I can only afford one language, which one?

Simplified Chinese. It was the most-used language on Steam in Valve’s full-year 2024 data (33.7%, ahead of English at 33.5%), English tolerance among Chinese players is low (32% in the Slay the Princess survey), and East Asian localization announcements drove the biggest wishlist lifts (over 90%) in Hooded Horse’s analysis.

Do screenshots and trailers need localizing too?

Steamworks supports localized text, images, and trailers, with localized screenshots uploaded via filename suffixes like foo_japanese.jpg. Prioritize text first; localized screenshots only make sense once the game’s own UI exists in that language.

How much does store page localization cost?

Typically $100-300 per language for professional translation of page copy. A sensible three-language starter set (Simplified Chinese, Japanese, Korean) costs less than most devs spend on a trailer edit, and a full ten-language set usually stays under $3,000.

Before you commission translations, make sure the English page itself is worth multiplying — run it through the store page checklist first, since every weakness in the source copy gets translated ten times. Then start with the first tier (Simplified Chinese, Japanese, Korean), publish, and watch your regional wishlist numbers for the next month.