Alyona Medelyan: 0:00

There's so much data around and technology to digest this data got so much better. Why, why why survey people? Instead? Why didn't we look at overall call center data all the customer chats, complaints, sales notes reviews.

Rick Denton: 0:22

You're listening to CX Passport, the show about creating great customer experiences with a dash of travel talk. Each episode we’ll talk with our guests about great CX, travel...and just like the best journeys, explore new directions we never anticipated. I'm your host Rick Denton. I believe the best meals are served outside and require a passport. Let's get going. Oh, I am so excited today. Yeah, okay, I know I'm really excited for almost every guest right? Today though, we get to do something new. A guest from a country we've not covered before in CX passport. It's a place I've always wanted to visit so today we'll get to do that vicariously as we talk with our guests. Alyona Medelyan, co founder and CEO of somatic talking to us from New Zealand. That's right, New Zealand, we finally get a guest from New Zealand and I'm extremely excited about that. Alright, so we know that AI continues to be the hot topic out there, I recognize that the term AI is quite general. And for many of us, we've been involved in AI for a while now bots, data sentiment analysis and others Yeah, for a bit now. Yet, for the most part, our mass cultural awareness of it came about to the amazing progress of generative AI over the last year or so. Well, I Leona has us all beat back in 1998. Let me say that again. 1998, she was already consulting in the general language processing subfield of AI. She's seen the evolution firsthand. Landing in the customer experience base about eight years ago, Alyona knows how to actually not with hype, but with actual results we've AI and customer experience together. Today's episode records when I'm at peak southern hemisphere jealousy as it is a solid winter here in North Texas, and a glorious summer in New Zealand man how I would want to chase summer each year by switching hemispheres. And guess what I really want to did just that. always a treat to talk with a Global Traveler. Alyona. Welcome to CX passport.

Alyona Medelyan: 2:39

Thank you. Thanks for having me.

Rick Denton: 2:41

Let's have some fun here today. So in that intro, I mentioned you've been in this AI world since the late 90s. With the release of chat,

Alyona Medelyan: 2:51

small correction, I haven't been consulting from 1998 I was still a student, but I bombard this field of AI. completely randomly in a maths lecture. I was studying languages to become a German and English interpreter.

Rick Denton: 3:08

Oh, okay. Oh, wow. This is really different that can

Alyona Medelyan: 3:12

Ukraine Yes. But applied math was one of the subjects we had to take. And the professor said, Oh, by the way, there's also people who use computers to translate text automatically. So I was like, why am I studying than this? This is all going to be fixed within years. And I want those people working on the problems of building those algorithms.

Rick Denton: 3:38

Well, I'm glad that you corrected me and I tend to, you know, take good care with my introductions. I am embarrassed that I made a gap there. But I love that that gap has resulted in something I didn't even know as we talked before, of this origin story of involved in learning languages and all of that before even getting involved in something. And I by the way, I chuckle at the end, it'll be a couple years till we just have computers do all the translation for us. Well, if that was your exposure back there at university in Ukraine, you've seen this evolution. And you probably even chuckled at folks like us who think that, you know, you hear the AI, air quotes, for those of you that are just listening experts that are out there. You've been doing this for decades. So talk to me about those early days of AI, and what was it like then and then how it's evolved to where we are today?

Alyona Medelyan: 4:25

Yeah, so back. In the days people have been working on using computers to understand language from the moment computers existed. So it's, it's been, I guess, soon, close to a century. But when I started studying in this field, people were trying to figure out how can we gather as much text as we can, that is that a computer can read through within a reasonable time. And from the stacks How can we infer language ideas? I'd like to share this really fun example. So there was this notion of patterns, like a speech pattern, for example, banana, apple and other fruit. Right? So, and other could be a speech pattern to recognize that an author wrote a book called lawyers and other reptiles. So this whole thing fell out of the window.

Rick Denton: 5:32

Because it's it, attributing that oh my gosh, that was so you've seen that those early days? It's obviously a lot different. Now. Talk to me about that evolution. What do you say now?

Alyona Medelyan: 5:44

Well, it's it all became more and more statistical. So those early patterns were first examples of how can we use How can we scan through a lot of text and figure out the word meanings and it then it has then has moved into something we call word embeddings. It's where a every word can be represented as a vector in a multi dimensional vector space. And then you can do things like King minus man, plus woman equals, and the computer tells you, Oh, it's a queen. Okay, something that basically gives us an idea of that, actually, computer understands the meaning of the words King, man and woman, and can make these inferences and with Chet GPT, just all became even more advanced. And now GPT, or, you know, the large language models, can analyze texts and create texts better than many humans can, which I have not expected to happen in my lifetime once and neither did a lot of my colleagues who have PhDs and even the guy who came up with the whole deep learning, there were just people who kept throwing more and more data and throwing more and more computing resources. And even they were like, like, we thought something cool will happen. But this has surpassed our expectations and how amazing it is. This

Your CX Passport Captain: 7:18

This is your captain speaking. I want to thank you for listening to CX Passport today. We’ve now reached our cruising altitude so I’ll turn that seatbelt sign off. <ding> While you’re getting comfortable, hit that Follow or Subscribe button in your favorite podcast app so you never miss an episode. I’d love it if you’d tell a friend about CX Passport and leave a review so that others can discover the show as well. Now, sit back and enjoy the rest of the episode.

Rick Denton: 7:44

I tried to it would be impossible, but it's almost if we could have the late 90s. You and the current you see what it is today. And just watch the shock and the the amazement of what was expected back in the late 90s. And what we're seeing now, you you've made that move over to customer experience. And I know that that was part of your reasoning for starting somatic several years ago, I amplify that a little bit more what inspired you to start the Matic? How was it going to solve some of the challenges that you were seeing out there?

Alyona Medelyan: 8:18

Well, I wanted to, I knew that the skills that I gained at university, and working at a few different companies, including Google, were really, really useful for companies but I wasn't sure what is the problem we're solving. And so I started consulting in this general area of NLP machine learning people asked me to create all sorts of different applications for different use cases. And then within a year, several companies reached out and they had the same problem. They wanted to understand what is driving their net promoter score. So this was an this was the moment where I dropped everything and started working on this problem, because I realized that it's, it's an area where both the customers benefit from having their voice heard. The companies and the companies who are listening to their customers will also benefit by building something more useful and growing faster. So I felt like it's a it's a great thing to work on. And I haven't stopped and it's been now eight years. Yeah,

Rick Denton: 9:24

well, so that was the inspiration for it. How How was like at your very beginning, when you were thinking of this? How was the medic going to solve some of those challenges that you were seeing?

Alyona Medelyan: 9:35

Well, I applied, were diabetics to start with to try to figure out okay, so one person might say, your service is really expensive, and another person might say, your service costs too much. And it's very critical to realize that these two people say the same thing. And when you use language model you can do do it and you can kind of group the feedback. And then once you can group it, organize it into a taxonomy, you can quantify it. And then you can present statistics about what's driving the NPS, and what people should be focusing on in a way that is digestible. And people will then make decisions based on this data.

Rick Denton: 10:22

I want to come back to that decisions based on the data. Before we get there, though, one of the things that you shared with me when we were getting to know each other and exploring being on six passport is, you talked about your experience working for a tech startup in the past. And you've shared with me that they raised a bunch of money, and they they raised it by focusing on tech, and then quite focused on solving customer problems, which is you're describing dramatic, you found an issue you started solving for that. It also kind of reminds me of how companies today go out there and just buy technology thinking it will solve everything without thinking about how it's going to help customers or employees or perhaps even hinder customers or employees. So how are you going to avoid that particular risk with the Matic, the idea of this is technology. But how are you helping to bring it to where it's actually addressing customer and employee needs?

Alyona Medelyan: 11:16

Well, I'm a huge proponent of content and writing blog, blog posts about things that people are interested in. And what this means is that people come to us rather than go seek out and try to sell them something that we have, when people come to us, we talk to them, and we try to understand what is the problem they're trying to solve? And we always offer a trial. So you can upload your data and see what it returns. Is this useful to you? Does it provide you with things that you didn't know that were in your data? So this is how we personally try to avoid building things that people don't, don't need? Because it's very easy. And we try to make this into like a company culture. Because especially in the field of AI, it's so easy to get distracted by cool technology, a cool demo, it can do this, it can do this, or ultimately, what is in it for for the customers. And for people who need this from these problems solved.

Rick Denton: 12:19

I chuckled when you mentioned the cool demo, I believe was it Google's AI demonstration, there was something very recently where they were they had put their cool demo on stage. And then it was revealed that a lot of it was not actually live demo it was calculated and sort of scripted in the background, or you weren't actually testing the engine, but rather it was responses that had been pre programmed. So you're right, that cool demo that that technology focus unless you're there solving for the customer's needs, then I can see how that might have a shallow or short lifespan. You had mentioned something to me earlier. And it was this phrase that is stuck with me. And I think it's related to what I'd said we talked about a little bit later. But you you talked about the democratization of insights, would you amplify that for me? Why is the democratization of insights important in a business?

Alyona Medelyan: 13:16

So what I mean by democratization of insights is making sure that everybody in the company who who is making decisions can access the data required for that and, and has the support they need to be able to interpret interpret the data. Today, this is easier than ever, because companies like ours, have created natural language interfaces where you can simply ask a question, and the data summarizes you the answer to your question. So you don't even need to be a an analyst or an expert to have the data to support your decisions. So if what we've noticed is that among our customers, those who don't have a centralized function for for insights, but instead what they still have the insights function, but that function basically works on to providing as many people with insights as possible, those companies grow faster and are more successful. And so I believe that the less barriers are there, that we remove between the data and people making the decision, the more effective will be their decision, rather than using gut feel and saying, Well, I spoke to this customer, and that's my most recent right of nugget that I will no influence my roadmap, right? It's very easy to kind of be being swayed by the loudest voice, but ultimately, you need to look at the data and do the bigger picture. Yeah, the

Rick Denton: 14:53

risk of the loudest the and even in some cases, perhaps most frequent isn't necessarily the same as the most more impactful can be written. What about the the other side? By the way? I'm totally aligned with you. So let me before I even ask this question, know that I, I like the idea of providing more information out there. Have you seen it be a challenge for a company with this democratization of insights, perhaps information overload, that there's so much available that because insights are no longer this closely held secret and accompany that it can overwhelm departments or teams that aren't accustomed to really absorbing that level of information? How can a company make sure that they balance here's all of our insights with here is just a tiny few that we think you need?

Alyona Medelyan: 15:41

Well, I think the decisions that they're making every day should be driving what information they should be getting. Or they should be requesting, right, it needs to come from them rather than here's data, data, data data. Like I agree that too much data is not a not a good thing. And you will basically waste time trying to make sense of it. But if you are actually need to decide between A, B and C, everybody has a limited budget, right? Like we can't do everything we need to decide and as you as you say, impact on growth, NPs CSAT.

Rick Denton: 16:27

It really captured me there at the end, you may have seen the AI is kind of getting wide or not. It's it is that it's not that there's just this firehose of democratization of insights, it's actually in the sense of Be thoughtful as to what business decision, you said that, you know, what is the business decision that you're making? And then how, what information do you need to help support that business decision, the light bulb went off in my eyes, or in my eyes, the light bulb went off in my head. And I have a much better understanding of the democratization of insights. Now, I want to I'm talking to somebody from across the globe, right? So I want to mix this up a little bit. If I were to fly to New Zealand, and I believe there actually has, there will be nonstop flights from DFW to New Zealand, which shocks me. Well, if that's the case, that's gonna be a really, really long flight. And so stopping off in the first class lounge would be really nice. And so that's what I want to do with you here today. Is join me here in the first class lounge. We'll move quickly and have a little bit of fun here. What is a dream travel location from your past?

Alyona Medelyan: 17:35

For me, it was Hawaii. Okay, the most exotic location you can imagine from, from somebody being from Ukraine and being able to swim with turtles. Oh, yeah, natural habitat. To me. This was mind blowing, I still think about it.

Rick Denton: 17:53

I have had the opportunity to do a swimming with the turtles, if you will. And it wasn't something that was scripted. You're just in an area in the US Virgin Islands. And there's something so magical when you're just following them. And you see the the the elegance and almost the grace that they have underwater. I can imagine that that's a memory that has absolutely been tattooed with you in your brain of how delightful Hawaii is. Let's go the other way. What is a dream travel location you've not been to yet. Not

Alyona Medelyan: 18:19

all occasion, but I I still haven't seen whales. And one day I hope to be able to be somewhere on a boat and experiencing a whale in their natural habitat. I love animals. So to me this is something magical. When

Rick Denton: 18:35

you're you're certainly in a great place of the world for animals and you know, things that I might never be accustomed to you know, it's funny. Eliana when you first said Wales, I was thinking you were talking about in the UK? Sorry, my echo. No, no, it was not the accent at all. It's the same sounding word. And I'm thinking well, do you like the Wrexham Football Club is that but no, it's the seeing the the animal of whales, the whales. You know, I've never had that chance to do that. And I think that would be a lot of fun. And you actually you may be the first guest that has pivoted that question on not a location but a trip and I like that

Alyona Medelyan: 19:08

this experience and experience. Somebody is

Rick Denton: 19:12

in the customer experience space. I like that. Well Well played there. I like that. I like that a lot. Well, what is a favorite thing of yours to eat?

Alyona Medelyan: 19:23

Chocolate ice cream fruit. I'm I feel like the more you eat the same thing, the less you like that. So I think that diversity is important. And I also enjoy making cocktails and drinking There we go.

Rick Denton: 19:37

Okay, I knew you and I had some some very similar we both like customer experience and now we both like cocktails and travel. Absolutely. So you are walking. I'm gonna give you a little bonus first class lounge question here. You're walking into your favorite lounge. What is the cocktail you request?

Alyona Medelyan: 19:56

Either Clover Club or aviation. Okay, Hey, since were in in flight, we're almost in flight.

Rick Denton: 20:04

Oh yet I will have to. I will want to get the recipes for all of that so that we can we'll get that off the off the air here and find out what that is the other direction you love. There's lots of different foods. I imagine there's some things though, that you might have been forced to eat as a kid, but you just hated growing up. What were those?

Alyona Medelyan: 20:24

I had wonderful parents who thought that a child shouldn't be forced to eat anything but yeah, wow. But I still dislike onions even now and I never ate them. There's

Rick Denton: 20:38

no onions for you. Is that true if they're raw, cooked fried or is it just all onions in general?

Alyona Medelyan: 20:46

The more fried they are the better.

Rick Denton: 20:50

Yeah, you're you're you're you're absolutely talking to this Texans heart when you talk about fried food absolutely had me some chicken fried steak on on Saturday, actually. Let's go back to travel. As we exit the lounge here. What is one thing not including your phone, not including your passport that you will not leave home without?

Alyona Medelyan: 21:08

Headphones. When I travel, I think very useful to keep the noise out and feeling like you're in your space.

Rick Denton: 21:26

Well, I can completely agree with that answer. As I'm wearing the noise cancelling headphones that I was thankful to get when I was consulting with a a well known audio brand Bose and I absolutely getting to shut out the plane is quite nice when you've got those going. One of the things about thematic and it's very relevant in kind of a lot of the work discussions and style of work discussions that we're having as a as a business community today. And one thing about the Matic it's not about work from home or work from the office, it's exclusively remote. Now, I admire and actually kind of envy your approach of living permanently in the summer by moving around the world. So making sure that you're living in summer, how does whether it's moving for the hemispheres or just in general, with your exposure to all sorts of parts of the world? How has that influenced your approach to customer and employee experience?

Alyona Medelyan: 22:27

Yeah, I think any trouble builds empathy. When we travel, we meet people from different cultures, people who have a variety of issues that are maybe different, but then we see how many of them are the same. And we ended up connecting with different cultures. And in CX, and even in any any company would be so critical, because this is empathy for customers is important. That's kind of what we're trying to do in the CX space. But ultimately, there's people in the company who need to do the right thing, by the customers may actually actually act on the insights that we provide to them, right. And having empathy about what they need in order to act on those things is important. So for example, a technical person is unlikely to respond to a single customer quote, they will straight away like, well, but how many customers are affecting, right, right? So you need to like understand they are, there's people who work differently from you. And this is what they require. A technical person might want to see the data and some sort of an estimation that when they work on this, it will impact that many customers. And this is why they should be working on this. So that's kind of how I think about the connection between travel and working in CX. Yeah,

Rick Denton: 23:59

and that spirit, I've, I'm being a little too glib here with this. But I often think that there would be a lot more, there'd be a lot lot less conflict in the world, if everybody would actually travel, and even amongst our own, inside our own countries or inside our own communities, and that sort of thing. If we would just be willing to expose ourselves to other thoughts, ideas and experiences, we might find ourselves with less conflict, I realized that's a little bit too late. But at least it's something that this this this guy can can hope for. Have you found I want to ask you a little deeper around because the employee element, especially in a company, like thematic, that has global employees, if I understand that correctly, you've got ways around the globe. And so there might be cultural differences. There might be motivational differences. There might be experiential differences. How How are you attacking it? And now let's not use that word. Let's how are you addressing that inside of thematic, recognizing, hey, we're one company one culture, yet we are built of people from many different experiences and many different cultures.

Alyona Medelyan: 25:01

Yeah, it's, it's tricky. And building that connection between people remotely is, I find it's very tricky and I don't think we've we've nailed it is there's just like benefits and of being remote for for both the company and the employee. But the disadvantage is it's much more difficult to build this connection, how we have been doing this, we actually take our team to an annual retreat. We actually have met in Hawaii. And when

Rick Denton: 25:31

I was gonna say hurdles to get to Hawaii,

Alyona Medelyan: 25:33

we did. We did two in New Zealand as well. We flew people over here. And for me cooking together is what kind of will try to mix people up. And they they cook recipes, and they cook, cook a meal and I cook dinner together. And I feel like this really builds empathy.

Rick Denton: 25:56

I'm frozen a little bit because what an absolutely genius idea from a corporate perspective, because we all have the foods that means something to us growing up. That's even why I asked the question, the first class Lounge, which is a favorite food, because there's something that there's emotional and it's a a symbolic element of who we are and what our culture is and what our experiences. And to use cooking. As a way to bring that together. I actually have a story from when I was an employee at a at a US Bank. This was decades ago, I couldn't tell you the exact year. But my manager actually invited I was traveling to a town in which he lived, invited my wife and me over to his house. And and we actually walked through the market. This was in Seattle, he got scholarships and other things for a meal. And we just gathered together in his house while he cooked. We talked we shared the meal the meal together, there's something really unifying and and I felt so connected to him. And certainly my employee engagement shot through the roof through that. And so that idea of cooking, what a brilliant idea who I like that. Yeah, keep it and you clearly have just there's a lot of folks listening to this going well, I'm definitely going to put in my application to automatic my annual trips to Hawaii or to New Zealand. Next time, you need to have a whale watching trip. Yeah,

Alyona Medelyan: 27:15

creating these unique experiences, right? It all comes back to like, how can we really create something unique that they haven't experienced before? And I love cooking food from from different countries. And

Rick Denton: 27:29

that really is, yeah, I want to ask you about because you've talked about empathy. A lot of what thematic is doing is listening to cause or helping your clients listen to customers, and then be able to discern the themes around that. Well, that requires knowing how to listen to the customer. And one of the things that I really value Bill Staikos is, if you don't know him, he's on LinkedIn, and he's out there for listeners, he's a great follow when it comes to customer experience inside he'll be on there has been on the show as well. He coined and frequently reminds us that we are in a post survey world. And so I'm curious as to how we're in this post survey world? How do you think about how we can understand what the customer thinks and what their desires are and their needs? In that post survey world?

Alyona Medelyan: 28:20

Well, I think what Bill means is that there's so much data around and technology to digest this data got so much better. Why why why survey people because when you survey people, only a certain type of person is likely to respond. And instead, why don't we look at overall call center data, all the customer chats, complaints, sales notes, reviews. And I think it's incredibly important. But I also think that some people just like being asked, and to service will become shorter, or already becoming shorter. You don't need to ask 25 Question rate this rate as we just what did you dislike, what can we improve? And then lets people actually say the specific things rather than needing a rating for everything. And similarly, analyze all of the call center data because a lot of people need to have their issues solved in more streamlined way. So just a more holistic view of all of the ways in which customers want to communicate and are communicating with us.

Rick Denton: 29:34

Huge fan of the contact center, right? There's so so much information not only the call transcripts or the email content, whatever the contact mechanism is speaking with the frontline agents themselves and providing a mechanism for the frontline agents to inform the company Hey, guess what this is what the experience really is. Here's where it's working. Well, here's where it's broken. I also like how and I push back on it a little bit. I still think there's good wouldn't be a place for some version of survey. What that means, though, is still certainly reduced, like you're describing. And I think there's an element of the survey is a way if I've got an experience that I got to communicate with a company, either if things went really well, or this is so off the rails, the survey gives me that opportunity to communicate that off the rails aspect of it. And if we go completely away from it, I wonder if customers will miss that opportunity like you're describing for customers like me who want to have that opportunity. Well, early on, I'm looking at the clock and we're out of time here. And that saddens me there's more that I'd want to ask you. I do want to ask you this, though, if customers wanted to if customers. If CX passport listeners wanted to get to know a little bit more about you and your philosophy around customer experience, or to get to know a little bit more about thematic, where should they turn?

Alyona Medelyan: 30:54

Well, they're welcome to connect with me on LinkedIn, I post regularly and I post a lot. And just Google my name, and you'll find me otherwise get the magic.com is our company's website. And there's a blog and useful resources around CX. And you can also request to trial.

Rick Denton: 31:13

Awesome. Well, I will get all that in the show notes. No need to Google or use perplexity or any of your other new tools that are out there. I'll have that link there available to you. I Liana, it really was a delight chatting with you today. You've you've certainly opened my eyes in some areas, and certainly like in things like democratization of insights and how that manifests. And you know, you broke the rules and I like it on what's your dream travel location you've not been to yet and I do hope that you get to see those whales very very soon in your future. It has been fun having you on the show today. Alyona thank you for being on CX passport.

Alyona Medelyan: 31:51

Thank you for having me. It was a pleasure. It was always always fun talking to Rick.

Rick Denton: 32:00

Thanks for joining us this week on CX Passport. If you liked today’s episode I have 3 quick next steps for you Click subscribe on the CX Passport youtube channel or your favorite podcast app Next leave a comment below the video or a review in your favorite podcast app so others can find and and enjoy CX Passport too Then, head over to cxpassport.com website for show notes and resources that can help you create tangible business results by delivering great customer experience. Until next time, I’m Rick Denton and I believe the best meals are served outside and require a passport.