For years, the recipe for taking a game global was simple: build a solid game on a proven formula, translate all the text, ship it everywhere. It worked—for a while. Developed, English-speaking markets drove so much success, with the app stores doing part of the discovery work, that a lot of studios never had to think much harder than “translate and launch.”
That era is over. There are now far too many games chasing the same players, and a well-translated game that plays like a copy of something familiar doesn’t stand out. As Artur Davydenko — Product Lead at Meta, previously at Google and in venture capital — put it during our recent LinkedIn Live, leaning on that old playbook has stopped working:
Just copy the big American game, translate it all, use the simple ideas — it’s just not enough nowadays. It will not fly. And if it would fly, those big companies would launch it themselves.
So if translation alone won’t carry a game into a new market, what will? The honest answer is: it depends on your game.
Localization isn’t one decision—it’s a series of them. And one of the biggest is how much of the game you need to localize for the experience, and the market test behind it, to actually work.
Translation and localization aren’t the same thing
The first thing to get straight is that translating a game and localizing it are different jobs. For some products, plain translation is genuinely enough. A news article, for example, primarily needs to convey information. A slightly awkward sentence may still do that job.
A game isn’t just information. It’s an experience. Players need to understand what’s happening, connect with characters, follow progression, respond to humor, navigate the interface, and—depending on the game—spend hours inside its story and world. See also localization vs. translation.
As Artur put it:
In entertainment and gaming, translation is absolutely not enough, because it's a story. You can skip a story and find another one in a second.
Adapt to the Culture or Skip the Market
Why entering an entertainment market seriously means thinking beyond transferring the words.
This is also why localizing half-heartedly—doing it just to tick a box, without really committing to what players in that market need—can waste money. The choice worth making consciously is whether you’re ready to localize because you genuinely want to grow into a region, or whether it makes more sense to hold off on deeper product localization until you are.
Bringing Artur’s market-entry observations together with what we see across game localization projects at Alconost, there are three useful scenarios to think about.
1. Casual games: localize the marketing, not necessarily the game
If your game is casual—intuitive enough that players can skip the pop-ups and windows and simply start playing—you may be able to test a new market without immediately localizing the entire product.
It works really well in casual games, which are intuitive enough from a UX and mechanics point of view. People skip all the windows and pop-ups anyway, and they just play.
In this scenario, a studio can begin with the player-acquisition layer:
- Marketing creatives
- Ads
- Store-page copy
- Store-page visuals and metadata
while initially keeping the game itself in its original language. That can be enough to answer an important early question: Can we attract players in this market?
It is a lighter-weight way to test demand before committing resources to full product localization. For a genuinely intuitive game, the first sessions may still give you useful acquisition and engagement signals even if the whole experience has not yet been localized.
The important limitation is that this tests interest and early behavior, not necessarily whether the game can retain players long-term.
Outcome: you can test market demand before committing to full localization
This approach can help a studio evaluate:
- Click-through rates
- Store conversion
- Cost of acquisition
- Installs
- Early-session engagement
If those signals look promising, that gives you evidence for deeper localization investment.
2. Retention-driven games: localize what players come back to
The picture changes once you care about retention—about what happens after day zero. Everything a returning player actually reads or interacts with may now matter: features, menus, progression, events, offers, tutorials, character interactions, and narrative.
When you think outside the first couple of sessions and day zero, and you want retention — features, windows, narrative — all of these are highly affected by language and culture and the context of where people live and why they play.
This is where localized marketing stops being enough. You may have successfully convinced someone to install the game, but now the product itself has to keep them there.
From a localization perspective, this is an important distinction: Marketing localization gets the player through the door. Product localization helps them stay.
If the experience after the first sessions remains inaccessible or culturally distant, weak retention can also become difficult to interpret. Did players dislike the game? Or did they never get the intended experience because important features and content weren’t localized?
Outcome: you can test whether the actual product retains players
For games where the business depends on repeat engagement, localization needs to cover the parts of the experience responsible for that engagement. That could mean:
- UI
- Progression systems
- Live-ops events
- Offers
- Tutorials
- Player communications
- Narrative content
The exact scope depends on the game. The principle is simpler: localize enough of the returning-player experience for your retention data to mean something.
3. Narrative and mid-core games: localize early
For narrative-driven or otherwise language-dependent games, localization needs to enter much earlier. The more players need to understand dialogue, story, character relationships, humor, quest objectives, tutorials, and complex mechanics, the less meaningful an English-only market test becomes for players who don’t comfortably read English.
This is especially clear in narrative games, where the story is part of the product. Localization cannot simply be bolted on after the team has already tried to validate whether target-market players enjoy the experience. If players cannot properly experience the game, you are not really testing the finished proposition—you are testing a reduced version of it.
The same logic can apply to mid-core games when systems, progression, narrative, and retention mechanics depend heavily on text and context.
Outcome: you test the game players are actually supposed to experience
This is the key difference. A casual game may be able to test acquisition before full product localization. A narrative game often cannot test its central appeal without localization because so much of that appeal exists in language.
That means localization should be considered earlier in production planning, market testing, content pipelines, UI design, and narrative workflows—rather than appearing only as a pre-launch translation task. For readiness and process, see the game localization workflow.
Localization can also mean adapting more than the text
Even complete linguistic localization does not automatically mean a game will feel right in every market. Depending on the audience and game, adaptation may touch art style, animation, text presentation, pacing, small mechanics, and overall rhythm. Cultural traps in presentation are easy to miss if you only ship strings—see cultural traps in localization.
Artur points to Brawl Stars as an example of how far adaptation can extend.
Even Supercell Took Years to Adapt
Artur discusses why adapting a game to its audience can involve much more than translated text.
His point is not that every studio needs to remake its game for every market. It is that players experience the whole product, not its translation layer in isolation. The way you present mechanics, visuals, animation, text, and pacing can influence whether the game actually connects with the audience.
For studios entering culturally distinct markets, that means localization questions may sometimes expose broader product questions too.
So when does localization actually happen?
Knowing how much to localize is half the answer. The other half is when. Here, Artur lays out a particularly actionable sequence:
- Start in your home market — if it’s big enough to give you real qualitative feedback on what players like and don’t. You may not earn much yet, but you learn.
- By soft launch — usually a handful of countries across your target regions — you need actual monetary data, and you should already have your localization strategy ready. The end of soft launch is when you scale hard, so the strategy has to exist before then.
- Localize marketing fast, close to launch — it’s quicker to turn around than the product, so roughly two to four weeks ahead of a market launch can be enough.
Localize During Your Soft Launch
Artur’s practical sequence for moving from home-market learning to soft launch and international scaling.
There is an important distinction here: Your localization strategy should exist early. Individual localized assets do not necessarily need to.
Marketing creatives can often be produced relatively quickly. Product localization—especially when the game contains substantial narrative, complex systems, voice-over, or ongoing content—needs more preparation. And if the game cannot be meaningfully experienced without the target language, localization should enter even earlier than the general soft-launch sequence.
Match localization depth to what you’re trying to learn
This is the part where Artur’s framework becomes particularly useful when combined with Alconost’s localization perspective. Rather than asking only “Should we localize this language?”, ask: What are we trying to learn from this market, and what does the player need localized for that test to be valid?
If you’re testing acquisition
Localized marketing and store presence may be enough.
If you’re testing first-session gameplay
You may need essential UI, tutorial, and gameplay localization.
If you’re testing retention
The features players repeatedly interact with need to be localized.
If you’re testing narrative appeal
The narrative needs to be localized.
That gives you a much more practical way to scope localization investment.
Localization depth is a spectrum
Based on those principles—and on how localization projects tend to evolve in practice—it helps to think about localization in layers rather than as a single yes/no decision.
Layer 1: Marketing
- Ads
- Video creatives
- Social content
- Campaign copy
Layer 2: Store presence
- Store description
- Screenshots
- Promotional text
- Metadata
Layer 3: Core UX
- Navigation
- Tutorials
- Essential gameplay strings
- Key system messages
Layer 4: Retention content
- Progression
- Events
- Live ops
- Offers
- Feature explanations
Layer 5: Full player experience
- Narrative
- Dialogue
- Humor
- Character voice
- Worldbuilding
- Voice-over, where relevant
Not every game needs every layer in every market immediately. The question is which layer is necessary now—and which one you need to be ready to add if the market performs. Indie Steam titles often face a related trade-off between breadth of languages and depth of adaptation; see Steam localization for indies.
Don’t let under-localization invalidate the experiment
There is one risk worth watching closely. Imagine that a studio:
- Localizes ads for a new market.
- Gets strong acquisition.
- Sends players into an untranslated product.
- Sees weak retention.
- Concludes that the market is not promising.
That conclusion may be correct. But it may also be a testing problem. If localization materially affects the player experience, then an under-localized product can distort the result.
The studio may think it tested “Do players here like our game?” when it actually tested “Do players here like our game when parts of it are difficult to understand?” That is not the same hypothesis. The more language-dependent the game is, the more important this distinction becomes.
A practical four-question framework
Bringing Artur’s market-entry logic together with Alconost’s localization experience, we can reduce the decision to four questions.
1. Can players understand the core game without text?
If yes, acquisition-first testing may be possible. If no, product localization needs to happen earlier.
2. What are you measuring?
If you only need acquisition data, localized marketing may be enough. If you need retention, monetization, or deeper engagement data, more of the actual product needs localization.
3. How important is narrative and context?
The more value sits in story, characters, humor, complex mechanics, or worldbuilding, the earlier localization should enter the project.
4. What happens if this market succeeds?
If a successful soft launch would trigger rapid scaling, the localization infrastructure needs to be prepared before the results arrive. That can include:
- Target languages
- Localization workflow
- Context for translators
- Terminology and glossaries
- Update processes
- Multilingual QA
You do not necessarily need to translate everything in advance. But you do need to know how you will scale if the answer from the market is “yes.”
The takeaway
There is no universal rule for how far to localize a game—and that’s the point. A casual puzzle game and a narrative RPG are different products with different player experiences.
For an intuitive casual title, localized marketing may be enough to test whether a new audience is interested before committing to full product localization. Once retention becomes the question, more of the product needs to speak the player’s language. And for narrative-driven or otherwise language-dependent games, localization may need to happen from the earliest meaningful market test because language is part of the product itself.
Treating all of those games the same way is how studios either overspend on localization they did not yet need—or underinvest in the localization that could have made their market test meaningful.
The better question is not simply “Should we localize this market?” It is: How much needs to be localized for the result we’re measuring to actually mean something?
Match localization depth to the game, the KPI, and the stage of market entry—and localization becomes much easier to treat as part of the growth strategy rather than a box to check before launch.
For the bigger picture on testing, community, funding, marketing, and global expansion, see How to Grow a Game in 2026: What Actually Works Now.
Planning your next game market?
Alconost can support localization at different stages of market entry—from store and marketing content to full game localization, ongoing updates, contextual workflows, and multilingual QA.
Artur Davydenko is Product Lead at Meta (Privacy & Compliance / GTM), previously Sr. Product Marketing Manager at Meta. Over 10+ years in gaming, tech, and investment he has worked across marketing, growth, people ops, bizdev, project management, and entrepreneurship—including raising $1M to launch casual games and advising a LiveOps-focused app that raised $700K. Before Meta he worked at Google (Ads and Firebase), as Entrepreneur in Residence at Flyer One Ventures, and in growth roles at Playgendary and game studios including Alpha Games and Kefir. He co-founded HoneyLab Games and advises Balancy and HookRadar.
