Managing localisation at Ferryhopper

Vasso: Thank you again for joining us for this interview. How did you initially land in the localisation industry? Sometimes I meet people who landed in our industry out of the blue, and other times it’s people who have been in it forever.

Maria: Thank you for the invitation. I’m definitely not in the industry by accident. I wanted to become a translator, and that’s how I got here. I did a BA in English Studies and then started a Master’s in translation. As part of the Master’s, there was an internship at a translation agency. It was supposed to be mainly translation work, but they offered me a job as a project manager, so officially, that’s how I was initiated in project management. I’ve done a lot of things since then, but it was an intentional journey. I expected to do more translation work—I still do a bit—but at some point, I realised I’m good at project management and I enjoy working with people in different roles. That’s how I got here, basically.

Vasso: A nice, planned journey. It does seem like project management comes into a lot of localisation professionals’ lives after they start as translators, right?

Maria: Yes, I think so. I completed my PhD in localisation last year and I tell my students that you can’t do just one thing nowadays, especially with how fast the industry is changing. Not everyone is meant for this kind of focus on process or communication, but if you have that ability, it’s definitely an asset. Even if you don’t move into other roles and just stay as a freelance translator, it’s helpful. When I worked as a translator, I found that my first job as a project manager helped me understand the other side, the communication barriers between stakeholders. Understanding other roles is always valuable, even if you don’t pursue a career in that direction.

Vasso: You raise a good point—we’re all part of an ecosystem, and being more aware of the needs and challenges of different roles helps us be more helpful and valuable.

Maria: Definitely. It makes us more adequate too, because sometimes, if you don’t put yourself in the other person’s shoes, you don’t know why they do what they do in a particular situation.

Vasso: Great. Now you’re the Localisation Manager at Ferryhopper. How would you identify Ferryhopper in terms of size?

Maria: I think we are more like a scale-up and we have moved forward from the start-up phase. We’re not a corporation or a huge company, but we are definitely growing and moving into new markets.

Vasso: If you had to describe the localisation programme at Ferryhopper with three words, which would they be? And could you give us a bit of context?

Maria: Hmm, that’s a hard question, but I’ll say: diverse, flexible, and dynamic. And I feel that this is what we should all be striving for. “Diverse” isn’t just about the diversity localisation brings to a product, but also about the types of projects and people we work with. Our work at Ferryhopper is interdisciplinary and depends a lot on each specific project or team—Product, Customer Service, and so on. By “flexible,” I mean there’s a different solution for everything, so it has to be customised. We can’t keep doing what we did five years ago; each language or project might need a new strategy. And “dynamic” because everything is always changing—in the industry and the company. As we grow and add more languages, things keep evolving and each year we set new goals and come up with new plans for localisation. There’s always something new to explore, within my role, too. That’s probably one of the main things I like about this job.

Vasso: I totally agree. I’ll keep your point about the industry changing for later in our discussion, but staying with Ferryhopper for now, how is Ferryhopper’s localisation programme different compared to when it started? Did you join at the very beginning?

Maria: I didn’t join right at the very beginning. But I joined to build the localisation strategy. We already had different languages when I joined, but things were more ad hoc. Over the years, we had to establish processes and a team.

Now it’s much more complex compared to the beginning when we just covered the main localisation languages (outside of English and Greek)—German, Spanish, Italian, French. At the start, we onboarded freelance language managers for these languages, and created style guides for Ferryhopper, translated and localised the website, app, all documentation, and started localising the blog, which is our main organic channel. We’ve learnt a lot since then, and now the setup is much more diversified.

There are languages for which we don’t have language managers and for which we don’t translate organic content. For the ones we do, it’s much more focused—we do market research, SEO research, and we know what works better now. For bigger audiences, we have people who are full owners of the language—they do everything, including translation. Their role isn’t just to check translations or be a “human in the loop.” For smaller markets, we might rely more on LSPs. The decision is based mainly on our audience and the number of people using our products in each market. That’s the main difference: we’ve grown, and along with the company, our localisation approach has evolved. For some languages, we just provide coverage in our products and communications, turning to a local translation agency as needed. It’s a sort of two-tier approach, but more nuanced than that.

Vasso: Do you have any specific metrics or experiences showing how localisation has impacted Ferryhopper’s market presence, positively or negatively?

Maria: Everyone wants to see proof, and every client wants to know if they need to localise. In tourism, it’s not really a choice—you have to localise. The bigger question is how you’ll do it: will you focus on quality localisation, or just push your content into as many languages as possible? I can’t share precise numbers, but for some languages, after launch, bookings in that nationality doubled in the next year. Still, it’s hard to say how objective that is because tourism is a very specific industry. Regardless of nationality, our users trust us, we have good ratings, and they know they can use our service in their own language if it’s supported. We try to be as local as possible. So far, it’s working: people trust us, and I think that’s enough to go on.

Vasso: It’s hard to make that connection, since business results are so cross-functional and some business results can be claimed by different departments, like localisation but also marketing.

Maria: Yes, definitely. Officially, we’re under the Marketing Team, but we work with every other team,too. You can’t isolate results—localisation is part of everything. In this industry, localisation is a must, and you can’t do it in a rudimentary way, just using machine translation and expanding as fast as possible. We’ve seen that what we’re doing works, and people are using our products in the new languages we launch. There’s an interesting thing: when we launch a language, especially one that’s not major in Europe, people tend to use English at first, even if their language is available. But over time, with trust, the percentage grows, and more people use our service in their own language. As a Bulgarian speaker, I can confirm that myself—it does make a difference. But it takes time for people to trust the product in their own language, especially if they are used to doing things in English because that’s what was always available to them. As time goes on and they see the quality, more users will shift to using the product in their native language. That’s what we’ve observed.

Vasso: That’s a really interesting insight. It does take time because I think most products also launch with English despite their home market and despite the launch market. They may have support for one or two languages, but they do launch with English mostly until they get everything together to start localising. Also users seem to have that expectation of using apps in English, but when we do find out that we have a well-localised version, then we switch. Do you have any idea of how much time it takes for your users to adopt their localised version?

Maria: I’m not sure to be honest, because we haven’t been around for that long, but it definitely increases with each year. I think we’ve seen it for Polish, we’d had Polish for like 3 years before we saw an uptake in usage. But now the majority of users who are Polish nationals use Polish as a language. Again, I might be over-generalising here, but these are some baseline estimates.

Vasso: And it would be interesting if at some point we could connect that shift, users adopting their localised version, with new subscriptions or new growth in that market. I think that would be an interesting figure to monitor.

Maria: It should be. I’m not sure if it is something that’s applicable everywhere, like for example in our industry, but for certain products, it will definitely make a difference.

Vasso: Going back to what you mentioned previously about the industry changing—honestly, pretty much all industries are changing a lot in these past one or two years. There’s so much hype right now around AI, LLMs, and all the related buzzwords: GenAI, language-specific LLMs, agentic AI, and so on. There are just so many terms floating around. You see hype everywhere, and everyone is trying to figure out how to use these technologies. What’s your overall take on all of this?

Maria: Yes, it’s all the buzz words! It’s definitely a huge topic, as you said. Everyone is trying to figure things out, and I think the main thing is that AI is clearly here to stay. Whether we personally like it or not, or whatever our preferences might be, every industry—not just localisation—is looking for the best ways to use it.

What’s really important, in my opinion, is to approach all of this analytically and critically. I’ve noticed that, in most cases, people are quite chaotic about it: “Oh, AI can do everything, so let’s make it do everything for us.” But that’s not a good approach. For example, we know that neural machine translation and similar tools have been around for a while, and they have real added value in localisation. Still, it’s more about finding ways to build on that, and it will take time. People are experimenting but aren’t always sure what to apply where.

The bigger issue for me, especially in localisation, is that there are a lot of stakeholders involved, and communication is not always clear. Not everyone fully understands what can be done, what’s possible with a particular type of GenAI, or what a language-specific LLM really is. I believe it’s essential for us to inform and educate ourselves and others, try new things, and test them thoroughly. We need to be more data- and results-driven and keep all stakeholders—clients included—informed.

I’ve even seen this in Ferryhopper. Sometimes, other teams say, “Oh, we already did this with ChatGPT,” and I have to ask, “Why?” At least let’s see how we can improve it, write a better prompt, or try something new. Sometimes people think, “It’s just translation” or “It’s just the next generation of this technology—of course it can do it!” That mindset is limiting. So, educating people is going to be one of our most important roles in the next few years.

Vasso: How do you feel when other teams (not from localisation) come to you and say they already did something—just because the tools were available and they could do it? To me, that’s one of the biggest changes LLMs bring, they give everyone the sense of control, the feeling that, “I can do this myself.” Sometimes, just because we do it ourselves, we think it’s good enough, or we’re willing to compromise. But if we give it to an expert, our expectations for quality shoot up—which is another discussion, since quality is very subjective. How do you feel about that?

Maria: Honestly, I think it makes sense from their point of view. It’s not like people say, “I did this with ChatGPT and it’s perfect.” Usually, they use it when something is not very high-priority, or they need a quick result. It’s not for content with high visibility to customers; it’s more about needing something now rather than needing it done in the best possible way.

So, I don’t get the sense that people believe they can do it better than the Localisation Team. It’s more about prioritisation, which I actually understand. But this is where the real discussion starts—about when it’s enough to just “have something quickly” and when quality becomes important. That’s the balance we need to talk about more as these technologies become part of our everyday work.

There might be problems that non-linguists simply won’t notice because they don’t have that kind of perspective. They might be Product Team members, engineers, etc. So it really comes down to communication—identifying these cases, trying to improve them, and educating people.

For example, I recently saw a prompt used for translation that was something like “translate without any changes.” Obviously, the intention was to prevent ChatGPT from altering information—because we know it hallucinates or makes things up. The prompt was written by engineers, I think, who tried to ensure the AI wouldn’t introduce changes. But from my localisation perspective, that prompt effectively says, “Don’t change the syntax; don’t change the words; just translate them directly.” This shows the importance of linguistic sensitivity. You do need someone with a professional perspective to guide how to practically use GenAI every day.

Vasso: Yes, you’re right. Moving beyond linguistic expertise and education, with more self-service approaches becoming common, do you see risks that this might create gaps or impact business practices? As an in-house localisation team, you have workflows—you collect and store content, keep translation memories or databases to maintain consistency across markets, languages, and content types. With the DIY approach enabled by LLMs, do you see risks or gaps developing in your processes or even in the overall brand image?

Maria: I think it definitely depends on the company and how communication about localisation happens within it. In our case, I don’t think it will create huge gaps because many people are responsible for different content types, they are aware of style guidelines and potential risks, and even if someone decides to use machine translation quickly for something, there is usually a language manager to check it afterwards.

That doesn’t happen often, but when it does, at least for cases where we don’t have the capacity to fully control everything, the output will be “good enough” for its purpose and we’ll improve it if we get time. I’m not sure if this applies to all companies. I’m not saying that we’re completely different or that nothing will change, but at least for now, we’re trying to make informed decisions. Of course we use AI and machine translation—we just try to use them in a sensible, controlled way.

Vasso: Good points. So, to wrap this up: what is it that you love about the industry? You consciously chose to become a language professional—what do you love about it?

Maria: Some of the things I mentioned earlier stand out for me. It’s the diversity of people and teams that I get to work with. There’s always something new to learn, whether it’s about software product creation, marketing, or market research. Of course, it depends on the person, but for me, it’s extremely rewarding to have something new to learn. So, the diversity and all the different types of people you meet and collaborate with—that’s what I really value.

Vasso: And what is one piece of advice you would give to peers in the localisation industry?

Maria: I usually give my students this advice: don’t be afraid to try new things, and don’t be worried about the state of the industry. There’s always something new happening, and there are always people saying we’re doomed. Anyone who wants to work in localisation needs to stay curious and open. Be proactive and explore how you can make use of what’s happening and apply it to your work. That’s the most important thing. And I don’t mean just in terms of automation and technology, though that’s big right now, but also in communication. If you’re open when communicating with project managers, other translators, reviewers, or software developers—as often as possible—you’ll learn the most that way. The industry is dynamic, so flexibility and willingness to learn are really valuable.

Vasso: That can apply to almost any industry! I completely agree. Today’s professional environment is so different from what it was twenty or thirty years ago. People used to start in a company and retire there.

Maria: Absolutely. And now, with remote work—something we already take for granted—we feel that comfort of our own time and space, wherever we choose to work from. But communication is still the most important aspect in any job. Even if you imagine yourself as a translator sitting at home all day, communication is still crucial. I consciously go out to work from a place other than my home. I may choose a friend to work with, or go to a place I like—it helps. Of course, it depends on what works for you, and how you feel about your alone time, but I think having a little bit of everything is the best mix.

Vasso: I think we’re a flexible crowd. Most people you need to communicate with in this industry are aware of different expectations and communication practices.

Maria: For sure. I see this a lot in localisation. Compared to other professions, localisation professionals are less set in their ways, probably because we have to work with so many different people. Over time, you learn and adapt.

Vasso: One final question: do you have a big, bold goal you want to achieve in the next two or three years? It can be either professional or personal.

Maria: Honestly, based on my experience so far, both professionally and personally, I don’t think it’s a great idea to put deadlines on big goals—especially the really big ones. As I mentioned, I finished my PhD last year, so now I’m trying to go with the flow, see where things take me, and have that attitude. Right now, not only is the industry changing, but so is the whole world—the sociopolitical climate included. I won’t even get into that topic! I think we should focus more on our well-being and our communities. That’s the kind of goal I have now—not a specific big thing, but more of an ongoing attitude to find balance in life and focus on community.

Vasso: I like that goal. I think it becomes very important after some years of professional experience, and we shouldn’t postpone it. When you’re younger, you want to push for more, get more pleasure and adrenaline from accomplishing things, and moving quickly. But after a while, you stop feeling like you constantly have to prove yourself.

Maria: I think everyone should do that, no matter where they are in their career. It’s tricky though—when you have that passion, it’s tough to slow down. I still find myself thinking, “This is the next thing I should try.” But you start finding a balance, prioritising, and deciding what’s truly valuable and worth your time.

Vasso: Yes, and you start enjoying other things too. Perfect. I’ll leave us with that. Thank you, Maria. Your insights have been so helpful, and I’m sure readers will appreciate the advice and the honest perspective on what localisation looks like in a growing company.

Maria: Thank you! It was a pleasure to chat and share my experience.