Localization is the single biggest thing any game can do to expand its player base and boost its revenue.
But it’s also a hefty investment that comes with scary risks.
Because taken at face value, localization seems like a simple rip-and-replace task. But under the surface, it’s a complex beast that demands collaboration between translators, editors, testers, marketers, developers, and even regional ratings boards. And, at times, this complex web of stakeholders and systems is overwhelming (especially if you’re new to it). And when the success and revenue of your game relies on the effectiveness of your localization, you want to know that investment is worth it.
So, here it is: every question (with answers) you need to ask before you invest in localization.
We got a bit excited at the chance to write about video game localization, and in our excitement this guide became quite chunky.
So, to combat all the chunkiness, here’s a quick navigation. Each section stands on its own, so there’s no pressure to read the whole thing cover to cover (but if you do, thanks).
In short, localization prepares a game for release in a new country or region. On the surface, this might sound like a simple translation task, but dig deeper and you’ll discover it’s so much more:
The simple answer is that translation is one part of the wider localization process. Translation is the part that turns a game from one language into another.
The entire localization process consists of mission-critical parts, each vital to successful translation. You can think of the process like this:
This depends on the size of the game and the quality of the localization you’re looking for. But a good rule of thumb is:
The more words in your game and the more languages you’re translating into, the longer it’s likely to take.
*It’s also worth noting that any changes during localization will add more time to the process.
Ever heard the phrase “All your base are belong to us”? (If not, it’s worth a quick Google.) Sure, that’s a translation disaster, but poor localization is even more disastrous. Get a cultural reference wrong and you might offend your entire market. Or worse, violate compliance standards and your game might not even be published in your target market. Bad localization is financially and reputationally damaging.
Localization helps you reach more of your game’s ideal audience. And the more of your ideal audience you can reach, the more exposure, popularity, and revenue your game will generate. So if you’re looking to maximize the success of your game, then yes, you’ll need to localize it.
Often localization teams group languages such as EFIGS (English, French, Italian, German, Spanish) and CJK (Chinese, Japanese, Korean) to hit major gaming markets. These groupings are a useful guide, but your ideal market depends on the nature and audience of your game.
Finding the perfect market is a tricky proposition for any game, and especially in another language. Existing data from similar games released in particular markets will give a steer on which genres and styles are more or less likely to succeed in certain territories. Here’s a starting point for the things you should look at:
This depends on the style of game. If your game is an online first-person shooter without a story, often there’s no real need to translate the game’s audio. But if it’s a story-heavy RPG that relies on dialogue, the improvement to a player’s experience that audio translation adds is worth the investment.
If you’ve already honed in on the languages and regions you’re targeting for localization, then doing all of it in one go can really pay off.
Using one provider for all six languages will make for a more consistent translation across all of the languages because they’re all working from the same source and reference material. What’s more, if there’s a bug, error, or query found in one language, teams can cross-reference things to make sure it’s covered across every single translation.
Additionally, future patches, updates, or DLC are much easier with a single provider that can push out changes across all languages.
Yes, absolutely. The best localization teams are the ones who love games and understand the role language plays in them. The more games they play, the more they’ll understand what a good gaming experience is, and the better they can make your game.
The cost of localization varies based on a number of factors such as quality, speed, and volume. But in general, the more complex your translation, the more expensive it’s going to be. Here’s where that money goes in a full localization process:
Familiarization: Per-hour rate
Glossary creation: Per-hour rate
Translation: Per-word rate (numbers and placeholders are words, but if there are repeated phrases, repetitions will cost you less)
Revision: Per-word rate
Test plan creation: Per-hour cost (this can be reduced if you have your own plan)
Localization quality assurance: Per-hour rate
Project management: Per-hour rate
The genre and audience of your game should dictate the value you get from localization. But here’s a rough outline of the costs and reach of some languages:
Simplified Chinese is relatively inexpensive and can open up a huge potential audience.
French isn’t just for France, but also for Canada and other former French colonies. And if you’re releasing your game on consoles, you should keep in mind that France is one of Nintendo Switch’s biggest regions.
Russia has relatively low levels of English fluency, so if you want your game to make money in there it’ll need to be in Russian. Fortunately, localization from English into Russian is on the cheaper end of the spectrum.
Spanish is one of the five most widespread languages in the world and offers good video game revenue.
Limit your changes. Even minor changes introduced after the localization process has started will add to your costs.
Say, for example, you’re translating a game into five languages, and your game has a key weapon that the entire plot revolves around, so the weapon’s name comes up 100 times. A simple change to that weapon’s name looks like this:
100 instances x 5 languages = 500 words to translate
It’s easy to think that a search and replace function would suffice here, but word endings are more complex in other languages, and changing one noun can impact an entire sentence’s structure. (Take Polish, for instance: a noun can have 14 different suffixes that can influence the adjective, which can also have 14 different suffixes.) So each occurrence of the weapon must be replaced manually. What’s more, the new name will influence the structure of the remaining part of the sentence, so the whole string must be retranslated—and on average a string consists of 10 words.
So 100 instances x 10 words per string x 5 languages = 5000 changes
And when you’re paying per word, these costs can add up fast.
On the whole, freelancers tend to be cheaper, but localization agencies have a broaders variety of languages and services, such as localization quality assurance. One of the biggest advantages of using a larger provider is that you have a single point of contact for every single language you translate into, reducing the amount of project management on your end.
Tools like Google Translate are great for small interactions like ordering a coffee in another language, but their tonal and contextual understanding is extremely limited—two things that are absolutely necessary when translating a game’s dialogue and story. Though there’s no replacement for human translation yet, there are tools out there that make the process much easier.
Any localization professional worth their weight should be able to rattle off a long list of tools they use on a day-to-day basis. Here are a few you should check that your localization team is using.
Computer-assisted translation reduces entire game scripts into smaller, more manageable chunks of text. Good CAT tools should also be able to protect tags and placeholders so they are not accidentally edited. The right translation management tool will speed up any localization.
Imagine a single word has been changed in your game’s script. Instead of retranslating the entire sentences, change management tools identify changes in the original language, then mark and filter them in the translated language to speed up the process.
Glossaries (also called termbases) keep key information—terms, proper names, place names, etc.—about a game consistent throughout the translation process. They also give you a uniform approach to translation and allow you to set rules for working with new and invented words (which many games, especially story-heavy ones, have in abundance).
This tool identifies identical pieces of text in identical contexts and translates them in the same way prior to presenting it for revision to the linguist—a huge time and money saver on any project.
The best way to ensure a game’s tone remains consistent throughout the localization process is through familiarization. This is a process where the translation team will learn and document key information about the game—like its tone and cultural elements—and then build a glossary and style guide that ensures everyone on the translation team can keep things consistent throughout the entire project.
There’s no sure-fire way to know you’re getting a top-class translation. But there are some things you should check out with a provider before you engage:
Culturalization is how you adapt content for other cultures and other geographies beyond just normal localization. Culturalization looks beyond language and considers other aspects of the content that might be sensitive in particular markets.
It could be the representation of characters, history, religion, the use of symbols, gestures, body language, or color usage—all of the things that go beyond language adaptation.
An 18+ rating for a game that is intended for a more family-friendly audience is one of the most damaging outcomes for any title. It misses the target audience entirely and the game might not be appealing to the audience that can buy it. Culturalization picks up the culturally sensitive elements in a game and alters them to make sure your game is compliant to a specific region’s cultural, social, and legal requirements.
Think of LQA as the last line of defence between your game and a bad localization. LQA protects you in these ways:
One of the best ways to analyze LQA standards is through submission rates. Submissions are all the checks and balances you need to tick off before a hardware manufacturer will accept a game—for example, a controller disconnect message needs to display the right terminology at the right time. If a submission rate is in the 95-100% range, you’re probably in good hands.
The sooner the better. If you start developing with an idea of which regions you’re planning to release in, it saves a lot of technical issues later. If you’re thinking about localization from the start, you’ll avoid language- and culture-specific issues, like noun/gender issues, special characters, and non-breaking spaces.
Thinking about localization early also sets you up to finish a project on your timeline, rather than force out a lower-quality rushed job at the end of development.
The single biggest thing you can do to cut your translation costs is to make your content consistent. Say, for example, you write three different strings that all have the same meaning, and are used in similar contexts:
String 1: I’m from London.
String 2: I’m from London
String 3: I’m from London (Yes, removing a full stop counts as a separate phrase.)
Computer-assisted translation tools will identify this as three separate phrases, despite it really only being one. So instead of translating a phrase once, these phrases will be set aside as three separate translations.
This doesn’t sound like a big deal, but think of it like this:
String 1 + String 2 + String 3 = 10 words
10 words x $0.10 (generalized cost of translation per word) = $1.00 (the cost of making a single change).
Now multiply that by, say, another 1000 inconsistencies in your project and the costs stack up.
But if these phrases are all the same, the calculation becomes a whole lot different. Instead of 10 words, a CAT tool would only identify four. And because the cost of a repeated string is lower than an entirely new translation, the cost of any change is reduced too.
Additionally, clever coding can cut costs. If you are using the same string all the time, code it in so that it can be called up in different parts of the game.
Finally, avoid hard-coding text. This is super, super mega important. If your text is hard-coded, it will take you a considerable amount of time and effort before you can even send it off for translation.
Spreadsheets are king. Handing over your project in a clearly formatted spreadsheet makes importing it into translation software fast and easy.
The best formats are multilingual compatible. A perfect example is an .xliff. These are files designed to be multilingual and allow for contextual information, etc to be included.
Still, in many cases spreadsheets are still used. In this case, it is vital that the spreadsheets (and its tabs) are all laid out using the same template or format.
Pro tip: Don’t put your strings in alphabetical order.
The more context the better. If you can give the localization team a full version of the game to play through, give them that. If not, any information about the tone, style, and background information of the game will go a long way to producing a better-quality result.
Every question you need to know before localizing a game, answered.
BUT BEFORE YOU GO,
WE’VE GOT ONE