Latest / Reframe First / Matthew Ferguson, Managing Director at eManaged | How AI is Transforming E-commerce Operations

Transcript

Matthew Ferguson, Managing Director at eManaged | How AI is Transforming E-commerce Operations · Reframe First

• Oleksij Rak: Hey guys, welcome to another episode of Free Frame First. I'm Alexey, the host of the podcast and let me welcome our today's guest, Matthew Fargusson, Managing Director of Emanaged. Hi Matthew, it's great having you here. Matthew Ferguson: Hello, thank you for having me on the show. Oleksij Rak: Yeah, thanks a lot for joining. from what I learned from your website, it says like V-scale marketplaces. It's Amazon, eBay, Walmart, Shopify. You manage all the marketplaces, channels, end-to-end, like operations, marketing, growth, and stuff like that. Could you tell us more about what you managed and about what you do there? Matthew Ferguson: Yeah, E-Manage is 10 years old this year. My business partner and I founded it ⁓ long time ago, it feels like. We didn't even mean to ⁓ found agency, I guess, I always like to sort of say as a little starting point. I think a lot of people set out to create an agency. We didn't even mean to. We were moving jobs ⁓ and we a number of clients contact us privately and say, could you? Oleksij Rak: Yes. Matthew Ferguson: moonlight or work on the side and we realized between us we had a starting point and the rest is history. We're a full service management marketplace agency. We will you into new territories or new channels or we'll just help you get more out of your existing channels. Amazon's very important channel so we're obviously an Amazon full service agency but we can replace Amazon with eBay, BNQ, Tesco, DevNom, OnBuy. Zalando and a multitude of other channels in excess of 150. I unlike a few other agencies though, we're very malleable. We have a very modular approach, so you don't have to ⁓ have full service. You can just have PPC or maybe we just create your listings and run your channel engine or your Linworks and we don't do anything else. So it's all very flexible to what you need. And yeah, we've got a really good track run of growing a lot of brands, punching well above our belt at times. So ⁓ the is definitely much smarter than me at delivering work these days. I just represent how good the team is. That's my role in the company. Oleksij Rak: That's a good problem to have, that you're being spooked on your day. Matthew Ferguson: Yeah, I'm very blessed. I feel very proud to them in the company. Oleksij Rak: So all you got to have is the product that you want to sell and you help people get that product to the market, to the marketplaces. Matthew Ferguson: Yes, we help them get it on the marketplace, we help them optimize it, and then we help drive traffic and promotions to get the most out of it. And we help scale the backend operations so that you're not just adding more work to add more channels, that ⁓ are and manageable, and especially with the rise of AI is automated and as efficient as they can be. Oleksij Rak: Yeah, with regards to like Amazon, Amazon is flooded with cheap, low quality goods and how easy is it to get to Amazon nowadays? when you have a great product. So they're in the territory of like ⁓ price wars. Matthew Ferguson: Well, okay, so there's a few things I guess to talk about there. So having a good product is definitely make or break in a lot of situations. you've got a very bad product, you're going to have a tough time in today's world because there's just a lot of people with great products out there. So that is almost a given in order to start anything. Amazon not as price sensitive as a lot of people think. You don't have to be the cheapest. Oleksij Rak: Yeah. Yep. Matthew Ferguson: a pair of socks or hats or phone. not necessarily about that. Amazon is much more of a convenience ⁓ marketplace. It's a marketplace where the buyer is ⁓ of the A to Z claim once the next day delivery and willing to pay a little bit more for a good service and ⁓ that foundational layer of trust. Really on Amazon, what you want to do is create the best value proposition. And that comes down to the branding of the product, the imagery, the title, the bullets, the tone of voice, but also the design, the A +, the storefront and the tone of voice you're adopting and the type of USPs that you're sort of accentuating. And a lot of people do well on Amazon by just selling alongside products that are very similar. because Amazon is a very big ecosphere. There's millions and millions of buyers daily coming to it. And the Amazon algorithm is always trying to find new wins and reward good successes. there are definitely difficulties at launching on Amazon, but to say that you have to be the best priced and that it's flooded with cheap product is not necessarily against you because... Oleksij Rak: Yeah. Matthew Ferguson: people who buy those cheap products probably won't repurchase, they might leave a bad review. And are you ⁓ cheap product? Are you a good product? It's about creating that value proposition first and foremost and making sure you're targeting the right buyer ⁓ interested and wants those USPs. Oleksij Rak: Yeah, I see. Yeah, that's also what I heard before, but it still good to double check. long does it take for me as a customer when I come to you and, okay, I got a few products in some niche, how long does it take to get on the marketplaces ⁓ how long is onboarding process and what does it take? Matthew Ferguson: well, we're very quick. So to work with us is very quick. ⁓ It really just depends. So on Amazon, ⁓ I will say that we have two models. there's the traditional managed service where we are your retained team to manage your store. We also operate our own Amazon store. We have our own eBay store, our own other marketplace stores where we can take your stock and sell it through our stores. Oleksij Rak: that they're very, concerned about. And a good point. And I that's a very So, I'm gonna end this off with couple of questions. and take it to the next level. you Matthew Ferguson: That is probably the fastest way to get started because we're pre-set up, we've got all the infrastructure and all we have to agree on is terms for taking your product. But if we're talking about specifically the managed service route where we're helping you, it really can vary. Amazon, you can be up and running within several weeks if you've got all the paperwork ready, ⁓ the the certificates of incorporation and whatnot. ⁓ Oleksij Rak: So, to bit training process to see we about the collaboration. We're to be lot of work on the places where we can do this, and we're lot work the this. So, we're going be doing a lot of on the places So, we're work the places this. And we're going on the places do this. And where we can do this. And lot the places And lot on the places this. And Matthew Ferguson: Other marketplaces can move much faster. They can be within a few days. Some marketplaces just have a very long vetting process. So Zalando can take months to launch your catalog because they have very stringent data requirements. They do testing and they've got a manual vetting process for each one of your products where if they reject it and tell you to make changes, they will then waiting two or three weeks for that flow to sort of go back in. So it really just does depend. It also depends on your internal setup. Oleksij Rak: that that we need to do in the future is learn about the benefits of the back end. So it's important to the benefits the back terms of how it uses those. it's only going to be made with that approach. And I would say it's really important to understand the benefits of the reason that you don't want to put your on a great box. That's what most likely going to be the most effective side of the job. And I that's important thing. Matthew Ferguson: your systems. We help you figure all of that out. And what I always try and say is, we're not going to be the delay. We're not going to be the reason that you don't launch on the marketplace in the timeframe you want. ⁓ That's most likely going to be something on your side that you don't have prepared, which is why we go through an extensive sort of ⁓ expectation setting period, make sure that you've got everything. ⁓ And we also align our commercials to make sure that Oleksij Rak: So, we're a lot research on We're going to a of this. We're a lot this. We're be a lot We're a a lot this. this. this. We're going a research We're going lot of a of We're going lot of research this. We're going a lot of this. Matthew Ferguson: If there is a delay, you're not paying to just have us waiting. So there's almost a project phase and we don't really start driving day-to-day performance and charging you for that service until you are actually live and running. yeah, to go back, it just can take quite a lot of time or it can be very fast. It really just depends on all the variables. Oleksij Rak: So usually what fails is something on the client's end, not being ready to go to the marketplace. Matthew Ferguson: ⁓ not having the right paperwork, just the basic, Amazon, ⁓ again, it depends. So Amazon is strict and the types of paperwork to set the account up. Some are very relaxed. ⁓ the technical side tends to be a bit of a problem. I would say the one thing that holds a lot of sellers back on marketplaces, and this is, ⁓ this is more about launch and is that. they want to dictate terms to these entities on how they want it to work. And what they have to appreciate is that these are big channels, they're SAS solutions at the end of the day, they have a linear way that you set up and you work with them. And then they have a fixed ecosphere where you have specific tools and algorithms that work in a certain way. So you can't push your objectives onto the channel, you have to Oleksij Rak: and they want to support it. And the thing that we appreciate is that the YouTube channels, they're very successful. They're successful. So at the end of the day, they come up in your way. And you set them up and you work with them. And then they come up in your way. So you can see results. And you're working with them. So you can follow the work. Your objectives, what's the challenge you have Matthew Ferguson: of the channel. And sometimes there's friction in terms of how brand perceives how they should launch on marketplaces versus the reality of how they need to. Oleksij Rak: Yeah. once you mentioned recently, in the beginning of the conversation about some automations in place, and like, it's quite trendy nowadays. So what are you automating in this process? What have you automated? And maybe how did it help you? What did it say, Will? ⁓ That's a new reality. Matthew Ferguson: Yeah, good question. mean, everyone I think is trying to get automation just to be competitive with everyone else. ⁓ It is, there's a gimmickiness to AI and automation because most of AI of today is actually just ⁓ with an interface to of communicate with and BI has been around for quite some time. Oleksij Rak: So. Matthew Ferguson: I mean, just, ⁓ adopted automation in lots of little ways and some very big ways. So the biggest thing we've done is automate our data management. We've built our own AI platform for data automation and management, which allows us to take a file from a client's website or an Excel or whatever it might be, a PIM solution, Magento, an ERP. And then we can take that file Oleksij Rak: Mm-hmm. Matthew Ferguson: virtually in its raw format without any changes, map it into our system, and then we pre-select the channels that we want the AIs to run on, so eBay, Amazon, France, maybe &Q, maybe Debenhams. And then that, ⁓ our system will produce the categorization, the item specifics, and the data formats we need to then... Oleksij Rak: Okay. Matthew Ferguson: upload that data in directly to the channel or through the third party system like the Linux or the rhythm. And that is massively improving our scalability for sort of the larger catalogs and just the day to day management. So new product gets added onto the feed three months down the road and the system just makes it ready without a human being needing to push things along. And we're improving that all the time with ongoing new channels and getting into other sort of levels of optimization with regards to descriptions, titles, and bullets, and having a control for different brands. That's massively reduced of our overhead. A lot of our time was going into the data population. I think the reason behind all of is that you can list a product quickly, but if you don't optimize it, it won't go very far. and the difference between optimizing it well and filling it all in and just listing it quickly, there's a lot of extra data and a lot of extra valid value requirements and mappings. So we are trying to create those listing from inception that have all that depth in them. But we've automated lots of other stuff. ⁓ have AI makers, of note takers on all of our client calls. And the moment the client call ends, ⁓ Oleksij Rak: Yeah. Okay. Matthew Ferguson: It translates the entire call into a series of bullet point actions and points and requirements that then gets fed into our client Slack channels, which allows the entire team, the entire company to see focal points of the discussion, the mood, the, you know, what's going well, what's wrong. And then a subset of that gets automatically pulled into the Asana project board so that then the team is aware of what those events are without anyone having to manually go in and log all of them. And I mean, that's just the tip of the iceberg. We use AIs for variation recommendations now. That was a new idea from someone in the team, ⁓ which is a really good idea. And event logging and sort of understanding that if we see this, we should consider that. So for promotional activities and seasonalities, there's lots of AI use that we're adopting. And we're just trying to create a mini ⁓ Oleksij Rak: Awesome. That's, yeah, as you mentioned, AI is a type of iceberg. So you have a lot of automations in terms of data pipelines, transformations and like summarizations, whatever. Matthew Ferguson: almost trained AI to do a specific thing that just helps the team not have to do it manually, or that gives them the information straight away so that they're not blind to it. And it just allows us to improve our service and our speed and our scale. Oleksij Rak: and enrichments probably. Yeah. Yeah. Awesome. Awesome. How long did it take to build it? did you start building? Like you're on the market for 10 years. Quite likely ⁓ been doing that manually for a long time. Matthew Ferguson: Yes. We were, and our tech journey has been learning everything the hard way and making a lot of bad decisions the hard way. The first version of our feed management tool we built, we built eight, nine years ago, I think, but it didn't have AI at the time. It was a rule-based system and ⁓ adopting sort at the time similar. Oleksij Rak: Yeah. Okay. Matthew Ferguson: components of sort feed management tools with regards to automation. So things like scheduling and event logging. We were re-platforming that into a new web-based sort of front end because we wanted to eventually put this on the market at some point or at least have the option to do so when AI started to explode. And so it was very easy to adopt those AIs and layer them in and sort of architecture it around the AIs. Our platform is naturally built around AIs ⁓ in many ways that other systems are trying to add it on as an extra. ⁓ But the 10 years, we went on all sorts of tangents. We tried to build some dashboards and reporting tools. We went on an advertising thing. We built various apps and sideshow components. It's only in the last two and a half years we've been focusing on this and as well the agentic-based dashboards. Oleksij Rak: So, going to go ahead start with the first question. Matthew Ferguson: ⁓ Most of those years was just wasted time and effort, unfortunately. One of the reasons why I have gotten so white, because of all the self-financed funds that have been put into wasted events. But you learn a lot in the process. Oleksij Rak: Yeah, that's the usual way to do You from your own mistakes. Awesome. What do you think still takes the most time in your business? What would like to get? automated or fixed with the snap of your fingers. Matthew Ferguson: Those are probably two different things. Oleksij Rak: Okay. Matthew Ferguson: Our time goes quite a bit into the client calls, because we have three or four people joining that call. So, so, ⁓ we try and have lots of those touch points. Even though we have emails and dedicated Slack channels, nothing really stops, ⁓ replaces ⁓ call where you can talk through things and understand each other's perspectives. And as a team, as a managed services team that wants you to consider us. Oleksij Rak: Thank Matthew Ferguson: what's, because there's an extension of your team, are important things that need to be there because we're not in the office with you. So you need to feel like we're close to you. And I don't think that can be fully automated. We're trying to streamline around it, but I still think the call is very important. We're trying to automate more of the reporting and the ⁓ actions. A time I was speaking to someone who was very senior in tech ⁓ and Oleksij Rak: Yeah. I it's so important to very clear about that. I that the of New is a place where the technology is there to help these people. The technology of tomorrow is going to tell them how they're going to do it. I think a great way to see the longer-term experience of reporting this moment. And think it's still a very, very rare opportunity to see it. Matthew Ferguson: He said it years ago, he said that the technology of today is helping a user do something. But the technology of tomorrow is going to tell them why they need to do it and or do it for them. ⁓ And what we still see in a lot of our ecosphere is reporting that is kind of stagnant and fixed and we're ⁓ just as guilty. reports that show you a snapshot of what's going on and look at metrics, but Oleksij Rak: Okay. Matthew Ferguson: don't really explain what those metrics need to be turned into in terms of next month actions. And that also a human being has to go look at the data to find what's important from it. So what I want to see us automate more this year, which we are working on, is basically spinning that whole idea on its head, where we have automation in place that's looking at the events, looking at the opportunities, looking at the inputs, is sort of pre-trained on our knowledge and what a human brain would be looking for in that data. and is surfacing the action and even at times do performing the action in certain environments. That will be very valuable for us, but I think just to produce a better service in the market for clients. Oleksij Rak: That's a great observation actually. Yeah, that's absolutely great observation. you basically have a system that looks... Matthew Ferguson: Thank Oleksij Rak: through these dashboards or look through the data, the system doesn't need the dashboards. And raises some action items or adjust accent on the data and to improve the product placement or whatever is there. ⁓ Matthew Ferguson: Or advertising or stock management. ⁓ It can look at different things and produce different actions, but essentially, yes. One of the easiest ways to lose sales ⁓ on is to run out of stock. And you'd be surprised how many brands do that on a regular basis. But forecasting is just one way of fixing that. It's also about price controls and promotional strategies. Oleksij Rak: Yeah. So, So, Matthew Ferguson: which can help you get the most out of stock, but also make sure that you can cater to the widest array of customers, especially in a multi-channel. If you're just selling on your website in Amazon, you have two channels, you have two bases to get a customer. If you're selling on 15 different marketplaces plus five wholesale accounts, that is a much bigger animal to tame and to understand and to forecast for. Oleksij Rak: Yep. Yep, yeah, for sure. Awesome. against the stepings, ⁓ out is like, how do see the market changing nowadays and what challenges ⁓ are? there for the marketplaces, especially like with this age of AI with agentic exploration stuff. What's now? Matthew Ferguson: well. actually argue with people about this quite often because I think okay so I'll preface by saying I could be really wrong on this I could be really really wrong and I'm wrong about a lot in my life so ⁓ definitely I be. I have a gut instinctive belief that AI right now is still in a gimmicky stage and we don't see it as a gimmick. So right now we're seeing a lot of technologies, fast track getting AI plugged in to solve problems or create efficiencies or fix manual gaps, whatever it might be. And these are useful adoptions of AI. They're the right way to go about it. E-commerce has been built entirely over the last 25, 30 odd years a human being in front of a screen interacting with a device. Oleksij Rak: Mm-hmm. Mm-hmm. Matthew Ferguson: typing keystrokes, navigating with the mouse, entering a search, navigating the search returns. does the button go that caters to me wanting to buy? Everything around site optimization, content optimization, digital marketing is all around a human being and the constraints that come with it. Be in the first three pages of results because a human being won't look at page seven, which is true. I think what we're seeing is AI is plugging all these little gaps and tools of the e-commerce landscape of today. But I don't think the e-commerce landscape of today is set in stone. I think the entire infrastructure is going to shift. an I like to sort of talk about, which I don't have a good answer for, but ⁓ don't know where it's going, but just here's an example of it. Digital marketing, as I just talked about. ⁓ I paying... Oleksij Rak: Mm-hmm. Matthew Ferguson: click. I am paying for someone to click and by doing that I'm trying to inflate my product in a search position. So pair of trainer shoes I enter into search. A million ⁓ 100 I'm only going to look in the first three or five. Organically speaking not everyone can get into those five pages so then it becomes how much does the Does the company pay to inflate their position to get into the first few pages? Do they have the margin? Do they have the budget? But that's what they're doing. They're catering to a human being who's not going to go past the first three or four pages of search results. In an agentic-based world where I'm asking an AI agent to search for me, they're going to search the full 1,000 pages. And then they're going to tally all the results together in a second in a way a human would never do. Where does digital marketing now fit in a landscape like that? Oleksij Rak: So, Yep. Matthew Ferguson: where it doesn't matter if you're on page one or page 50 in terms of does the product fit the brief, will the users still want to buy it? so there'll still be advertisements, let's say, so to speak. But that's one example I feel where AI is going to change the infrastructure and how we interface with the internet. I don't it's unrealistic to consider that our kids or our kids' kids will be asking their personalized AI agent that's in a little pin on Oleksij Rak: Mm-hmm. Matthew Ferguson: to, you know, that's tracking their lives, seeing where they're going all the time, listening to their conversations and say, I want to, need a new pair of shoes. I need to be size. It already, actually the AI would already know the size. would know your credit date shows. It would know your address. So I'm looking for a new pair of trader shoes. I kind of want them to be, you know, modern within this price range. I want them to be black. I want them to be comfy, but also good for running, let's say. And that agent just goes and it looks across the entire landscape. Oleksij Rak: Yep. Matthew Ferguson: and it produces based on your criteria, what are some good pairs of shoes? It surfaces them to you. You say, yeah, that sounds good. Purchase it for me. There's no human being in the equation. There's no human being in front of the screen. There's no human being with inflated PPC to sort of guide search. There's no display ads. Even the transaction itself could be occurring off the marketplace. So what is the nature of a marketplace now? ⁓ do I even need to go to Amazon? Oleksij Rak: Yeah. Matthew Ferguson: the whole convenience of going to Amazon is click, click. I purchase it. It comes next day. No hassle refunds. Well, what if an AI could just mimic all of that behavior for you? Because it's scanned, you know, trust pilots and make sure that always will refund if the order gets lost. So I ⁓ guess, so long story short, I really don't know. ⁓ Oleksij Rak: . . Matthew Ferguson: I think that what we're seeing right now is an explosion of AI use and then a counterbalance of we've gone too far. Some people have embraced AI so much that they've realized it can't fulfill the need of the rehiring. And that's why there's this resurgence of hiring developers right now and bringing staff back in for these big tech companies that thought they could just have AI replace everyone. So, so there's definitely a balancing act going on about how good AI is and what its limits are. But I don't feel like anyone's thinking about how AI is going to change our interfacing with the internet and the infrastructure and how that will become a bit of a paradigm shift in terms of what even e-commerce means. That was a long answer. Sorry. Oleksij Rak: Yeah, no problem. absolutely. Digital marketing will change for sure and buy a lot. the Yeah, there should be agent right there. ⁓ With to big marketplaces or like... Matthew Ferguson: SEO, product SEO, website optimization. Suddenly it's about an agent and that's a very... Oleksij Rak: comparing to the travel industry, big travel agencies. still questionable because the rest of the world is really unstructured ⁓ ⁓ I would say. And they put a lot of effort into bringing all of that together. So they will survive, but will Google or will digital marketing, this form that we know right now, will they survive? I don't know. do you see this change happening already? Do you feel like, regards to your clients, what's ⁓ ⁓ regular acquisition channels that you notice? Did the acquisition channels change Is it by PPC? Is it Google? Is it like LLMs? don't know. What's in there? ⁓ Matthew Ferguson: No, is definitely changing for and for the industries around them. So we have clients increasingly asking about how do I make sure that Amazon's Rufus surfaces my product in selection versus by competitors, which is much more about responses to questions it is about specific USBs or keywords. It's a tonal shift. And it's a tonal that is still unclear in terms of where you embed the data and... do you cater to one or the other because in some situations you can't have the room or the ability to have it cater to both. But that's just a small example. There's all sorts of shifts going on in terms of the tools that are being adopted and what's expected from a tool to be able to do. And I think four years ago, what you would expect of a customer service ⁓ AI agent, is very different from what you would expect it to be able to do now in terms of its ability to interact and answer specific questions. So I think we're seeing a lot of shifts in various dimensions, but I guess the vast majority of the industry is still very far behind. ⁓ offices still have someone manually note taking in meetings. Most brands ⁓ are themselves terms of finding ⁓ systems and automation are still quite manual and ⁓ Excel driven. ⁓ Oleksij Rak: So, So. So. Matthew Ferguson: To say that I think that the tip of the spear of the industry is really cutting edge and exciting. And I literally love the way AI is getting involved in the medical field and the legal field to help make these places more accessible to people and also advance research or finding new chemical bonds and engineering and all sorts of things like that. But that's just the tip and that's the stuff you hear about in the news. Everyday people, for the most part, don't even adopt AI. Oleksij Rak: So. So. Matthew Ferguson: Most people on Pioneer don't really use AI yet or use it in such a simplistic way. So I think there'll be long, there'll be generations and it'll be an adoption. I think each generation comes and adopts new things. And I think the next generations, seeing as how we're both gray, maybe not us so much, but our kids and our grandkids, they're the that will be pioneering adoption, which will pull the industry further in these steps. Oleksij Rak: I'm not sure about that. like adoption is still low, it's moving so rapidly and ⁓ it can you know, like Matthew Ferguson: That's true. Oleksij Rak: Imagine tomorrow, Ciri is not so dumb anymore. Ciri is your AI agent and you can say, hey Ciri, bring me a pair of shoes and it goes and does it. everybody, half of the world has Ciri in their pockets. Matthew Ferguson: No, you're absolutely correct. ⁓ A counter-argument. Yeah. And a good counter-argument for that is the horse and the car. When the car was coming about, everyone said, ⁓ it's a gimmicky, it's a fad. The sort of business that sold, was it horseshoe work? I forget ⁓ the specific professional Oleksij Rak: So, and all of a sudden, you have the technology, which is available. Matthew Ferguson: many of them felt like, we've got a decade, we've got maybe 15 years, we'll have plenty of time because it's gonna be slow in adoption. been using a horse for thousands of years, there's no way it's gonna get replaced overnight. And in some situations, like you just pointed out, the shift is so extreme, within 10 years, it from ⁓ one 10 are using car to nine out of 10 are using a car on the street. whole industries died overnight and whole industries exploded overnight or around that adoption, happened just in a very short period of time. So you're absolutely right. Smartphones, I guess, is another good example of that. ⁓ I just feel that there's a counterbalance going on against AI that is more symptomatic of the economic system and models that we have in place where Oleksij Rak: Yeah. Matthew Ferguson: full AI adoption would actually break entire countries in terms of economic models because they're already, countries are already struggling to have enough jobs to their people employed, let alone tax revenue. And when have this funneling of scalability and productivity channeling into sort a few specific areas, you'll create economic meltdowns in certain areas if it's adopted too fast. almost a reality that even if we wanted to fully adopt everything, we'd have to create a proto utopian state to a degree to be able to live with it. Because the the you know, I don't want get into politics, but but like that the current economic model is not designed for the level of automation and scale that AI could bring. Oleksij Rak: We had that with Industrial Revolution. Somehow we survived it. whole landscape has changed since then. ⁓ Matthew Ferguson: There's a really interesting book on exactly that argument that says that's not a good comparison. I understand what you mean and I agree, ⁓ some, can't the details. It was a very interesting paraphrase that I read from it. The nature of which the industrial revolution occurred was very quick. what a lot of people misread in the history is it created a lot of poverty as well and a lot of dysfunction and a lot of people having to move to urban locations and lose their jobs. A lot of stuff we know about. With AI, the problem is that it's almost like a multitude magnifier. it's ⁓ exponential curve in terms of the scalability is some industries is really sharp. Oleksij Rak: Yep. Matthew Ferguson: And although over generations people can just learn new tasks or not learn to become a, you know, a horseshoe specialist, the people that live through that time have a livelihood of 30 years where it's difficult for them to retrain or to lose their livelihood without being very resistant to it. And those people have kids, they have family connections, they have loyalty and customers. So... knows? Who knows? I have no idea. I don't pretend like I know it all, but it's a very interesting time we're living in. Oleksij Rak: Yeah, absolutely. Yeah, with regards to the future and the time we're living in, how do you see it? What are your bets about the future for your business, what you do? Matthew Ferguson: Huh? Oleksij Rak: because the world is changing so rapidly. yeah, really be able to help your clients like you do them, like you do it now. Matthew Ferguson: Well, yeah. It is, it is. ⁓ No, I don't think anyone working in e-commerce should be complacent. Anyone. I don't think anyone's safe from disruption ⁓ potential job loss. I think everyone needs to be on their toes with things that are changing. I think inevitably from a managed service and a ⁓ perspective, you can put all the best AI tools in place. There's still going to, for at least the foreseeable future, need to be some human being to run it, to navigate it to some degree. Oleksij Rak: . I think that the idea of the United States of is that not a new country. It's a And it's a going to place peace and stability for all of us. And it's a country to be of peace and for all of us. And it's a going be a place of for of us. Matthew Ferguson: Now, that might mean that instead of 20 people doing it, one person can do it. That still means one person will be able to do it. ⁓ And there's always going to be people that would rather someone else do it than them figure it out and learn it themselves. We've seen vibe coding explode, but we've also seen a lot of people just not want to create their own app. They want to just license someone else's. ⁓ So I think there's always going to be a time and place for services, consultancy, but I think it will be a very different shape. Oleksij Rak: So. So. Matthew Ferguson: And one of the reasons why we're investing in technology for ourselves is because we want to be, ideally, on the tip of the spear. We want to be on the cusp of these things so that as things move to AI, we have tools for that too we can monetize as well. Yeah, ⁓ I no idea, though. I really don't. I would love in a perfect world for us to get to some sort of Star Trek utopia over the next couple of generations. But that's a whole educational problem that is probably a... podcast for another time. ⁓ I just hope humanity starts to embrace for its ⁓ and sheds its very outdated ideas ⁓ of and frankly, if I was going to say. Just whatever we do, let's do it together and work together. Let's not be Darwinian and selfish ⁓ how get our future. Oleksij Rak: Okay. Yeah, that makes sense. you're trying to move fast, did you experience like, I've heard from a ⁓ people that ⁓ you move too fast ⁓ few months, there will be another like AI or AI tools that does exactly what you invested like your last half a year of time. Did you that because like you're building something, you're investing. Matthew Ferguson: Yeah. Yeah. Oleksij Rak: to some new tool set, new something, new technology, whatever, and then like in a few months time there is another AI tool doing that. And it's almost free. Matthew Ferguson: ⁓ I'm under no illusions that we're building something unique. What we're building exists in other solutions already, but you have to license that tool with everything else and all the costs that come with it. We're building it for our needs and most of the time that we put into building was actually not the AI related stuff. It was the things that wrap around it that make it useful in a scalable sort of automated manner. ⁓ But that's good point. When AI first came out, ⁓ when was exploding three, four years ago, Oleksij Rak: Yeah. Yeah. Matthew Ferguson: There were so many tools that were launching overnight about we can do this to a PDF or we can automate this. And then AI just embedded it in, moved on, part of the license. And those companies, they were around for six, nine months and they just died. ⁓ And I feel like there's another round of those companies coming as AI sort of continue to mature. I also think that there's just limits ⁓ how much sort centralization you can do in these things. ⁓ And I think we're seeing the limits and the infrastructure and cost that goes with it becoming apparent with things like OpenAI, is that company ever going to make money right now? And it can't just burn the money it's burning on a continual basis because it's trying to recreate this ⁓ all AI that can do everything. And I think what we're seeing from very, very different models coming out of China is they're pinpointing their eyes to be very good at doing one thing well. and to have a much more scalable infrastructure behind it cost-based to it. Who knows? Who knows where it's all going to go? And you're right that you've got to be very cautious because I wouldn't want to build an AI-specific tool right now. We're building tools that have AI embedded in them, but it's still a tool you need with or without AI. It's just that AI makes it such a better tool. ⁓ I not want to build a tool that is entirely built off of AI. ⁓ Oleksij Rak: Awesome. Matty, that was really interesting discussions. Yeah, we touched a lot of points. Matthew Ferguson: I'd be worried about the longevity of that model right now. Thank you. Thank you. It was great to chat. Oleksij Rak: Before we wrap it up, do you have anything else to add on top of? Matthew Ferguson: No, I mean, hopefully I don't need to plug what we do. And if you need help on Amazon and marketplaces, we're just a cool way. But no, I mean, I would just say that, just going back to what I said a minute ago, what if we do humanity as a whole? You know, ⁓ we're to have to figure out a lot of changes to come, I think, with AI. And it's a societal thing. It's an economic thing. And that both those things will translate into politics. Let's not misled by Oleksij Rak: Yeah. Matthew Ferguson: foolish rhetoric, you know, let's be pragmatic and do what's good for humanity, not just what sounds good for some and not others. Oleksij Rak: Yeah, that's a great wrap up. Thanks a lot, Matthew, for joining and for this interesting discussion. Matthew Ferguson: Thank you. Thank you. It's nice to be here. Oleksij Rak: Then have a great day. Matthew Ferguson: YouTube.