Today, we’re going to talk about how to localize a game properly without using Google Translate, whether it’s possible to balance business imperatives and creativity, and the sorts of cultural values and norms you need to consider when entering a new market.
When Should a Publisher or Developer Start Thinking about Localization?
Imagine you’re a European studio who’s created a game in your home country. In this case, the answer’s clear: as soon as you want to expand outside your country.
But what if you created the game in both your country’s language and English? When would it be time to take the next step? That depends on what your company and its internal processes are like. Localization tends to be an afterthought. When the game’s almost done and the money’s running out, people suddenly think, “Hey, people don’t just play games in my language and English—there are other languages too!” But here’s the thing: if you’re planning to expand into other markets, the sooner you start thinking about localization, the better.
Take Arabic or Hebrew, for example, which are written right to left. That’s going to require another, very different interface layout. You’ll need to revisit the menu, buttons, notifications, any scrolling text—essentially every element of the interface, since the text has to run right to left. You’ll also need to consider your choice of fonts and the complex interaction between fonts and numbers: words might run right to left, but numbers run left to right. When things are this complex, it pays to think them through early.
And let’s not forget Japanese, Chinese, and Korean characters. For example, in Korean, predicate goes at the end of the sentence, and that creates another restriction for the interface. You can’t just chop the text without losing its meaning; it might cause confusion. That’s why localization should be planned in advance.
If you’ve got a complex engine or intricate game mechanics with lots of random generation of items and tasks and characters of different genders, you’ll likely encounter issues translating your game into any language other than English. You should be sure to consult a specialist about this early on.
Proactive Localization and Cultural Norms
Imagine you’ve made a game in Spanish, translated it into English, and want to release it in other markets. You’ll want to analyze how profitable that will be for you, and thanks to our many years of experience in the gaming market, we can advise you on that. We can also help you take different countries’ cultural norms into account.
Here’s an example. We once worked on a project involving a funny character called Goose. Everything he said in the game was randomly generated, and most of his speech was memes that every Russian player would know. So when we were translating the game into Brazilian Portuguese, French, German, Korean, and so on, we had to find equally funny equivalents in each language for everything Goose said. That’s the job of translators and editors: they find equivalents and adapt texts to different cultures. This kind of thing is our bread and butter.
We should note that localization can be proactive or reactive. Reactive localization usually has to do with avoiding cultural taboos, whereas proactive localization is more about adaptation. The Goose example would be proactive localization, whereas reactive localization aims simply to satisfy the demands of audiences and platforms.
You will sometimes have to redo something from scratch to get around different restrictions. For example, in the Middle East, you can’t represent gods or divinities in images, as religious norms prevent this. If your game has angels and demons, you will have to rethink them, perhaps turning them into good and bad robots, for example. And if the game is called Goddess Wars, you’ll have to change it to something else, say Hero Wars.
There are some specifics that wouldn’t occur to anyone until they start working on localizing a game. Take place names—even they can be different in different countries. Say our game features the Sea of Japan. Korean people would say, “What is this ‘Sea of Japan’? That’s the East Sea!” So here too, your approach should be tailored to each of your different audiences.
In China, for instance, images of violent, bloody murders are forbidden, so when a character is killed, a sad face appears above them, or else they just disappear. There are lots of specifics like this.
That’s why you need to start thinking about localization early, not once your game is already released and you’re entering a new market.
The Technical Side of Localization
We used memoQ, a server-based solution that we believe is more reliable in terms of information security than online options. It also has more functionality than many alternative products, as well as flexible license management. We have memoQ licenses that we distribute between translation teams. Meanwhile, projects’ term bases and translation memory are kept on our server because they are our intellectual property.
If you’re a first-time indie developer and want to localize your game but can’t stretch to hiring a studio, there are alternatives. For example, Crowdin lets you automate the localization process by uploading translations straight into the game’s code. So you translate your text, upload it, test it, and that’s it—you’re done. If your game has a loyal community, they can help you with the first stage. Just bear in mind that this approach only works if your community is mature enough, and you’ll still need to check the quality of the translation you get.
We should also say a few words about machine translation. Machine translation is a tool to be used with caution. Generative AI, for example, can do translations, but there are often restrictions on the text’s length, and you’ll start to get errors if you use it for too long. It’s like trying to cut wood by smashing it with a hammer: it’s possible, but the result will leave a lot to be desired.
To translate interfaces, the best option is to use specialized solutions such as DeepL. Our tests have found it to be stable, and it also lets you use a glossary to ensure quality.
A glossary is a dictionary where every term in one language strictly corresponds to a term in another language. For example, in a game about bandits, the Russian word dengi (“money”) might be translated as “dough.” If you go with that translation, you should use it throughout the game. Using a glossary makes a big difference to localization quality, as it keeps your terminology consistent, which will make your game come across better.
What If You’re a First-Time Indie Developer or a Small Team?
You’re pushed for time, but you want to do things well. We know how protective creators are of their games, and we also understand how important high-quality localization is in ensuring that your game is understood and well received in the country you’re planning to promote it in. So you have a choice: try getting your community to help localize the game, even though it’s not that popular yet; use machine translation and then check it carefully; or turn to a professional studio that can oversee the whole process and ensure the result is up to standard. Which option is going to be the safest for you?
A lot depends on your game: how much text it has, what the context is, and what languages it’s going to be in. For example, machine translation might struggle with rare languages like Uzbek, but depending on the translation engine, it might do slightly better with Farsi. Some language pairs produce acceptable translations, whereas others have real issues that will leave your users unimpressed.
It’s also important to consider your audience: how fired up are they? Do they want a translation at all? Sourcing localization through your community is quite an unpredictable option, and can be tricky to manage.
Whatever option you choose, the whole process should be overseen by a specialist who knows their stuff. If you have someone like that, you’ll be fine.
If you’re planning to enter new markets and want to know more about localization, just get in touch.
We’ll be happy to advise you and answer any questions you might have.
