Yana Tarasevich on Designing Beyond Translation
Yana Tarasevich on designing beyond translation.
Your product works in English. But does it work for everyone else? A conversation about language, culture, and the UX decisions that make a product feel native.
Published October 19, 2026 · Hosted by Nick Mann
BETTER EXPERIENCES.
Good UX speaks your language.
Think globally.
Design thoughtfully.
Design for more than English.
Spot the hidden assumptions in text containers, forms, fonts, and reading direction before they become barriers.
Test before you translate.
Use pseudo-localization to expose layouts that break with unfamiliar characters or longer strings.
Give translators the whole picture.
Bring localization specialists into design reviews and share visual context alongside interface copy.
Use AI with a quality check.
Understand why content type, context, and human review matter when evaluating AI translations.
It takes more than
the right words.
An interface can feel effortless in one language and become frustrating in another. A label no longer fits. A form rejects a phone number. A phrase sounds unfamiliar, even when every word is technically translated.
Yana Tarasevich, a localization team lead at Alconost, joins Nick Mann to unpack the design decisions behind those experiences. Drawing on her work in translation and localization, she explains why adapting a product means thinking about culture, context, and the everyday details that help people feel at home.
The conversation covers practical ways to prepare designs before translation begins, how to collaborate with localization teams, and where AI can support the work without losing consistency or user trust. If you design for an international audience, or might someday, this is a useful place to start.
“good localization is something that you just don't notice.”
YANA TARASEVICH / Localization Team Lead · Alconost
Explore the
conversation.
From the first design decision to the final translated word.
Select a chapter to jump to that part of the recording.
Yana Tarasevich
Localization Team Lead · Alconost
Yana Tarasevich leads localization teams at Alconost. In this conversation, she shares more than 14 years of experience in translation and localization, from working as a translator to managing projects and teams of localization managers.
Resources mentioned
Pseudo-localization
A way to stress-test designs before translating into real languages.
Figma and translation management systems
Sharing visual context with translators through connected workflows.
Linguistic quality assurance
Reviewing localized content for meaning, consistency, and cultural fit.
EVERY WORD, AT YOUR PACERead the Full Transcript
Supplied transcript, lightly formatted. Speaker labels were not provided; wording is preserved.
Hi, this is the UX Growth Podcast, the podcast that helps people learn and grow in the UX design industry. I'm your host, Nick Mann. So I'm here with season six, and I'm gonna be joined by Yana, a localization team lead at Alconost. So what is Alconost? It is a full stack localization company that helps teams adapt apps, games, websites, software, marketing, and product content for international markets.
So Yana brings a localization perspective to a UX conversation, which is exactly why I'm excited for this episode, because sometimes the biggest user experience problems do not show up when a product is being designed in its original language. They show up when that product has to work for people in another language, another culture, or another market.
So today we're gonna talk about how localization goes beyond translation, what product teams often miss when they go global, and what UX designers can learn from localization experts before the designs break for users somewhere else. So thank you so much for being here, Yana
Thank you for having me.
Yes.
So let's begin tell me a little bit about yourself and how'd you get started as a localization team lead.
Sure. So I've been working in translation and localization industry for, I think, more than 14 years now, because I started as a translator myself, and then I, became a localization manager, which is a person who manages localization projects, including translation.
And now I'm a localization team lead at Elkonost, where I manage not only the projects, but also the teams of localization managers who take care of localization requests.
Mm-hmm. Yes.
What Localization Really Means
So for UX designers who may think localization just means translation, how would you explain what localization actually includes?
Well, it's a common misunderstanding, I would say, not only for UX specialists, but all, around the digital industries.
Mm-hmm.
And, the easiest way to explain it is maybe to, think about how you use the apps, how you use digital products, how you use, I don't know, how you play games maybe in your own language, and what really makes you feel that this product, this game, this app was made for you, in the first place.
So when you have this feeling that it was made in your language, that it was made, like in your country, for example, this is when the UX and the localization are done perfectly. And when you start to notice some small things, not only language things, but, for example, the, French, date format you are not used to, the lack of, your currency of your country in the currency list.
So this is where you see that UX was not really ready for localization. And, things like that, they are also part of localization. So we're not only talking about translating something word to word, but we are also, talking about adapting your product to make it feel native, to make it sound native, and look native.
And, even when we are talking about translation, we need to make sure that the translated elements, they fit, the cultural, atmosphere in the country, in the region, for example. That they fit, at the layout of the app, that they fit the, tone of voice you really want to convey, you really want to use in your app.
So all these things, they cannot be only achieved by, word to word translation, but they need more, like, transcreation efforts and more localization efforts in the end
Mm-hmm. Yeah, I can see that. Well, it's a much bigger field than so many people would think because even I'm guilty of that when I hear localization, I just only think of translation.
But of course, that's a really good example when you brought up currency because the world doesn't just run on just one currency and it's always changing. So what would you look at localization as instead of just being about words, but also how it affects the overall product experience from the very inception of using a product and to actually fulfilling what is the problem the user is looking to solve?
Why Start Early
I would say that the earlier you start thinking about localization, the better, the product will be in the end, will feel in the end for the target audiences for your users. Because, well, it is often another, like, mistake made, when developing anything is, just, thinking about localization as, like, a handoff somewhere in the end of the way.
Companies start to think about that when it's really, maybe not too late still, but, it takes a lot more effort and money to fix the things that could be foreseen at the very beginning. So my general feeling is that, when you start designing your project, just keep in mind that it may be localized in, like, 5 or 10 or 20 languages.
And, just try to think about, as many unusual things in the languages you can imagine, like, a right-to-left direction, like hieroglyphics, like accentuated, letters in some languages which are not supported, in some, fonts, for example. So, Yeah, good example ... yeah, there are so many things that can, like, break in a specific language, that it makes sense to think about that, even when you start designing.
And also, when you start designing, you need to, understand that languages are really very different and that the... some languages are shorter, some are, longer, and that you just cannot, fix, everything, for, English phrases, for example. So the, size of your containers cannot be fixed for English, because if you try to insert any German phrase inside it later, it will definitely break, and you will need to redo it from scratch.
So, the earlier you do that, the more successful localization will be.
Yeah. I do like that German example because I've been studying German myself and can confirm that there's some words that are much longer in their language than in English, and I think a lot of people don't really think about these sort of things because it- it's not apparent in our common day language.
I also know the, Chinese language, they write their letters right to left compared to how many so many of us, especially in English, where we, do left to right. So that's also an interesting change-up in how people are gonna be consuming information on websites and apps.
For Chinese, it's, basically still, I would say, left to right, but it's hieroglyphics, so, the size of, text, will definitely be smaller.
The common right to left English, which is often used for localization, is Arabic. So for that, yes, you will need to, definitely adapt the interface, adapt the design just to not only for the text, but, overall, how it looks, so that it is not confusing for right to left, direction of the text.
Mm-hmm. And I know, and know you mentioned that to prepare for this localization in the first of the design of the app, so, you know, we don't run into these issues later down the road. And I'm curious to know, because I think what happens with a lot of product teams is that they don't anticipate to go to other markets with their app.
They just think about just dominating this one market in, like, say, in the US, and then they adapt it for the other markets. So I'm curious to know, like, how can we make that transition smoother? Because a lot of product teams don't want to think about localization from the get-go because they don't think they're either gonna be ready, or they just wanna see that the m- if the market just, an- any market accepts w- what they are developing right now, and then they can go into other markets when they feel ready.
So I'm curious to know, like, what would you say to those teams who, may struggle with that aspect?
Well, I need to,
Making the Business Case
when I need to persuade anyone of anything, I always try to, give some data, to, to do that. And in this case, the statistic is that 75% of people still prefer to use digital products in their own language.
So that gives us only, like, 25% of people who would accept, the products in, like, in English, for example, in English only. The rest will look for another product in their own markets. And, so what it tells us is that localization gives a great opportunity for expansion, and it's maybe the biggest, in the end.
So, it still makes sense to think about it from the start because it's-- for me, it seems inevitable in our global world. Another thing- Mm-hmm ... is that, some countries really, like, if we are talking about US, for example, of course, there is English, but there are some other languages spoken in the country, like Spanish.
And you, basically can, ignore and can lose a big percent of the audience just by staying English only again. So with this data, with these numbers, I think, in the end, it makes sense to, think about localization earlier and, then we can also
So another thing that, can help persuade people to do localization earlier is showing them how much effort it takes to redesign things, after localization is coming in the picture. Because, redesigning things for a right to left direction of the text, redesigning, pictures where, text is hard-coded in a picture, redesigning containers which are rigid and cannot move to accommodate for longer German phrases, for example, that takes a lot of money, a lot of effort, a lot of time.
And, when you want to speed up, you definitely don't want to, spend all it, on just redesigning.
Mm-hmm. Yeah, and that takes up, you know, resources that takes, you know, takes up and you may have to change up other design interfaces too in the process and, like, that's a lot of time and effort that can be easily spent because of all this, search of things.
So anyway, it can be, it can definitely be expensive if not done properly.
Yes. Yes, fixing is usually more expensive than doing something, like foreseeing something from start. That is why basically for, especially for UX professionals, it is
Pseudo-Localization
highly recommended to try a thing which is called pseudo-localization.
This is a process when you actually, like, fill in your design with the, pseudo language, so to say. It is just filled with all the possible problems you can... that can occur in real languages, so that, before you start translation, you can actually see what are the weak spots in the design, and thus you can fix them before you start translating.
That really helps to save on translation efforts. That helps again to save time and resources. But that is something not, not everybody knows about, so I hope this may be a good thing, to learn about today.
Oh, yes. Absolutely. Think because that's so useful, can you tell me a little bit more about that so, like, UX designers can already start implementing that into their workflow?
Oh, so that is the process. So, so why doing pseudo-localization? Again, it's, to expose, all the errors, all the problems in design that may occur when you do actual translation in actual real languages. So what usually happens is during pseudo-localization, your, translatable strings are filled in with, some symbols, so to say.
Yeah, some... They're not proper words, but they are symbols that may cause some problems. So they are usually like accentuated letters. They usually, they can be hieroglyphics, that can be right-to-left direction, languages. So, so that can be anything that can break your design. And you don't really have to spend a lot on that, so it is not a translation.
You don't need a translator for that. So at any translation management system, you have this opportunity to do this pseudo-localization, and then to upload the results of pseudo-localization back to your designs and to see how the design works with it, if it breaks, where it breaks, and then you can fix all the problems you have encountered.
Hmm. I think that's super important to know. Also, another issue, not only can this be very expensive problem, but I think that, it could also affect users' trust, and that is so huge when it comes to new products on the market that no one has ever heard about. So I'm curious to know,
Localization and User Trust
how does poor localization affect user trust?
Well, we can see a lot of examples, of that, especially in game dev, because, games are expected, often to have, several localizations when they are out. And, recently, especially with AI developing and with, developers just applying raw AI translations, in games, it really causes a lot of backlash from gamers, from users when they see that, the content that was supposed to be creative, is, not only not creative, but it, like, lacks even general, some general ideas.
It is hard to understand it, or it has, for example, some, information, that any, like, AI tool gives you, that should not be in the string at all. So, this really affects the reputation a lot. And, like a recent example here in the Polish market where I'm located is, a game by 11 bit, which was, again, which caused a lot of, backlash from gamers for,
who discovered the, AI translator strings, inside the game. It is not to say that AI shouldn't be used for translation at all, but, some human, participation is still expected, especially when we're talking about creative content. And anything that is creative, which is usually marketing, which is usually ad campaigns, which is usually, some narrative content inside, games and apps, this is where you can expect, like, the biggest issues if you do it wrong.
Mm-hmm. Yep, I absolutely agree with that. So I'm also curious to know, so what happens when a a team treats localization as something that only happens at the very end?
Well, let's imagine that we have an app which was adapted for English, which has, containers with a fixed width, which has, for example, the user information form which only accommodates for US, addresses and US phone numbers, for example. So, what happens if we try to translate it into some languages, in the very end of production, is that, we see that the design, itself is not accommodating, any other cases, any other cultures.
For example, if, I try to translate it into Polish, and I try to, insert, like an example of phone number which has, not, 10 digits but less or more, the design will break, the translation will not be accepted, and the user will not be able to provide their phone number at all. So in here we have a user who just cannot, sign up, for example, and the localization breaks in this way.
So this is why, we're trying to foresee such problems at the very beginning. But that is why it is really advisable to think about that, about designing for, a lot of countries, a lot of regions at least, when you're doing when you're doing design.
Mm-hmm. That's a pretty good answer. I never really thought about that.
Wow. It, it very puts into perspective how deep, localization really can be in terms of, like, all these errors and potential ways to expand, and how to even do it properly. I'm just, because of that, I'm just so curious about how, how, like, UX designers can hand off their work to translators, linguists.
Hopefully, I'm pronouncing that right. You know, people who say languages for a living, and other, like, localization teams can have, like, the better context. So, like, how can UX designers best help support, you know, people who do what you do?
Well,
Handing Off to Translators
first of all, it's really great when localization specialists, like project managers even, are invited to design review so they have a look at Figma, for example, at designs in Figma, and they have just this opportunity to say from the very start what can go wrong even before we start translation.
And, this is a great point for us. And another one is, using integrations to connect, the platform where you do design, like Figma, to translation management systems where translators do their work. This can easily be done today practically with any translation management system. So, what happens in this case is that all translatable strings you choose in Figma, they are uploaded to a translation management system and then a translator not only sees the strings they need to translate, but they also can see, the Figma designs, inside translation management systems.
So they can see the context and they can see the actual, like windows they are translating for. This is super important- Mm-hmm ... because, like, the context is everything for translators and especially for UI, for example, where screens are usually very short and it is really not always easy to understand what you are translating for.
Mm-hmm. Yeah, that also makes sense. I'm also very curious. I think one issue that I think a lot of teams deal with a lot is because their product is also constantly changing is they're always thinking of new testing, new features, new ways to build, especially like we're in a, you know, the era of AI, and being able to build on something is now easier than ever.
So I'm curious to know, how does that work into a localization? What should teams think about that? And, you know, we can also bridge into, you know, like an AI discussion as well.
Well, what is good about AI is that it speeds up not only developers but translators as well. So, all the new features, all the changes, they can be handed off quickly.
For example, by means of integration I just mentioned, then it shouldn't be a problem to set up workflow where translators will take care of new strings for translation, like practically during the same day. So, translations also become faster and faster recently, and, translators can handle, more text, than ever.
So here we are ready to do it, as quickly as the product really requires.
Yeah, no, that is actually really incredible how AI just affects, like, all types of work. I'm just curious to know, like, where in, like, AI translation is its most usefulness?
So,
AI in Translation
actually AI translation starts with quality evaluation. So you first to understand if you can use AI translation at all for your product, for your content, you need to evaluate if the quality of translation is high enough to just leave it at as is or even to proofread it by a human translator.
Sometimes the quality can be so low that it just doesn't make sense even to proofread it. You just need to retranslate every single string. So when we do, AI-powered workflows, in localization, we usually start with, evaluating the quality and choosing the right model for a translation that, works the best for specific languages.
Then I can say that, it also is different for different kinds of content because for for example, if we're talking again about creative and marketing, content, AI is usually not the best choice here. It is not... still not so good with, for example, wordplay, with humor, with things like that.
And here you want to have still a native, translator or a native copywriter who will write the copy in the way to make it sound and feel really native. But for, the repeating content for UI, for example, it really makes sense to use AI translation, but it still should be proofread by, humans, because, AI cannot grasp all the content usually, so it just cannot, take in so so much information at the same time.
And the more nuances there are in the project, the more details you need to understand to translate a single string, the more difficult it is for AI to make the right choice because the content window is not endless, and at some point it just breaks and, starts to, translate as if it didn't see any context at all.
Yeah, that's a really good point, about like context memory. That's essentially it's definitely not infinite in how, that works. Also in fact context in general, like it requires quite a bit of work, and I don't think a lot of people really know how to use it to the fullest 'cause I've been practicing it myself and like it can do absolutely amazing results.
However, the setup for the context, that require quite a bit of work to really make sure that it all comes together and it's all accurate, and then piece that together. And I know like a lot of teams, like they just don't have time for that. So yeah, no, I do think that's like kind of like the issue of the fact that a lot of, teams can just use AI translation and just say good enough and just copy into their prototype and, you know, and say like, "Okay, good enough," because they don't know the language themselves.
They don't know like how well it works. They kind of have to put their faith in the AI knows the best, you know?
Yeah, I actually saw recently quite a lot of, cases when people just translated with AI and just, uploaded it and launched the product without any proofreading at all. But in the end, finally, at least, some QA is, being done on these strings.
At least some LQA, testers are invited to look into it because, what I saw, yes, AI lacks context a lot and, you need to put in a lot of information to keep it on track, and it is not always possible. Sometimes it is possible when you just break your content into small chunks and translate it, like, chunk by chunk, but that also doesn't guarantee that you will have consistency, inside, your translations.
And consistency is another thing that is very important when you do localization because, for example, if you're localizing a game, you need to, so your gamer, your user will expect a character to be named in the same way throughout the game or some items to have the same name throughout the game, and it can break, user experience, completely if they are not consistent.
Mm-hmm. And also to put into perspective of AI translation is, like, when we look at our own language, say English, like, we can see, like, the words it uses, the phrasing, it all sounds very like limited and uses what your average human speaker wouldn't say. So now we can imagine, how would that look like in another language?
Because those same principles are still gonna be applied. So that's how a person in another language will be able to see how we are writing when we put out from this copy and paste AI text. So it really puts into perspective the fact that, like, non-native speakers, when they read it in their language, like, they can tell, like, "Oh, a human didn't write this."
I think that's also very important to remember.
Yes, sure, and it affects, like, the reputation. Again, it can just, cause a backlash, not because the user experience was broken or somebody wasn't able, like, to sign up, but just having this feeling, sometimes, makes people leave and, look for another product.
Yeah, because, I think that's also kind of the issue about, this copy and pasting whatever AI it generates because if they can tell that you did this for the text or copy, now they're gonna be thinking, "What else did you do where you just copy and paste AI into without any kind of thought process or, I don't know, any kind of human review?"
And, you know, that does ruin that trust 'cause now it's like, how else am I gonna trust you with everything else if already when I'm reading I'm not, you know, I'm already feeling a little wary, especially to a, like, a company that I've never heard about before or a product, you know?
Sure, and, actually there is a case, talking a lot about game dev today.
It's just a coincidence, I think, because maybe it's, just the latest news, for me is, that another game company, a Czech company, has, fired their translator who was translating from Czech into English and doing localization, like striving to move to more AI-powered, solutions. And that also caused a lot of reaction from the community because the game they're developing, is, very, like, heavily narrative.
It has a lot of going on. It has a lot of text. It has a lot of characters and items and everything. And now the users are just, concerned about what else, again, will be replaced by AI. So if translators are being replaced, how will it affect, like, the next part of the game? So this is, not only about the trust in a product that is already on the market, but also the trust in a product that will, be delivered in sometime soon.
Mm-hmm. Yeah, I know. I don't... it's interesting how the video game community typically doesn't really like AI. Whenever whenever a game developer does something with AI, I always hear a lot of negative feedback, and it's always interesting to hear about. And I also do think a lot of it is because there is just so much bad use of AI, because I always like to say good use of AI is, like, invisible.
You hardly notice it because it works so well. So I'm,
I'm just curious- So relevant, actually. Yeah. This is something that people usually say about translation, that good translation,
Good Localization Is Invisible
good localization is something that you just don't notice. That if you don't notice any difference between, the product developed in another country and the product developed in your own region, in your own language, this is a good localization.
Wow. That's actually a powerful quote right there, where good localization is also invisible. Wow. That really does make a lot of sense.
And, when I was, studying to become a translator, that is what was, like the motto is for translators to be invisible, both for, simultaneous interpreters, who work with people and, for translators who translate books, who translate games, who translate...
Now, today, they translate apps, and all digital products. So if you can keep your translation so native it is not visible, this is a good work.
Yeah. Absolutely.
So I'm just curious to know,
Advice for UX Designers
what do you say is the one biggest thing that a UX designer can make to for their product to become a little more localization friendly for the future? What do you say is like probably maybe even the easiest way?
So I would say this is actually a collaboration with localization team, if you have a localization team in place, if you have a vendor you're planning to work with in place. Just start collaborating with them and sharing information with them, as soon as possible, as early as possible so that they know how to prepare so that they can help you maybe identify some problems before they become problems.
If you're just, designing a product and you're not really thinking about localization, just thinking about delivering a product, in your own native language, my advice is to do localization we talked about- Yes ... which is really like a powerful instrument to see how this will go when you decide to localize in real life.
And, another advice maybe is just to try to think out of the box really here, because as you said, there are some things that we don't really think about, when we are filling in the forms for users, for example, when we're filling in phone numbers and addresses. Just don't think that they can look in some other way in other cultures and other languages.
And this, is just a simple example but that really goes for the whole design. Just try to look at it through somebody else's eyes maybe, and to see how it can be treated differently by different people.
Mm-hmm, that's a very powerful way to look at it I'm also very curious to know What is a one thing localization teams understand about users that UX designers can learn from?
I think the localization team includes a lot of, specialists, a lot of native speakers who understand their culture, in the best way. You cannot be a native speaker and a person living in all countries of the world and all cultures of the world. So, Mm-hmm ... I think this cultural expertise is something, is, like, the most valuable thing that localization specialist brings to the table.
And, it is worth, listening to it when a translator suggests that maybe some wording should be changed, maybe some appearances of characters should be changed, because that would just not resonate with with their culture, with the target audience in this country.
Mm-hmm. Yeah. So also I'm curious to know, Yana, what should a team look for when choosing a localization partner?
So it really depends on whether you are starting, your first localization project or if you have, for example, a portfolio of brands and you already have done some localizations internally and just want to find a vendor who will help you scale. These are, like, different things, different approaches. But if you're just starting, with the localization, I think you will need to look for a partner who is really flexible.
I know that when you start developing a product, costs and expenses, they mean really a lot, and you need to, maybe compromise sometimes. You need to find solutions to fit localization and localization budget in your, like, general budget. So in this case, your localization partner should be able to find a solution which would still provide you the quality you want, but maybe, apply some shortcuts like, introducing AI translation here or AI proofreading there just to make sure that, the costs, and the expenses are a bit lower and that they fit your expected budget.
Mm-hmm. Yes, and as we're drawing close to this episode,
Where to Learn More & Closing
where can people learn more about you and Alconost?
You can learn more about Alkonost at our website. It it has been revamped recently, so a lot of new information there. And you can, contact me at all times, on LinkedIn. I'm happy to answer your questions, anytime.
Yeah, and, all links of Alkonost website and Yana's LinkedIn will be found in the show notes, so you can easily check them out. All right, and yes, Yana, is there any closing words you'd like our audience to know about?
Well, I would say that, Doing localization together with, , designers is really an exciting opportunity for any localization specialist. So if you're just starting with localization, don't be afraid to ask localization specialists for help. Don't be afraid to invite them to review your designs to provide their perspective.
And don't be afraid to do localization and to plan for localization because, in the end, it's really a great opportunity for any digital product, for expansion, for getting more global audience. So just keep localization in mind and,
You will achieve the good results
Yeah. Yeah, no, it does feel like the earlier you think about it, the better the end result it will be.
Yes, that's correct. Yeah.
Yeah. All right. Thank you so much, Yana, for being here. It was wonderful talking with you.
Thank you so much for having me again, and, it was great to talk to you too.
Yes. Please do support our guests, and until then, you just listened to the UX Grow Podcast. I'm your host, Nick Mann. Thank you for listening.
Great work deserves
a better story.
Your portfolio should help people understand the decisions behind your work, not just the finished designs. Explore how UX Growth can help you tell that story more clearly.
