Amanda & Sebastian: The UX behind unforgettable experiences

All the AI stuff that's unblocked us is definitely exciting because we just have so much information about what we sell that we should be leveraging better. And so finding smart ways to really do that at scale is something that we were not really able to do at the same level of speed and scope. This is super exciting. - Hello, I'm your host Liam, and you're listening to Behind the Journey. On this show, we dive deep into the world of Get Your Guide by chatting with our leaders and innovators about their stories, ideas and strategies. If your new here Get Your Guide is a market leading gap dedicated to connecting millions of travelers with unforgettable experiences around the world. Ready to start a new career journey and shape the future of travel? You can visit GetYourGuide.corris. And today we're talking about something every travel of fields, but doesn't always have the language for and that's user experience. So how do we make it simple, clear and genuinely enjoyable to discover and book the right experiences? And what does that actually look like behind the scenes? And to unpack that, I'm joined by two people who shape how our product works and feels every day. Sebastian Schnauf, research team lead and Amanda Mullenhoff, senior manager, consent design. Sebastian, Amanda, welcome to the show. Maybe just to start with, be great to hear a quick introduction to you both and your roles. Yeah, I'm Amanda and I am a Get Your Guide Boomerang. So I worked here from 2018 until the end of 2021. I had a bunch of different leadership roles throughout the UX team over the years, but I was initially hired to found the content design discipline here. And now after many twists and turns of life and a few years, actually away from Germany, I am back and leading content design. Once again, with a lot more focus, I currently have eight content designers on my team. They are absolutely fantastic. Yeah, we're in every single part of the business. It's awesome. My name is Sebastian. I lead the user experience research team here at Get Your Guide, joined three years ago. And it's been a fantastic journey. I actually started my career in the field of user experience a long while ago as a designer. During that time, that was called a web designer. And broke for various companies, then studied social science and after a postdoc, I was at another startup, a health care startup. And then 11 years at Google, where I had a chance to work across many different product areas, like Gmail, YouTube, and Waymo, shopping. And during that time, I also did a business degree, which then made me really eager to join an exciting startup. Like Get Your Guide. A great stuff. Thank you both for joining. We want to appreciate it. So let's start maybe by grounding everyone in the basics. So user experience of UX increasingly seen as critical to business success, but it can mean very different things, from one company to the next. So when you're talking about user experience, it gets your guys. What do you actually-- we think of user experience to get your guide as a longer journey, taking place over multiple weeks, starts with dreaming. So this may be a random conversation with a friend in a bar about your last vacation. It's really about exploring stuff and not having a specific kind of trip in mind and where you go and where you may stay. And then the planning is a bit more detailed. And planning of activities has different states, starting with an overview of what can be seen and done in a destination, exploring how activities fit into a place, narrowing down and deciding on a specific option. Then booking, right? This involves pricing, payment, and flexibility options, anticipating. So before the activity, people may need timely instructions like to bring along your sunglasses and getting reminders. And a popular last minute request is really, can I bring my other friend, which you probably realize can be a challenge for our suppliers, just to switch the site here and then also take the supplier into that journey. Then on trip is about people in the destination, when they actually do the experience and people may take pictures and them to Granny, right? And after the trip, people want to share reviews, memories with a friend in a bar. So you can see this entire journey, we are actually thinking of that as a cycle. And it's even more true when we look at it from a supplier perspective, because a supplier has, once you have finished your unique tour, he will go back to operations which are much more happening in a cyclical way. Yeah, just to jump in here and to clarify what we mean by supplier. We mean a tour operator, somebody who's running a business, selling guide tours, or excursion, state trips, even somebody with a boat who's doing boat cruises. That's a many of them. Yeah, in a specific place. So these are people who are like, have a specialty in not only entrepreneurship, but also a deep understanding of the place that they're representing and bringing people into. And that's definitely a very different user to design for than the average e-commerce traveler that we're building user experiences for. Also, I just wanted to say it's a journey. It's also like when you asked initially, how do we think of user experience to get your guide? And I think part of this is like, are we able to meet customers where they're at in that journey? Their needs are different. If they're just exploring, they're trying to find out what they can do in a place. They're trying to get inspired. How are we serving that need versus, I know exactly what I want to book. I know where I want to go. I want to take it to the Eiffel Tower. Get me a ticket so that I can skip the line and enjoy my day. So how can we make it easy for them, even delightful for them to do these things and also make it feel a bit more like someone has their back. Travel can be scary or out of your element. To your, it's emotional, it's high stakes, right? You want to make sure you're using your time well and our job is to come in and help people really do that. - Great, thank you so much. And how does that definition that you both share there of UX? How does that show up day to day for your teams? Like what does a good UX decision look like in your world? - A good UX decision is based on sound evidence, not guesswork. I mean, it's great to have strong opinions and we also need to be thoughtful that needs and pain points evolve over time. So we can be a bit visionary in our decisions, but I would say a good UX decision is based on sound evidence and really rooted in real user needs and problems. Like how do you coordinate at this decision making across your best three friends, right? That's very, can involve a lot of back and forth. - That's challenging. - And maybe a second point is also like, but maybe Amanda can speak more to that later on is a clear and understandable design should have no surprises and no hidden friction, I think here of like prices that are initially discovered, right? They should be in line with the expectation once you're at checkout point. - Yeah, I mean, part of it like I totally agree and we also hire really smart people. We hire people who we trust to make judgment calls based on the data grounded in evidence to the best interests of the customer first, the business and that's pretty much it and the supplier I guess too. Like at the end of the day, everyone should be successful. That is, you know, transacting with us. That's the customer and the supplier, everybody. - Yeah, and maybe to add the supplier, right? We should also make decision where we can sort of predict a different future mental state. What I mean by that is really like, for a traveler, it may be like having amazing experiences and trying to remember that and treasure that but for a supplier, it may be to have peace of mind, maybe a different state and be sure that the operation within a supplier business is actually running without friction. So let's dive deeper into research. So from the outside, people might imagine usability tests interviews, but as a research lead, like what's the real mandate of your team? I get your guide. So the role of our team is to bring in the perspective of our customers, suppliers and other partners as a user of our product. And it's really going beyond just like making sure the things work or look nice. It's also very important that we are involved in the decision what we're at. actually building. To give you an example, before I joined Get Your Guide, I was in Turkey doing one of those iconic balloon rides there. The pickup was scheduled really early at five o'clock in the morning. We had to be in front of our hotel. We came out on time and we just saw a bus driving by and I'm like, "Okay, that's it. We missed the bus." And my son almost had tears, right? And I tried to contact the supplier and my phone didn't work. It was really painful. And about 10 minutes later, another bus showed up. And this was actually our bus, right? So the balloon ride was amazing. And what really stuck with me at the beginning of that journey, it's really critical that travelers get like very precise instructions where when to be in also reliable ways how to get help in these kind of situations. And now it's interesting. I'm three years in at Get Your Guide. And I feel like we've made a lot of progress around pickup logistics, which remains a challenge. It's even when you talk with people from Uber, like this kind of finding a pickup spot is still challenging. But the ability for travelers to get directly in touch with suppliers is much better now. And we're really part of that journey from a research point of youth identifying the problem in the first place and then sizing it and actually testing new versions and more precise instructions to ensure we can actually do a better job. I'm just going back to the topics that you covered, Seb as well. How does that play out like practically? How are you working? Say, for example, with product managers, designers, content designers in a typical or standard project that you might work on. Research starts very early. We are now working on the themes and testing also visions from a design perspective that are relevant in 12 to 18 months from now. That may be a surprise for some listeners. And we really employ a mix method approach. We usually first start exploring problem spaces from a qualitative point of view, really understanding, okay. What is it that people experience? And with that, we get to a more refined understanding and may actually size the problem. Like in the example I've brought up before about pickup, it was really like, okay, how many times are people trying to reach a supplier and not reaching somebody? That was actually, that is very important. And once we enter this kind of development phase with a product, we work much closer with teams like Amanda, with the designers and product management in a more iterative way where we test different versions and come up with and refine the solution until we are launching it. And in addition, we do like work with other teams for more kind of tracking of what the experience looks like for people. So we can monitor how much progress we're making, like interesting frameworks around measuring ease of use. So this is really going beyond just satisfaction or NPS. It's really about quantifying and being able to track over time the user experience and the quality of that people have were using our product. Amanda, what it's a turn to more on the content design. So for people who don't work closely with UX content might sound like, I don't know, nothing more than writing microcopy. How do you describe the role of content design? I get your guide. Yeah, I mean, it's all in the name, right? We're designing with content. We work alongside product designers to use a UX toolkit. The same best practices, the same heuristics, the same even research insights to make decisions about how best to structure information across not just one little UI text slot, not just one button label, although we may we may agonize over the text on a single button. It's been known to happen, but you know, these are impactful places. But really looking at how are we disclosing and organizing information across an entire flow, across an entire app experience? How does that tie in cross platform? How are we thinking about terminology? How are we controlling our vocabulary so that people can easily learn how to use our site and our app to be able to accomplish whatever the task is that they're trying to do. So if they're trying to contact support because or they're trying to contact the supplier because they're in the middle of the desert in Turkey and they can't find the right bus or if they're trying to make a decision about, oh, should I book this Northern Lights Husky Dog sled tour or this other one and how are they different? Finding ways to surface information where it counts in a way that matches people's mental model. How am I already thinking about this? Helps them make decisions faster. And also like can we do it in a way that on the surface level, the actual wording, the verbiage level, makes them easily understand, build trust that also communicates something about who we are, right? Like we have a pretty cool tone voice framework where we strive to be empowering and relatable speaking in natural language while still being concise. We do everything from the structure and scope of what is the change that we're trying to make in order to solve this user problem to the surface level of what are the actual words on. Yeah. Thank you. Do you have of if you can like can you share an example of maybe without going sort of too deep, but where changes in consent design like really did move the needle for travelers and for customers? Yeah. So like one example and this is an older example, but it's it's still something that we're applying learnings from was we had these long descriptions of inactivity where it would explain in the description like you're going to go here and then you're going to go here and then you'll do this and then we're going to stop for lunch and then you're going to be able to get out and take photos and whatever else. Think of like a day trip from Rome to several different little villages in Tuscany like maybe some vineyards and stuff. You might want to know where exactly am I going. You might have never been to Tuscany and so not really be able to visualize this in your mind's eye. And so one of the things that we did was we actually took the itinerary and built it into a structure that is a map and then a timeline that shows okay first you're going to go to Siena and while you're in Siena you're going to see Viscathedrial and this in this and then we're going to go over here for a gelato stop and then we're going to go etc etc in this way that made it it took the same information and just surfaced it in a different way that made it really easy for people to understand at a glance really easy to parse and also frankly like a good way to get people excited about where they're going what they're going to do while giving them kind of the lay of the land. Yeah yeah that's like interesting you picked it to because I actually did I don't know if it was two years ago from Katania then it was to Mount Etna and then this food experience somewhere but I got down the mountain. It's the best food I've ever had in my life. Anyway it's still honestly still sticks in my mind and I don't even like olives but do now. I could have a million of them. I want to come on to a different topic as well so we've had another episode of behind the journey which was looking at AI across the company so I don't want to kind of repeat that topic as such but I'd love to kind of zoom in on your world for like content design and research. You're both working in disciplines that are feeling AI very directly right now I would say so how is AI changing the way that you're working and how are you helping your teams and navigate the changes that are coming with that. I would say AI is affecting each step of the core research journey and maybe I start with with insights right. We used to have reports sitting somewhere on a drive and and we actually pioneered a chat-based system where our teams can retrieve insight by actually asking over a chat UI like oh I would love to understand Italy case a little bit better that you just mentioned and it really enables team to retrieve insights from our repository including the sources very important and also the context. Was this a study about pricing? Was it about itinerary building what you mentioned before Amanda so it's very important that we can ensure that whatever we retrieve is actually makes sense from a quality point of view then on the analytical side I would say that is very established you know to answer unstructured survey responses or reviews that's something that other lamps do very well and for a longer time already these tools I would say or quiet is. established what is very hot these days is obviously synthetic testing, meaning the setup we have right now that you are interviewing us. That's a typical setup of a user experience study that you cannot only make a synthetic creation of the moderator, but also of the audience, which may sound a bit creepy to some of you. I need to get used to this, but it's really amazing to see how that field evolves. I would have this really profound impact on how we do research and how we enable our team to answer the second part of your question. We really need to carve out time, play time. I think it's very important to use these tools. We see on the research side there are sort of off-the-shelf tools, and there is also a lot going on in terms of actually scripting what is now called e-vails in the tech world. It's really fascinating to see how that evolves. I was just recently listening to another podcast, and when I heard about e-vails, it's like, a computer science who was traditionally a very deterministic discipline. You put in zero and one, you process, and you get zero and one's on the other side. It feels like computer science is discovering qualitative signals, like open unstructured text, and you're starting to classify this and use this kind of human in the loop process to classify different open text responses. Your judgment as a basis to actually scale it. We use AI a lot to scale stuff. I think that's a bit of a direction. We are going orchestration of sub-agent system is really important, and that we are actually thinking that way, "Okay, what are workflows that we can standardize? How can we break it apart?" Which is a bit counterintuitive, right? Because AI, the public perception is like, "Oh, here is this superhuman who knows everything. We can feed 8,000 hours of videos, and this traveler knows everything." When in fact, it's actually the opposite. We need to unpack those different kind of elements that we want to learn about and actually create sub-agents or swarms that can evaluate different aspects to ensure we get the quality. I think we're all learning about this, and I'm sure you agree, but it's really amazing how we can be part of that process. I wonder what parallels the UC from the content design side? It's allowing us to solve problems that blocked us for years. This is amazing. We have this massive catalog of inventory that spans every type of tour, travel experience, adventure experience, multi-day trips, like in so many places around the world doing so many different things. It is in the past, it was really difficult for us to create user experiences that could adapt to the type of content that we needed to convey. It's tricky, right? We're selling experiences. The words, the descriptions, the photos, these things become really critical in helping people understand, "What am I getting myself into here? Is this right for me? Is it right for my family? Can I bring my kids, these kinds of things?" Our inventory was so massive. It's so massive. It's only gotten bigger since then. We really struggled to wrap our arms around like, "What? How do we surface this information? How do we organize it in a way that makes sense to people when there's so many edge cases?" It's basically just a lot of edge cases. What you see also now in the AI, now we're using AI to look at the catalog, look at free text data that we have, pull insights out of that. What are the most important things that a customer would need to know in order to compare this balloon tour to another balloon tour? Oh, it's probably the amount of time that you're actually spending in the air, which is mentioned here in this random free text section, buried in the description, whatever. Being able to bubble those things up at scale is a huge game changer for us. Also because of the collection of edge cases and just the massive breadth of our inventory, and when we're working with AI, it's kind of squirrely. Seb's talking about leaving time for play. I feel this so much in my team because sometimes, especially with Gen AI, generative output, where you're having an AI actually right for you based on something that it's found in the catalog, producing content, producing summaries of reviews, producing summaries of, how does this compare to that or what should I expect from this activity? They will hallucinate strange things. It's so funny because it's just nonsensical. It usually takes a few tries, a few evils, maybe like a couple different eyes on a prompt in order to get it right, to be able to produce the content that you're looking for, or to be able to parse the content in a way that gives you some meaningful output. In the meantime, it's just funny. I can give an example if that's okay. Yeah, of course. For example, we were working on Caitlyn, a content designer on my team who's working on growth teams. She's working on how to organically attract new visitors to the site, basically. They were working on, "Okay, what if we have these landing pages?" We have AI-created summaries of, basically, content from reviews that's packaged into a single sentence that can explain something about a destination. I think what they were hoping for were things like, "Oh, Berlin is a really historical, interesting city with this kind of nightlife culture and this and that." Instead, what it spat out was in multiple languages. It would be one poll quote in English. Customers say Berlin has great guides, which is not really a proof point for Berlin itself. Berlin has great guides. Then it would be something in French. Also, they wanted more Brazilian music. It was so nonsensical. It was a Friday afternoon. We were all tapped out from a heavy week. We were just absolutely cracking up about this. I'll show it. Yeah, I need to hear it. When you look at your teams together, the content designer's research is, "What have you learned about getting people genuinely excited about AI?" Not just compliant with the new tool, but really embracing what's happening. This is my favorite topic right now because I was so grouchy about AI until last year. I'm a writer, right? By trade. I grew up as a UX writer. I worked at Amazon. I came to get your guide. It started doing content design here. Worked at a few other places. This has been my life for a long time. With LLMs, there's a lot of maybe perceived threat. Is this going to replace my job? This seems scary. I don't have time to learn this. These are the kinds of things that I was thinking. It wasn't until I started really playing around. You hit the nail on the head before I said, "Really? I had to be playful with it in order to actually shift my mindset around it and then start to see what are all of the capabilities. Once I discovered all the things that it could do, this totally changed my whole mindset around it." One other thing that I often talk about with regard to getting people excited about AI is don't stay in your lane. Go vibe code something. There are all kinds of apps and stuff to vibe code things like music. Go make the pop song of your dreams about your neighbor that has really heavy footfalls on top of you early in the morning, whatever. Just for your own enjoyment, whatever it is, go make mischief with it. I can add, in addition to providing a space to play around, we also share a lot of best practices. In our team, we had a recently assessed where we would upload a couple of reports on notebook ML and generate like a presentation. It was very much like, "Can you spot the Hall of Cination?" It was a really interesting discussion that we had because our designers now know a lot about the complexity behind our products. It was interesting to see like, "Okay, here it was really a shortcut to tell a good story and the reality looks much more complicated." I think sharing best practices, where it works, making time for it. And really, I would say use it or lose it because I mentioned before that lots of tools now popping up in the research space, like enabling you to run like, I don't know, 150 quality interviews overnight and so forth. So they're like really big promises, but we're testing all those tools and just like simple things like turn taking. How is an AI moderator doing the turn taking versus like a human? There's so much to learn here from that tool. So it's really about unpacking the black box and make it more glass box. One other thing that I think is really cool that the team is doing is like, they're just vibe coding. Whatever will make their workflows easier. So like on my team, we have some outdated tooling right now for managing localization strings. A lot of companies deal with this at the moment where on the brink of upgrading that system, but for now, we're kind of stuck with this tool that's super clunky. And one thing that really was bothering Annie, a content designer on my team who's based in Zurich was that the search would always break. Like it would always time out error out. There was something going on there where like it just wasn't working well. And she was so frustrated because as a content designer, you kind of need to know like, OK, I know I need to change this string. I'll just go into the backend and I'll change it. And then it will machine translate and then it will go into the human review queue and it will get updated on the site, whatever. If you can't find the string, you can't do your job. And so she was like at her with end. And so she just jumped in and started like talking to Claude and vibe coded a whole new search experience in data bricks that now engineers are using because they were also annoyed that this search would constantly time out and not work. And now she's unblocked like the entire tech team. And it was a week of work. We had a AI week was really a super. So everyone was like, it was way more fun than I thought it would be. And I totally agree. I was like, oh yeah, this is hype, whatever. And then I got in there and it was like, build your own anything. Solve any user problem. Here's FigmaMake. Here's cursor. All these tools, lovable, go. I am so impressed with where people built. Thanks so much. And let's talk about careers for a moment because a big part of this podcast is speaking to people who might want to join, get your guide. So when you're hiring for you, controls, whether it's research, content design, what are you each looking for in the people that are coming into the company? I'm not sure if it's my own bias, but I tend to pay more attention to the CV. Like very boring kind of what are the main places a person has been to and would roll and how did they progress during these times. So I think, I don't know, if it's an effect of AI, but I really pay more attention to the CV, then obviously if somebody enters our pipeline of interviews and the exercise, it's very important that a candidate brings up very concrete examples. That's usually the place where we really see somebody has multiple years of experience. This is an example, anyone analyzing lock data knows that you need to have a good overview, cleaning outliers and look at distributions before you come up with any fancy stuff like Max Day for something like this. So I pay a lot of attention to the craft and for me it's really important that this comes through into you with very tangible examples because yeah, ambiguity is a reality and aside from the craft, I think even earlier relationship building, how is somebody starting and creating and developing relationships with Mainstay Colors is a really important part. And again, I want to hear very credible examples and also some creativity because the research we can also put it in a shelf, it's interesting, but activation is a big part of research. And so I want to hear like really creative idea how to make insight stick within an organization and that includes also some reflection at the very end. Like how do they deal with setbacks because that makes us all human. We are in a work environment where it's quite dynamic and therefore learning about setbacks is really important as well. Yeah, I mean, plus one to all of the behavioral things that said brought up, like the collaboration is key, we're working in a highly cross-functional setup. So like we've got mission teams with a number of engineers, a product designer, a PM, a designer's are split across multiple mission teams, which means they might be supporting, you know, 40 engineers or something. And when you're a content designer, having strong relationships with engineers is actually really helpful because sometimes you need to find a string ID or your not sure what the logic is on something or they need something from you. We hire people who are curious, who are low ego, this is so nice. People who are self reflective, you brought this up to sub like really are you able to take a setback, learn from your experience and then integrate that in a meaningful way into however you move forward. Whether that's like at the same in the same project at the same company or somewhere else. So like for me with content design specifically, I'm also looking at craft like if your CV or your, you know, cover letter is riddled with typographical errors, it's just going to get screened out immediately. If you have worked on content systems, this is something that's very interesting right now in the age of AI because that's even if you weren't working with AI, creating a content system of patterns and guidelines across an entire user experience is the basis of being able to automate stuff. And so if you're able to think in systems, this already definitely sets you apart. Thanks very much. And for people who maybe already in UX, maybe listening from other companies or even other destinations than Berlin, what makes this moment a get your guide particularly interesting for your disciplines? I mean, I have all the AI stuff that's unblocked as. It's definitely exciting because we just have so much information about what we sell that we should be leveraging better. And so finding smart ways to really do that at scale is something that we were not really able to do to the same at the same level of speed and scope. Yeah, I agree. Maybe one example, we used to spend a lot of time retrieving information. You're like, oh yeah, two years ago, I had this really interesting interview with this case. We used to spend hours digging for that data and the chat system that I mentioned before is really helping us to be much faster and have more time for the more exciting stuff. Maybe one more thing I want to mention, what makes it very interesting, we're really creating a new product experience or a new product category. And it has a lot of service components. And I think that makes it very unique. It's not a standardized e-commerce product. We have a lot of service elements and it's a very complex space so that is something that is really exciting. We're so much to do. Apply now. Apply now. That's the message. Keeping me busy. Before we wrap up, so we always finish behind the journey episode with the topic that we do love the most, which is obviously unforgettable travel experiences. So I'm going to ask you both. You can go separately, set what start with you. Most memorable travel experience and Amanda will come on to you a second. I'm not sure if it's my most memorable experience, but it's one that I recall because fairly recently I was in southern Italy again. What a coincidence. I was also in Lecce. I did a food tour in Lecce. I was running late. My kids had to return a rental car again. It was a bit stressed to find a meeting point. But the guide was really welcoming and it was a really great group, a small group. We had a really good tour. Then in the final stop, I ended up chatting with another guest of that small group and it turned out to be this couple who made a road trip in Italy, seeking out smaller, less obvious places, like not the Colosseum, right? She said, "Well, I found in each place, in each town, I've been in a bulia, something to do with Get Your Guide." I didn't want town like. Posa class in another an architecture tour. And then this four tour is like really for me was really amazing because I remember when I started the inventory was a time of locker for people to use it outside of those bigger tourism hubs and and I think for me was really great to see that how selection what we offer and the good matching are just something really really important and for turning a destination into a longer memory. So cool. I'm going to Italy next week so I'll be right back but to answer your question, after I left get your guide the first time I was feeling like I needed to do something to like I felt like I was taking a big risk like I had this like job that I liked for a long time but I was really tired after the pandemic and I needed to go home and take care of some family stuff and I wanted to do something before I left Europe that was like gonna really help me feel like I could take risks and so I went to Chamonix in France in the French Alps and I jumped off the side of Mont Blanc. Yeah I went paragliding which like is an experience we offer with a pilot so basically you're like strapped to another person and you run off the side of Mont Blanc and you fly down into the valley and this was such a good metaphor in my body for the risk that I was taking and the fact that like it was gonna be okay and in the end it was definitely okay. I ended it back here again. Amazing. Yeah it was really cool. Brilliant. Sir but Manda thank you both so much for taking the time and taking us behind the scenes of how you work research content design shape to get your guide experience and for showing us well how you're navigating the next wave of change especially with AI in your areas and if you enjoyed this episode and you want to hear more stories from the people building the future of travel experiences you can discover more behind the journey episodes on all major streaming platforms and if you're interested in joining us whether you're a researcher content designer product design engineering visit getyourguide.ca to explore all our open roles thanks so much for listening and we'll see you next time on Behind the Journey.