How to localize your SaaS platform - The Mentimeter approach

April 04, 2018/10 mins min read

Despite being based in Stockholm Sweden, Mentimeter has a global audience with users from all over the world. Like many SaaS companies, trying to create a great user experience for many different audiences, operating in many different languages, with a small team based at one location can be a challenge. This blog post tells the story of how Mentimeter has tried to tackle localization during the past few months, without hiring a single translator.

To localize or not to localize? That is the question for many SaaS companies looking to reach a larger audience with their product. In 2017, Mentimeter, and in particular our UX Designer Maja Jakobsson, was also faced with this question, and decided to find a solution. Despite being based in Stockholm Sweden, Mentimeter, has a global audience with users from all over the world. Like many SaaS companies, trying to create a great user experience for many different markets with a small team can be a challenge. This blog post tells the story of how Mentimeter has tried to tackle localization during the past few months, without hiring a single translator.

Why did we prioritize localization?

Mentimeter has always had a global audience, and has always operated in English. This is standard for small SaaS companies as it’s easy to maintain, reaches a wide audience and makes the product much more scalable.

However, we realized that there were some limitations to only providing a product in English. For example, in the case of Mentimeter, the presentation may be in one language but the any buttons and information from the platform that your audience sees will be in English, creating a disjointed experience for the audience.

We also want to make Mentimeter accessible for possible for the widest audience, and having no localization whatsoever, felt that we were potentially excluding many potential users and audiences. Most importantly, we had received many requests from users for this feature, and as a company that is both product and user-driven, we decided that it was time to take action.

How did we localize Mentimeter?

In business and product development, localization is a broad term. When a company decides to localize their product or move into a specific market, this could mean any number of things. It could mean anything from translating a website to opening an office in a new country. Therefore, we knew that we needed to clearly scope out what our goals were and what we wanted to achieve with localization. So our first step in localization was to define what localization meant to Mentimeter.

How we defined localization

We started to define what we wanted to achieve with localization by reflecting on one of our core values: to make presenters look like stars on stage. With this in mind, combined with requested we had received from users, we knew that we had to focus on using localization to help the presenters feel more confident when presenting with Mentimeter and making the audience experience as best at it could be. Internally, we also knew that we were a small team, and that we wanted to remain agile and flexible - so localization couldn’t be something that would slow us down, or drain our resources.

From this, we drew the scope of the project. We decided that we wanted to focus on translating what the audience sees, rather than focusing on translating the editor or spending a lot of time localizing our marketing.

We decided to focus on translating the parts Mentimeter that are most important for the audience. This included: The voting instructions, i.e. the top bar on your presentation that direct users to go to Menti.com, Quiz instructions, translation of buttons and question-specific instructions in menti.com when the user has joined a presentation, information messages and other information.

How we translated voting instructions and menti.com

As a small company, we knew our limitations. We do not have offices in different locations or a team of in-house translators ready to take on the project. For us, the best way to start implementing this project, was to first translate some languages in house and then reach out to our active user base, and particularly users who had requested the feature to help us with translating this text.

At Mentimeter we have a very experimental approach to working. We usually create a hypothesis and then try out the experiment on a small scale before rolling it out. This is exactly what we did when translating the product. We first translated parts of the voting instructions and Menti.com to Swedish as we have internal knowledge of this language. We then reached out to out Swedish users to let them know about this update, and we received positive feedback.

At this point, we started to get a lot more requests to translate Mentimeter, including users who wanted to volunteer to translate for us. So, we started approaching those users who had enquired about translations and those who had volunteered their skills to see if they wanted to help. By helping to translate Mentimeter into their own language, users would be able to use the feature for themselves, thus increasing product value for themselves, whilst helping others who are also presenters.

We managed translations by inviting users who wanted to help into a platform to manage this. They could then work in their own time to create the translation which Mentimeter could review and then publish it - ready for users to use.

Crowdsourcing worked well for Mentimeter because we have amazing users who wanted to help out with the project. Crowdsourcing would also help us to have an agile approach to the project as once a language was translated, we could make it available for our users quickly.

What are our learnings so far?

As mentioned above, the project is still ongoing, however, we have learned some things already. First, we have learned a lot about our users. We have learned that they are engaged with us and the product, which made crowdsourcing translations a breeze. We also learned that amongst our users, there’s a strong sense of community. This has been shown through users’ willingness to help us improve the platform - which feels amazing!

We can also see that this was a good project for us to prioritize. There has been a fast adoption of the feature already.

What’s our advice for companies starting with localization?

The biggest piece of advice we can offer to companies who are looking to start changing their product or marketing for local markets, is to make sure you scope the project clearly first. For Mentimeter, this meant only focusing on the main pain points of the users - translating the audience-facing parts of Mentimeter.

You should also consider what your goals are internally. For us, it was important that localization didn’t slow us down, particularly in terms of product development and marketing.

Companies may be scared to start working with localization as they risk losing momentum when implementing new things, and as a SasS company, you don’t want friction. We drew a clear line and decided on what to do — we knew that we were not providing a full localization experience for our users, however, what we offer is enough to extensively improve their product experience.

What are the next steps?

We will continue to roll out more languages through crowdsourcing, as users need and request them. Once a new language is available, we will let the relevant users know! Learn how to change the Language in Mentimeter in this help article.

At the time of publishing this blog post, it is possible to translate the voting instructions and menti.com into the following languages: