Latest / Talk Commerce / Turbocharge E-Commerce Growth with AI: Secrets from NetElixir's Udayan Bose

Transcript

Turbocharge E-Commerce Growth with AI: Secrets from NetElixir's Udayan Bose · Talk Commerce

• Brent Peterson (00:03.374) Welcome to this episode of Talk Commerce. Today I have Udian from Net Elixir. Udian, go ahead, do an introduction for yourself. Tell us your day -to -day role and one of your passions in life. Udayan Bose (00:15.971) So wonderful to meet you, Brent. Thanks for having me on this podcast. Really excited. I am the founder and CEO of NetElecSair. Started the company along with my wife, literally out of our living rooms back in India in 2004. So it has been a little over 20 years now. And really living the dream. I mean, we have, from literally that day in the apartment, we have really scaled the company to be about a 150 people company in about three countries. working with over 150 customers at this point in time, all of them in the retail e -commerce space primarily. A bit of a trivia or fun fact about myself, I guess, is I really devour books. I read a lot, Brent. And one of the areas that I'm particularly interested is the world history part. At this point in time, I am really reading this very unique book called the Kohinoor written about when India was at that point in time ruled by the Britishers or the East India Company. And some of the facts that I am really sort of getting to know is just simply mind boggling. And again, a big believer that there is a lot which happens and which you can really find out there is a lot of knowledge in history. And because if you really look at humankind, the entire pattern is very cyclical. We can do different things and think about it in different way. But fundamentally, it's my belief that things go in cycles. And at this point in time, I think that's what I'm very fascinated about, reading those world history books and so on and so forth. Brent Peterson (01:50.862) Yeah, that's interesting. My major in university was history. And it is very true that it does nothing. It's repeated over and over again. And sometimes even the horrible things that happen in our history, we forget and it's done again, which is unfortunate. But it does help us to understand some of the new things that are happening based on that history. And the other comment I'll make is on India and how diverse it is and how interesting it is to Udayan Bose (01:55.875) wow. Udayan Bose (02:02.627) Yeah. Udayan Bose (02:15.683) Mm -hmm. Brent Peterson (02:20.334) for each region and each region had a king at one point and how they were all brought together. And I've been to India many times. I very much enjoyed my visits to there. Before we get started though, you have volunteered and I should have had a history joke. You have volunteered to be part of the free joke project. I'm gonna tell you a joke and all you have to do is say, or all you have to do is give us a rating of one through five on the joke. Great, so here we go. And it's a good one today. Udayan Bose (02:44.227) Sure, happy to do that. Brent Peterson (02:49.07) I love going outdoors. It's so much safer than going out windows. Udayan Bose (02:55.459) What a PJ! I would give it a 4 out of 5. Or is it out of 10 or 5? High five! Okay. Brent Peterson (02:58.894) Okay, it's five, yeah. Well, that's good because a lot of people would have given it a four out of 10. I know I preempted you with it's a good joke, but that was just to kind of set the stage. All right, well, let's. Udayan Bose (03:11.395) Sure, of course. Thankfully, my daughter was not listening to the joke because she is being a typical classic teenager. She would have probably a slightly different perspective. But anyway. Brent Peterson (03:22.542) Yes, I roll. All right, so I do want to talk about digital marketing today and how that has changed. And especially, I would like to kind of dive into the subject of how AI is changing digital marketing. So tell us a little bit about your background and get us kind of how you got to where we're at today. Udayan Bose (03:47.363) Sure, absolutely. So my journey in the digital space brand started in 2001. I know I am significantly dating myself. where I was invited by two of my engineering school mates. We were still in India at that point in time. And they had started this online gaming company. First, I didn't really... I mean, my brush with online Brent at that point in time was checking my Hotmail account twice a month. So this was like 2001, I mean... And then online gaming, I didn't even know as to what online gaming is, but in the spirit of we'll figure it out, dived in, and on day one, they said that... Hey, by the way, you are joining as a part of the founding team. Can you really start the bingo business for us? I didn't have any clue of what is the online bingo business, but that was my brush into the internet part. We did extremely well, Brent. We built... what eventually became one of the largest online gaming companies in the world. And one of our brands, Party Poker, was extremely well known. In fact, in one point in time, I think 2005, I had moved on by then to start Netflix. The company did a pretty significant IPO. The company was valued... more than Google back in 2005, essentially, did well, I mean, overall. I think the big learning from that first part, and I just keep on getting back to this theme of learning Brent, primarily because just purely as a person, I'm just all for us to, am I really learning every day? Am I really learning every moment and so on? And my big learning there was even though I didn't know anything about online or anything about bingo, Udayan Bose (05:31.523) Just the willingness to dive in and figure things out and suddenly sort of you really have to obviously hustle hard but things slowly come in place and when it comes in place the impact is quite magical. I think that was my big learning from that part. And that was, I think, the inspiration for starting NetElec, so that can we really build a company which is steeped in three things. One, we are pretty strong on the technology side. So technology and mathematics was the core. The second is we really wanted to build a company culture based on fail faster. And... Improving, improving, improving. Every day had to be that entire fail faster part. And the third part was we wanted to build a company which was truly a global company diverse in every sense. So for example, Brent, I mean, we hope to host you sometime at our office in Princeton. It's a small group, about 40 people, but we... trace routes back to about 17 different countries, right? We probably would be one of the most diverse companies that you can think of, at least in the United States, if not, just given our size effectively, right? So, total about 150 people, 40 are in the US and I mean, it's a, so those three things, the diversity, the analytics plus technology. and constantly keep on testing, experiment and fail faster and then sort of move on really has helped us to survive and thrive for over 20 years now. So that I think has been the success formula essentially. Always be learning. Brent Peterson (07:04.43) That's good. I mean, I think failing faster is such a good moniker for every person, especially in the marketing realm, maybe not in the athletics realm. Although if you look at some of our professional sports teams, they're really good at failing faster. But in the digital marketing space, failing faster means that you've learned something about why it failed and you can do the next one better. Give us some background about Udayan Bose (07:20.579) Yeah. Udayan Bose (07:32.579) Yeah. Brent Peterson (07:34.03) Why you focused on feeling faster? Udayan Bose (07:38.115) Again, I have to trace that roots back to party bingo or the bingo business that I built. I didn't know as much, right? So, I mean, I was starting with a clean slate, which was exciting because I could really test things and come up with the hypothesis, et cetera, et cetera. And there were relentless failures, probably about fail 20 times and get one win. But that one win gave you so much happiness. It's really even difficult for me to express in words. I think that was, I think, the key component. And over the years when we are really building Netflix, I mean, there have been like any company which has been around for over 20 years in the digital space, which, as I think all of us know, is arguably the most dynamic space which exists in the world or has existed for the last 20 years. That's the only way of moving forward. I mean, it was so interesting. When we had started off, there were two different schools of thought when it came to paper click advertising, as it used to be called back then. 2004 is when I'm talking about. One was those paper click bid rule driven tools, which were there. And I don't even remember their names, but these tools effectively had a certain number of rules. It was basically a role engine. And... Effectively, all of them were gone by 2007, 2008. We had the Kenshoo and Marine, and they had the next one. Efficient Frontier came before that, and that is gone now as well with AI. The second school was basically the entire school of campaign management. But unfortunately, I think they didn't really focus as much on the industry nuances and learning the industry and internalizing themselves as a part of your industry or as your team. As a result of which, I mean, even at Party Gaming, we were spending quite heavily. We were actually one of the largest advertisers or spenders effectively for, back then it was Overture print. I don't know if you recall or remember this Overture which became Yahoo and so on, and then Google Ads. We were actually one of the top 10 at any point in time in terms of how much we were spending. However, the agency that was working with us was... Udayan Bose (09:48.995) hell went on us using the same formula for campaign management as probably any company in the retail space or financial services space or legal services space. So I think that was the big idea that you really have to understand the domain of the customer and really understand that success cannot only be just you're being an expert in terms of campaign management or channel. you really have to understand the overall industry. And that was, I think, a big driver as to why we decided to focus on the e -commerce. Initially, it was retail e -commerce. Then we added the direct to consumer side. We are pretty big now in the B2B e -commerce as well. But everything has hovered around the e -commerce part as well. So now we are getting into health care. It has been about a year now. And we are seeing remarkable success because health and wellness again has been really, really sort of seeing a massive explosion in terms of growth when it comes to the e -commerce part as well. So we have stuck to the that e -commerce part and understood the nuances of the shopping cart as to overall the visitor flow. We have really understood as to what are the different nudges that we really have to make to the to the consumers. Why do the consumers buy? What are the intents? How do they evolve etc. It's a lot of work in terms of research and lot of failing faster as well because if you don't really do those tests, it can be only a theoretical knowledge. I know so much about, let us say Brent as the consumer, but I do not know as to what would get him to really buy from my website, right. So, it has to be a very collaborative effort and that can be achieved in my opinion only through failing faster and not repeating the same mistake again, but moving on to the next test. Because if you are repeating the same mistake again, then there is no learning that you are just doing it for the sake of doing it rather than really improving. Brent Peterson (11:36.077) Yeah, do you feel as though every merchant should be looking at those numbers and many merchants don't? Udayan Bose (11:45.475) Yeah, yeah. I think it's... And first of all, I really have to give a bit of a... This thing that I understand that the merchants are under tremendous amount of pressure, right? So, the juggling, the day -to -day work is just crazy and in an environment which is as dynamic as what we are currently sort of going through. I mean, it is not easy at all. But my request or literally a request is probably the one I can think of is for every merchant to... at least reserve about 7 to 10 percent of their budget only for testing. But not in, again pardon my mentioning this, but dumb testing, but smart testing. And that's where there is an opportunity to do that. You really have to carve out that time. And the reason I mentioned this, and I'll give you a very logical reason why I am sort of mentioning this, the space that we are operating in is moving at a very quick pace. So, if you don't reserve that time to take a pause, do the right experiments, understand as to where your consumer is going. Just by sort of constantly running on the treadmill at 15 miles an hour will not really help, right. Because if the gym itself is moving, understand if the floor is moving, you really have to ensure take a pause sometime and say that, am I even running in the right place? And that is, I think, one of the big reasons why this entire fail faster mindset has to be coupled with taking those pauses and really reflecting upon as to where am I even standing? Where am I even running? Is this even a failure, which is a smart failure? Or is it a dumb failure? So those are the questions that become important. But that 10 % of the time reserved. for testing as well as just assessing as to where you are from your consumer's eyes becomes an extremely important part of any business at this point in time. Brent Peterson (13:44.622) Do you think that the proliferation now of machine learning in your space is leading to the tightening of margins across all the digital advertising and everything that has to do with it? Udayan Bose (13:57.699) like. Udayan Bose (14:02.179) Sure, absolutely. So let me give a bit of a context for the listeners on this. So effectively, we primarily work on, we significantly work on about six categories, Brent. The first one is the beauty e -commerce. The second is fashion and apparel e -commerce. The third one is home goods, home furnishings and so on. The fourth one is pet products, pet supplies and so on. The fifth one is consumer electronics and the last one is food and groceries. Now out of this consumer electronics partially, but the others are very fast moving, the high frequency purchase. Now in this high frequency purchase, we have seen... a tremendous or an exponential jump in the overall competitive pressure. And it's not only new brands coming in, but almost like mega competitors like Tmoo and Sheen coming in and really cutting the prices to the level which is seemingly almost impossible. And they are able to manage it primarily because it's a different business model that they are deploying. I don't want to spend time analyzing their business model. That business model is there. The challenge is for you as a business person, if you are competing in any of these categories, you are essentially competing with Sheen, Tmoo, along with Amazon and the Walmart system. So, you do not have the liberty to keep on increasing prices. You can, but obviously people build defect to the machines and the T -Mos because they can get things cheaper. Even the quality often can be suspect. Let us just put it that way. Udayan Bose (15:35.299) So that's, I think, one trend. The second trend, effectively, the advertising cost, which is, let us say, talk about the variable expenses, that keeps on increasing constantly. We were really analyzing, so we have been having this tracker brand since 2008, August is when we launched it. So it's been about 16 years now. And we have been tracking those cost per clicks on a very regular basis, the CPCs and so on. Now, just in the first four months of this year, I don't have the main numbers yet, but in the first four months of the year, there was a CPC inflation of close to about 17%, right? If you certain categories saw a CPC increase of almost about 32%. So the cost per click, the same Google ad, costs a lot more now. So just think of it. You are unable to increase your prices, and the variable cost keeps on increasing. And this is leading to a massive squeeze in the contribution margins. And that contribution margin gets sort of thinner and thinner, almost like massively squeezed. So within that part, I think what is the role of ML which is being played? I think at this point in time, we'll get to know, but I think it's just more of the competitive landscape, which is really influencing those contribution margin squeeze as such. The challenges also are posed by a very sophisticated or advanced level of advertisers or marketers as well, who have decided to really utilize ML or machine learning. even generative AI to create content and all this stuff. And these sophisticated marketers are probably in the top 1 % or maybe 0 .1 % of the market, essentially. I'm talking about companies like Amazon and so on. So they have set up very extensive AI -powered experimentation in -house labs. which are effectively able to really, at any point in time, probably a thousand experiments are running on different channels for Amazon, right? And it is being powered by highly curated or high value audiences, which is really driving those experiments as such. Now, they are trying to squeeze out not suddenly a 20 % gain brand, but they are trying to gain maybe in this channel X, campaign Y, let me get a 1 % increase. Udayan Bose (17:53.667) But remember, this is one percent increase for a company of the size of Amazon. We're talking about big numbers out here, right? Maybe a half a percent increase here, maybe a two percent increase here, and so on and so forth. And all of them then are getting aggregated together. So that's what we are seeing. So which is leading to two things. One, the CPCs as it keeps on going up, these large companies have really built these AI -powered experimentation platform, which are essentially keeping their costs down. So that's, I think, one trend, which is very clear. All of these three are really converging at a very fast pace. And we saw that convergence happen starting at around the holiday season, 2023 last year. And that really kept on increasing constantly. For example, I was invited to speak at a large investor conference of this Roth MKM Group, which is one of the larger mid cap funds out there. And I mentioned that one percent, I believe that, which is Tmoo's share of overall US e -commerce has gone from 0 to 1 percent, which is a significant number by the beginning of this year, so Jan 2024. And it is my belief that at this pace, there is a potential for it to go to 5 percent pretty quickly. Now, that company did not even exist about 18 months back or 20 months back, Brent. So it's just that things are happening at a very fast pace. So yeah, so again, a long answer to your question. But yes, it's a combination of AI ML application to do smarter experimentation. It is basically very aggressive competitors like T -Boo's and the Sheens along with Amazon's. And the third component is obviously the increasing of the spiraling costs of Google Ads CPCs or Meta Ads CPCs, which are really squeezing the contribution margins for businesses. Brent Peterson (19:47.086) Can you explain the difference to our listeners who look at AI and they just think about writing content for your blog as compared to machine learning and AI that's actually helping you be a better marketer? Udayan Bose (20:02.211) Right, right. So, Brent, I can literally sort of geek out on this topic for a day, but I'll just try to summarize it in three minutes, I think, for the audience. So if you think of artificial intelligence in very broad sense, think of two types of AIs, if not anything else. One is called a horizontal AI, right? And the second is a vertical AI. Now, the horizontal AI is things like chat GPT or let us say Google's Bard or Gemini and all that stuff, which is accessing publicly available data. And it is just trying to use the data as well as the overall models which started all mostly as neural networks and then really evolved into smarter and smarter models to really do the same thing. much quicker and much smarter, right? So it's basically doing a lot more for less. So this is effectively the horizontal one. The horizontal is essentially accessing and open access to the data, getting all of the stuff in. And their big thing is the productivity part and doing more for less or more with less. So again, what Brent is mentioning, for example, the easy content creation, for example, maybe a physical copywriter. Ideally, what we have been doing at our team is our content development team writes about top 10 copies, let us say if you're running a new Google ad campaign. And then we really utilize that to seed this horizontal AI, whether it is a Google Gemini and 1 .5, et cetera, et cetera, and get them to really pre -populate. So we are generating literally thousands of combinations in terms of ad Udayan Bose (21:37.269) copy and then you sort of pick and choose. So it's just a different scale that you can operate on with horizontal AI. The same applies for, let us say, productivity tools as well. For example, if this entire ad copywriting creation of 1 ,000 ad copies would have taken any expert copywriter probably about a couple of days, in this case, you are able to achieve not exactly to the same level of... impact but I think a fairly comparable I would say probably it has gone to about 90 -95 percent of that level within a fraction of the time. So it just we are able to do more for less on the horizontal part. The one that we have been championing as well and that's where we set up our own AI ML lab in 2017 at our office in India in a place called Hyderabad is it works with your data. So first of all, I think that's the first classification that horizontally is with open data and this works for you with your data. And with that data, it focuses on solving one very specific problem which pertains to you. Right. Now, Chatterjee period solves multiple problems, but there is nothing very specific. The tool was not really built to solve your problem. The tool was built for solving a wide range of problems in productivity primarily, which exists. Vertical AI, I think of it, is very specific to solving your problem. And the problem that, for example, our tool, Alexa Insight, solves when... We just wanted to solve a simple problem that the entire e -commerce world or digital marketing was focused on new customer acquisition. We wanted to change it marginally and say that, can we use machine learning to help you do new high value customer acquisition? Udayan Bose (23:26.723) So when I say high value customer acquisition, I mean, again, I must give credit to my online gaming days. It's almost like we walk into a casino. The high rollers are treated separately. So very similar to this. So we were able to build that machine learning model, which has the capability of identifying who is a high roller customer, who is likely to spend three times to 10 times more with your website and really use Google and Meta to acquire that type of customer, more lookalikes of that and so on and so forth. So that's the primary difference. Vertically, we are solving is one problem, helping you acquire through Google and Meta, the high value customers. One problem using your data. Horizontal AI is, if we were to implement, we have a product called LXR GPT, which is used internally as a team, where we are utilizing the productivity enhancement tools or features, functionalities, which can be built on this part. But that is accessing a wider overall array of data as such. It's a wider problem. Brent Peterson (24:25.422) You talked earlier about the bigger companies, top 1 % doing things so well. Do you feel as though machine learning now is going to more, is going to democratize the ability for smaller or mid -size merchants even to compete at that same level or at least have the same tools if they utilize them? Udayan Bose (24:45.923) Yeah, I mean, I'm really glad you brought it up, Brent. I feel... In fact, I would say I'm so confident. that I am literally just betting the entire Netflix's future on that. And I'm not talking about that we have really run this company for the last two years. We have been running it for 20 years and five months now. And I'm willing to really bet my company essentially that the future would be driven by machine learning, which is really helping you really experiment, test, come up with a different module. This is just one module that we have started with experimenting. And I think the analytics insights can go to a different level using AI as well. You can really, obviously we have talked about the productivity. the content generation, the image generation, et cetera, which is there as well. But I think the future of digital marketing is going to be, I think, powered by AI. And one of the areas where we are seeing or realizing massive gains for our customers, and I'm talking about customers which are primarily mid -sized customers in the categories that I mentioned, categories like beauty, fashion, home courts, and so on. And we are seeing a dramatic increase in the quality of customers that we are acquiring just through the process of of this AI powered experimentations, where the same amount of money is being spent, but the quality of revenue that is coming in is significantly better. So I tend to think, and that would be, I think, the ideal situation. Because as we know, I'm very bullish, Brent, and I'm very confident also that we'll hear a lot about it in the next few months and years as well. There is this concept of, one -to -one marketing or hyper -personalized marketing, right? So showing Brent the message which resonates with Brent, showing Udayan the message which resonates with Udayan, and so on and so forth. We are getting to that phase. And I believe that through the experimentation only, you'll be able to get to that level where you know as to what point in time, what time in the day, what day of the week, what channels are you really sort of focusing on, what products will the customer buy, and so on, and make it so fine -tuned. Udayan Bose (26:54.403) Primarily because, as we all know, that the customer's time is becoming more and more premium because there is just so much which is really competing to get access to that time. of the customer in such a very fragmented attention span. Essentially, you really have to make an impact very quickly. And I feel that AI ML in a responsible way can really help us to get to what we typically at Netflix really call it a responsible engagement. So that's what I think we are banking on as well. Literally betting the company's future on this AI powered experiments, which are very hyper personalized to win the right type of shopping. developers. Brent Peterson (27:37.294) So you've mentioned now NetElixir is going to be, or is 20 already. How about, has your core principles changed over the last 20 years? Udayan Bose (27:49.283) That's a beautiful point. If you were to look at it, I mean, let me talk about what has not changed, Brent. We started the company with a core purpose. The purpose was helping businesses succeed online. That has not changed. That will not change. So that helping businesses, I think helping has an outsized meaning now, right, with all this chaos going in. But helping has had a tremendously important meaning back in 2004 when we started as well. So that core purpose is not changing. Our value systems have really been focused on four pillars effectively. The first one was respect for people as simple as that, right? The second one was an equal opportunity for all. And we were fortunate enough to really earn the trust of a wide diverse group of people who sort of decided to join and work for our company. So that equal opportunity for all really helped. The third one effectively was really being obsessed about customer delight as to how can we understand your problems, challenges, and really be an extension of your team and solve the problem. And the fourth one is understandably, I'm going to say it's not really a priority, just focusing on the quality part. because everything at the end of it. It's important for us as an agency to really be mindful that we are essentially shepherds of the messaging that you are effectively putting it out to the customers. So that, I don't think, has changed. In terms of core principles of running the company, we have run based on two core principles, which I can share with you, Brent. One, it's more of an anecdote, which is a very funny anecdote, which every team member who joins anywhere in the world, any of our offices, sort of goes through. Udayan Bose (29:30.563) It's our huge, huge belief in the incredible power of human beings. And the example that I give is think of any new team member joining us, think of yourself as tea bags and the color that you impart in hot water. will define as to, and that should be only your color print and that color should be different from the color that I impart and that is what really makes it special. So that principle has not changed. So what we as a company are providing is hot water, which is almost like a challenging assignment, engagement and so on and so forth. But the color, I can't really. define as to what color will you impart. So that principle has remained the same. And the second principle is more of the growth mindset principle. I mean, effectively, I think in a space which is changing and evolving so quickly, it's extremely important for us to really have as a part of the team. The same, the party bingo when I talk about it, we'll figure it out mindset, right? But that growth mindset and being open to change. Because in times of things like this specifically, AI unfortunately does not have its own emotions and it couldn't care less whether you are an expert in something. But the rules are getting broken and rules are getting reformulized and changing all the time. And at this point in time, having a fixed mindset is probably the biggest liability any company can have. So I think our entire focus on getting team members very fortunate to have that with growth mindset. And secondly, the entire philosophy of the company is we will provide the hot water, but you have to ensure that you impart your unique color to that hot water. Those two principles have really done us pretty good so far. Brent Peterson (31:13.358) Yeah, that growth mindset is so important. And I can remember a company owner founder telling me that he was too old to keep learning. And it's very hard to change. And I think that, especially nowadays, with so much change happening in our space and things getting turned upside down, if you don't embrace that change, you're going to get left behind. Udayan Bose (31:24.803) Yeah. Udayan Bose (31:29.379) Yeah. Udayan Bose (31:37.731) Yeah, exactly, exactly. Yeah. Brent Peterson (31:40.238) If you had a little bit of advice to give to an e -commerce merchant, store owner, what would that be? I mean, I think they should be planning for Black Friday today. They should have already had a plan in place. But what would it be that you would say, if they don't have it in, they need it now? Udayan Bose (31:56.131) Primed at least, yeah. Udayan Bose (32:02.979) I would actually sort of have two very specifically. The first one is. just creating, carving those experiments of those tests. Because you don't know, and also what has happened with the emergence of things like PMAX and so on. I mean, even Google platform or Google ads has become a bit of a black box. So unless and until you're experimenting or testing, just because you have had some great results today doesn't mean that... those really can be extended to tomorrow. So effectively, you really have to figure out what is going within the black box. And for that experimentation, I think is the only sort of solution to this. And if it can be powered by an AI platform like Alexa Insights, that even better. So that's, I think, one part as well. The second one effectively is as simple as that. I mean, you have to reserve 10 % of your personal time as the business owner or the CMO to just explore what is going on because in many cases the biggest mistake I have seen a lot of lot of business owners really make they're just so so much working inside out that they just never get the time to look at us to What season am I in, right? Is it even spring, or is it summer, or is it fall? So that's where don't do that, primarily because in this case, AI is accelerating the speed of these changes. And before you know it, you would really be struggling to catch up with this shift or change in environment. And unlike other linear technologies which have happened before, AI is more of an exponential technology. Udayan Bose (33:47.971) And once you really miss the boat, it will be really difficult to catch up. So that's where 10 % of your time reserved for just thinking, reflecting, meeting with experts, talking to experts, speaking with people, and just getting a lay of the land what is really going on in the industry. Because the seasons, I'm telling you, are changing very, very quickly. And you just can't afford to be sitting in your basement or to really, and once you come out, gosh, I didn't have my winter coat and I didn't, I thought it was summer. So that is, I think, the most depressing experience to have effectively. And unfortunately, that entire thing has gotten expedited multiple times because of artificial intelligence as such. Brent Peterson (34:30.958) Yeah, I'm a big believer in EOS. It's an operating system for businesses. And one of the principles that was in the book, Traction, was work on your business rather than in your business. Because if you're in your business, you're just doing the drudgery day to day. And if you step back and look at it and work on it some, you're going to learn a lot more and be able to pivot. Like you said earlier on those places where you need to pivot. Udayan Bose (34:36.419) Yep. Yep. Udayan Bose (34:41.443) Yeah. Udayan Bose (34:45.411) Exactly. Exactly. Brent Peterson (34:59.598) and you can learn when you need to learn. And I like what you said too about at least 10 % about around your thought leadership. And I think we've seen big companies like 3M and Google give their engineering department 20 % to learn on their own because they realize when you release some of that tension of having to do, when you can go and find out what's new and do something new that gives you that extra creativity and extra things that aren't. Udayan Bose (35:03.075) Yeah. Udayan Bose (35:14.563) Mm -hmm. Udayan Bose (35:23.331) Right. Exactly. Brent Peterson (35:27.822) quantified necessarily in a JIRA ticket. Udayan, we have a few minutes left and as we close out, I give everybody a chance to do a shameless plug about anything they'd like to plug. What would you like to plug today? Udayan Bose (35:34.435) Exactly. Udayan Bose (35:46.307) Well, I would, as I mentioned, for some reason, Brent, I'm really becoming more of a cheerleader for this entire concept of AI -powered experimentation. Because I, by now, am literally 100 % confident that this is not really good to have. for specifically the mid -sized businesses, actually every business, it's a necessity, it's essentially a must. So really understanding the AI technologies, and that's where I think the shameless plug would be our technology solution, LXR Insights, lxrin sights .com. We... or for all the Shopify customers, basically, there is a direct integration. It uses your first -party data. And it will, within a couple of hours, tell you as to how good is your customer mix. Do you really have enough high -value customers to really help your business thrive in the future? And if not, it will actually recommend experiments that you can run on Google and Meta to identify the right type of customers that you should really build your business with. Because... Again, in my, I wouldn't say opinion, I am seeing it firsthand, working with, since we work with so many companies, Shopify businesses, majority of them. is we are seeing that the only way for you to really come even compete in this landscape which has the team and the sheen brand is to build your community of raving fans. When you say raving fans, I'm talking about these high value customers who are very loyal to your customer to your business and they sort of want to buy from your business. And our technology helps you, Alexa Insights helps you to identify and win more of them and build a much more solid business foundation and a better strong Udayan Bose (37:38.085) customer mixes with. Brent Peterson (37:38.766) That's great. I'll make sure I get the links into the show notes. How can people get in contact with you? Udayan Bose (37:44.579) Absolutely. Udayan Bose (37:48.355) I am very accessible Brent. I have my LinkedIn just type in Udayan Bose my first name and last name or send me an email it's Udayan it's probably it's going to be a tongue twister for some of you but Udayan at netelixir .com Brent Peterson (38:07.022) That's perfect. And again, I'll make sure I put all those in the show notes and on the on our website to make sure that they can get a hold of you. Uri and both. It's been such a pleasure. Thank you so much for being here today. Udayan Bose (38:20.931) My privilege. Thank you very much. Absolutely. It was such a lovely conversation and wish everyone all the best. My last message to all of your audiences would be just simple three words, Brent. Yes, we can. So with that, thank you. Brent Peterson (38:32.366) Yes, we can. Perfect. Tend the avad. Udayan Bose (38:39.299) Thank you.