How to Localize a Game on a Tight Budget

Please translate this game into Russian! We need a Polish version! We’re crazy about this kind of ARPG!! I wish this game was in Brazilian Portuguese so I could recommend it to my audience.

The players were begging the developers and I relayed the message. The problem? There was no money left in the localization budget for any extra languages. The real tragedy: the game was translated at the last minute for regions and players that weren’t interested in the game.

Read on to learn how to minimize your localization costs while also making your game more accessible to the players who care about it the most. We’ll cover five factors that will save time and money:

1. Timeframe (Localization can take weeks/months)
2. Internationalization Practices (Preventing problems at the earliest stage possible)
3. Language Pack Size (Striking a balance between content and costs)
4. Target Languages (Who cares about your game anyways?)
5. Language Complexity (How difficult will the translation be?)

Main Factors in Localization Spend

Timeframe

Timeframe is the high up on the list because it is the one variable that can be easily managed without altering the game itself. Tighter timeframes = higher prices. This is true for most services. Want your visa expedited? Pay more. Want your new toy to arrive in 1 day instead of 5? Pay more. Want priority customer service? Priority boarding? You get the idea.

So figure out how large your game is (in words/characters). Get some quotes and estimates from language service providers. Plan ahead. Games take weeks or months to properly localize. If you didn’t plan ahead and need your game translated NOW, then the project manager has little option but to recruit more translators for the project. The more translators working on a project, the less the volume will be for each linguist. Lower volume means higher rates set by the translators. Furthermore, the more translators a project has, the more likely it is that consistency (in style, tone, and usage) will become an issue.

So plan ahead and allow for enough time to translate your game. Not only will the quality improve, but you can also save money this way. Translators and localization project managers alike love longer timelines and higher translation volume divvied up among fewer people.

Poor Internationalization (Technical Debt)

This can be a massive hidden cost that is difficult to quantify. The engineering required to fix a game that wasn’t built for multiple languages can potentially cost more than the translation itself.

Key Considerations

1. Hard-coded strings. These are bits and pieces of text that are buried inside code rather than an external file. These strings will need to be hunted down one by one, using up valuable time that devs could be using to refine the game’s mechanics, fix bugs, etc.
2. UI Flexibility. If your buttons, text boxes, or menus cannot adapt to the target language, you’ll either pay more for localization quality assurance (LQA) after the game is translated, or your UX/UI team will need to spend time re-making UI components. You might need to pay for both.
3. Concatenation. Let me provide a few examples:
   

• [Color] + [Item]. In Spanish, for example, adjectives change depending on the gender of the noun. Without knowing what the exact item is, the translator will need to guess the adjective’s gender.
• [Number] + [Item]s. In many languages, the noun will change based on the number.
• Subject-Object-Verb. English follows SVO while other languages are SOV. If the string is hard-coded in one language, it could ready very awkwardly in another language.

Concatenated Strings

String 1: "Greetings, " [playerName] +

String 2: "! You have earned " [goldAmount] +

String 3: " gold."

In this setup, the translator cannot freely change the word order to match their language.

Named Placeholders

"Greetings, {playerName}! You have earned {gold_amount} gold."

With named placeholders, the translator can now re-arrange the placeholders to fit their language’s natural flow.

Language Pack Size

The size of a language pack (total words that need to be translated) is the biggest variable under your control. Though tweaking this variable can yield the largest difference in price, I listed this as the second consideration because sometimes the amount of language in a game can’t be changed.

Beyond saving money, the lighter your language pack is, the more money you will have to spread across more languages. If your game is a 1+ million word behemoth, you may only be able to translate it into 2 or 3 languages. But if you have a leaner language pack, then you’ll be able to target more regions and thus more players.

Where you can cut words

1. The main story/dialog. Is all of this background information useful? Is this fluff, or does it really contribute to the story? How much will players care if this isn’t in the game? Will anyone notice?
2. Lore. Here I’m talking about those extra stories, legends, and background information discovered in the game. This lore might be found in a book on a shelf, or in a scroll locked in a chest. Character backgrounds, monster beastiaries, and all those little easter eggs contribute to the language pack’s size.

The game I mentioned in the introduction suffered from this exact issue. The game was dense with dialog, detailed character backgrounds, and really long beastiary entries. While these things are nice, the sheer size of the language pack forced the developers to limit the game to just four languages. If the language pack was cut in half, they could have used the same budget to translate into roughly double the amount of languages.
The main story + lore are the first places to cut text because they are often the most verbose parts of the game. In addition to this, the language in these strings will be less repetitive and more context dependent. This means that word repetition rates will be low, and translation will take longer.

Target Languages

Who actually wants to play your game? Do you have data from your Steam wishlist? From App Store or Google Play pre-registrations? From your own website or social media? Look at this data before you decide on what your target languages are.

EFIGS (English, French, Italian, German, Spanish) is a common set of languages that game publishers cover. But what if your game is receiving a lot of hype in Brazil? What if Polish gamers are signaling high interest in your game while almost no French speakers are? Focus your budget on the markets showing the most interest.

Here again, the ARPG I was working on had English, Chinese, Japanese, and Korean versions. The problem? This game was really popular in Russia and eastern Europe. Nothing could be done but cringe and apologize because the developers had no more money to spare for their actual fans.

Language Complexity (aka translation difficulty)

Does your game invent languages for a certain tribe, race, or kind or people? That’s hard to translate. Does your game deal with quantum physics, space travel, or something super technical? That’s hard to translate. Translation difficulty doesn’t affect the price as much as the overall volume of words, but it’s worth considering.

If your game is full of niche concepts and terminology, then finding an appropriate translator will be harder. And if you’re translating this game across multiple languages, this effect compounds.

For controlling localization costs, I think these are five solid variables to start with. The earlier you take internationalization and localization into consideration, the more time and money you’ll save over the long run!