Localization maturity in game development

Jasmin Jelača is a localization lead at Nordeus, a gaming company based in Belgrade, Serbia. Being raised in the multicultural environment of Berlin, it was unavoidable that he become a specialist in various languages and cultures. He holds a master’s degree in German language and literature.

Jasmin Jelača

Jasmin Jelača

Localization maturity in game development

Jasmin Jelača

Jasmin Jelača

Jasmin Jelača is a localization lead at Nordeus, a gaming company based in Belgrade, Serbia. Being raised in the multicultural environment of Berlin, it was unavoidable that he become a specialist in various languages and cultures. He holds a master’s degree in German language and literature.


ocalization maturity is a term that can be heard frequently in our industry. It indicates where organizations stand in relation to localization, how they can improve and make sure that this is one of the central functions in the company. This is vital for every team in our industry, whether it’s in gaming or not. Gaming may be a fun sector, but that doesn’t mean that processes should be kept on the light side.

What is localization maturity?

The localization maturity model (LMM) was brought to us by CSA Research back in 2006. It is a method to measure, as the name indicates, the maturity of an organization and their respective localization teams. It is utilized throughout the industry as a great method to see how teams and organizations can improve their workflows and deliver more quality content to their users. There are nine phases in the LMM model, from -3 to 5, with 5 being the highest level of maturity.

Level 1, or Reactive, means that roles and responsibilities are unclear, localization gets done when it needs to without too much planning. It’s basically just putting out fires when they arise. Level 2, or Repeatable, starts to show the usage of some basic workflows, and the use of outsourcers. Level 3, or Managed, means that processes and workflows are set up and used, as well as localization tools. Depending on the setup of the team and organization, localization work is either centralized or spread throughout the company. Level 4, or Optimized, means that standards are followed throughout the company, documentation is kept up-to-date, and probably the most meaningful of all: localization is accepted as one of the key pillars of product and company success. Level 5, or Transparent, means that localization is a central part of company plannings and product launches, the team is able to scale the work, and metrics are being followed and met. I will not cover -3 to 0, as even writing about these phases gives me anxiety

How is this connected to game development?

As the gaming sector matures, players expect localized text of higher quality, better cultural adaptation and more immersion, which low-quality text can render impossible. One area where the gaming industry sets itself apart from the others is creativity. Anyone who has ever worked on a game knows how difficult it is to come up with the core loop, the meta and a story that adds to that experience. Immature teams will have a hard time dealing with all of this, as they are constantly firefighting and making sure that they are able to deliver translations on time. The LMM is a great solution for teams that want to challenge the status quo, even when things are going fine. There is always room for improvement.

Another requirement from gamers, especially in the last few years, is to deliver a lot of in-game events, to keep the game fresh and attractive. Games as a Service is the new hot thing, and supporting teams need to be able to fulfill those requirements and operate at a very high pace. Continuous delivery is the new standard, especially for mobile games.

Understanding the current maturity level

There are different ways to get a score. A few companies are specialized in improving the localization maturity of organizations, offering researches and workshops with your in-house staff and key stakeholders. Another way to do the scoring is to define what each level means to your organization. This second approach is more tailor-made, cheaper and can still bring you a lot of insights into your current situation. Industry standards serve a great purpose as a baseline for measurements, but ideally every team would come up with their own aspirational level 5. It is important to point out that the scores themselves are not the main purpose of this exercise. The main purpose is to identify areas for improvement and kick off conversations with stakeholders. This is the most important step. Getting buy-in from the company leadership is, without a doubt, an area that should be focused on the most. Including other stakeholders into the initial scoring, by setting up surveys and interviews, will help as well. This process takes time, being thorough and gathering proper research will take a few weeks, but the results and, more importantly, the action steps derived out of this effort will definitely be worth it. It is important to highlight that it’s not possible to go from level 2 directly to level 5. It’s a long process, some teams even spend years rising the ladder, but it’s a climb that is vitally important to your product succeeding globally.

Time for action

Once all the areas for improvement have been identified, it is useful to brainstorm different ways certain problems can be tackled or workflows and processes improved. Things to keep in mind: what are the needs for the team and the company? Where are gaming and localization industries headed? What do the players expect? These are only some of the questions that should be asked during the brainstorms. The next step is to combine all of that knowledge and create action items. They can be by category, prioritized by effort or sorted any other way you see fit. Once the plan is created, company leadership will need to provide guidance and set short and long-term goals for leveling up localization in the organization.

What is the impact?

One of the great things about thorough research like this, is that it showcases everything that is going smoothly, as well as things that are not. So the list of action items can cover many topics.

Discipline promotion

Something that all disciplines need to work on, not only localization, is internal promotion. You want to make sure that everyone in your company understands your terminology. Education is a huge deal, and there are many ways of educating colleagues. Whether it’s blog posts, success stories, your aspirational goals for improvement or workshops for different target groups, all of this will have a positive impact on your collaboration with colleagues. Another way to promote localization is to make it part of the onboarding for new employees, especially the ones who will end up being key stakeholders in the process (game designers, UI/UX artists, developers, product managers, producers).


It’s always important to be involved, especially in decision-based meetings, plannings and retrospectives. It’s crucial for localization teams to have input on wireframes and early design documents, in order to make the games ready for localization. If there’s a possibility to collaborate closely with copywriters, or even own that part of the development process, it makes an impact in the feature design, which in turn makes the quality of the localized text better. The better the source, the better the translations.

Automation is another crucial topic, and has been over the past few years. Long gone are the days of just copy/pasting assets, manually publishing the latest translations for the games and pulling data in order to make strategic choices. As a hint for doing the rating process correctly: automation should always be one of the areas for improvement.


Since games are a special category on its own, there are no standardized QA metrics for game localization. Gaming companies tend to use the standard QA methodologies to measure their output. However, what can be done is to adapt those metrics and create benchmarks suitable to the genre. Of course, mistakes in meaning will always have the highest error severity, but I’d argue that a word-to-word translation is just as bad. If the source text is of high quality, the translation must be able to follow this. Another way of ensuring high quality is to work tightly with the functional QA team and educate them as well about localization mistakes and how to avoid them. Eventually, they will start searching for those on their own.


For teams that don’t have a clearly defined strategy, the first step should be to check in-game metrics and acquisition numbers. Based on that, benchmarks can be defined, as well as go-to-market strategies. Another thing that can be useful are language tiers, which can be defined either by industry standards or by your own products. They can cover what to localize, how much and where. Working closely with product leadership and the finance department will make sure that everything is transparent and choices are made in collaboration with the main stakeholders, which strongly relates to a level 5 localization team.

What does the future bring?

The gaming industry is evolving at a rapid pace, and is already the most lucrative entertainment sector. With mobile devices being more powerful than ever, it’s quite clear that all teams working on games will need to step up their game. The LMM is a nice guideline in determining which direction teams should go. Essentially its main purpose is to kick off conversations and make sure that localization is an important pillar of the company and that the localization teams, and therefore business, are set for success.