This week’s podcast is about AI Agents transforming ecommerce. Likely all commerce.
You can listen to this podcast here, which has the slides and graphics mentioned. Also available at iTunes and Google Podcasts.
Here is the link to the TechMoat Consulting.
Here is the link to our Tech Tours.
Here is the McKinsey article I mentioned:
——–
Related articles:
- Two Lessons from My Visit to Tencent Cloud (1 of 2) (Tech Strategy)
- Tencent Cloud and Mini Programs Go International. Lessons from My Visit to Tencent HQ. (2 of 2) (Tech Strategy)
- How Amap Beat Baidu Maps. My Summary of the Alibaba Playbook. (Tech Strategy – Podcast 252)
- Scale Advantages Are Key. But Competitive Advantages Are More Specific and Measurable. (Tech Strategy)
From the Concept Library, concepts for this article are:
- GenAI and Agentic Strategy
- GenAI and Agentic Strategy: Agentic Ecommerce ABCs
From the Company Library, companies for this article are:
- Alibaba Cloud
———transcription
Welcome, welcome everybody. My name is Jeff Towson and this is the Tech Strategy Podcast from Techmoat Consulting. And the topic for today, get ready for agentic commerce. Starting with agentic e-commerce and then probably moving offline and everywhere else. And this is kind of one of those, this is one of those moments when I’m happy I’ve spent the last 10 to 15 years studying this topic. because this is just going to be a huge thing. And I think retail, commerce, e-commerce, I think a lot of the world that we’ve known is getting reshuffled, restructured, and maybe a lot of it’s going away. And there’s some pretty important lessons for merchants, brands, and retailers that I’d hit the panic button, and I’ll kind of explain my thinking why that is. But that’s going to be the topic for today. I’ve talked about agent strategy before I’m sort of building that out AI plus agent strategy but this is really specific to commerce e-commerce online offline yeah that’s I think where it’s hitting right now let’s see standard disclaimer here nothing in this podcast or my writing a website is investment advice the numbers and information for me and any guess may be incorrect if using opinion express may no longer be relevant or accurate Overall investing is risky. This is not investment legal or tax advice. Do your own research. And with that, let’s get into the topic. Now no real concepts for today, but this is just a whole big new bucket of, you know, of strategy. Most of what I’ve been doing is digital strategy. Well, this is kind of the next thing, which is generative AI and then agentic strategy. I’m getting tired of saying all that, so I’m just going to say AI strategy, but… Really not talking about the AI we’ve known forever for 20 years, probabilities, analysis, prediction, stuff like that. No, we’re talking about generative AI and then AI agents. So, kind of different, but I’m just going to say AI for the purposes of this podcast. Now, to sort of take all of this apart, because a lot of things are happening, a lot of new tools. But let me just sort of go through how I think about it. Why this is such a seismic change is because it changes the user interface. mean, the user interface for a long time has been for a buyer, I’m saying user, I’m saying consumer. You go to the website of a company, you go to the app of a company, or you go to a marketplace platform like an Alibaba, like an Amazon, or maybe you go to a big retailer, know, Walmart, something like that, again, can be the app, can be the website. But the whole thing is based on humans interacting pretty much directly with companies. And maybe they’re going through a marketplace like Alibaba, so you have somewhat of an aggregator platform in the middle. at least the user is a human. OK, if we’re talking about agents, it’s no longer the human doing the shopping. There’s different degrees of it, different variations. It’s an agent. It’s like having a personal assistant. Go get my groceries for the week. Well, the person that the merchant in brands is dealing with is no longer the consumer. There’s another party now. That’s a huge change. we sort of had these agent things is going to be a problem. It’s going to disintermediate a lot of established players. There’s a race to sort of be build an agent that’s the best friend of your customer right now but it’s going to dramatically change e-commerce for merchants and brands. I’ll give you some of the frameworks that are coming out of McKinsey and BCG. I’ve been kind of keeping an eye on that. I’ll tell you my simple version which I think is actually better. My version I think is simpler. It’s just ABC. Keep your eye on the A’s, the B’s and the C’s. The As are your assistant. The Bs are brokers. The Cs are concierge. So, a person may have an agent working for them, the buyer. That’s like having a personal assistant. Could be your travel assistant, could be your general-purpose assistant. That’s the, go get me my groceries for the week. B is a broker. B sits, you know, the brokers sit between the buyers and the sellers. These are independent parties. They may have their own. Agenda things like this this when you buy health insurance most people use a broker and they sort of sit in between and then the third week concierge This is what the business creates the merchant the brand you go to the hotel They’re going to have a concierge that can help you with various stuff, know various things. So that’s ABC’s Assistant broker concierge most of it fits under that I think quite nicely so We sort of start with the user experience. I’ll give you sort of three points for today. Point one, there is going to be a completely new shopping experience process for buyers. And lots of stuff is going to change. And most everybody is going to get intermediated to some degree. There is going to be someone between you and the buyer. Now you could look at it from the other side, between the buyers and the sellers, there’s also going to be an intermediary. If you’re the personal assistant to Donald Trump or the prince, who has personal assistants today? uh Big rich people, companies, billionaires. It’s something you get at that point. Warren Buffett has never been in a Walmart. They just tell their assistant to go get me stuff. uh We’re all going to get those. Those assistants, yes, they represent the buyer, but they also kind of have their own self-interest somewhat. You know who can shape Donald Trump or Prince Walid or someone like that? The assistant. They can sort of filter the options that that person sees. So even though they work for you, you got to keep in mind, if you’re selling to someone with an assistant, you really are dealing with two different people. And yes, the assistant can have their own agenda and self-interest. So, you kind of got to, it gets more complicated. But that’s going to be in the middle. There’s going to be intermediation. That primary user interface is going to change. Now, let’s step it back one level just to AI, not agents, just AI, generative AI. We already see this playing out, changing e-commerce. And it’s simpler, so we can talk about that first. Right now, it’s singles day. I’m actually in Shenzhen right now. I’m sitting in a hotel in Fujian. It’s Singles Day, which has now really become a month. And the big event that Alibaba is doing on Singles Day is to integrate AI tools into the experience of e-commerce at scale for the first time. And they do this. When they’re trying to do something new and significant, they will often force it to be launched on Singles Day because it sort of creates a momentum in the whole company. You know, it’s the difference between saying, I’m going to go to the gym and practice boxing to get in shape. Okay, I’ll probably do it. It’s very different if they say, we’re going to go to the gym and get in shape and you’re going to have to get in the ring and fight someone in three months. Okay, that creates a lot of urgency. I’m in the gym every morning at 6 a.m. because I don’t want to get punched in the face. Singles day is kind of like that for internal projects at Alibaba. They use it as sort of a forcing function. And so usually there’s one big goal every year. This year it’s to deploy AI tools at scale within e-commerce. And that’s what they’re doing right now. And out of that you can get a pretty good sense of what the AI playbook is. AI meets e-commerce. And it’s kind of three buckets. Number one, conversational search. I’ve talked about this before. wrote an article about Alibaba Cloud and what they’re doing in e-commerce. It wasn’t a very good article. It was kind of wordy and not too clear. But the main points were there’s a significant difference between keyword-based search for e-commerce and conversational search for e-commerce. Keyword search, which is not generative AI, not a chat bot, is you put in the words you want. I don’t know, a Huawei D2 smartwatch. And it’ll give me options. In that case, it’s pretty specific. But what it comes back with is very narrowly defined based on what I searched for. And there’s not a lot of wiggle room there. They might say at the end of that, people who bought this also bought that. Conversational search is more like, I’m thinking of getting a smartwatch that can measure blood pressure. The A, I might come back and say, well, you should consider the Huawei D2 smartwatch. It has a blood pressure cuff within the wristband. It’s actually really pretty cool. However, you might also think about getting a normal watch and getting a blood pressure monitor. And you might think about, here are some low salts. It opens up a conversation. Have you considered low salty snacks? Have you considered sort of moderate exercise to bring your blood pressure down that way? The conversation opens up so many different consumption possibilities. Maybe it’s going to solve the bigger problem. This person has blood pressure. Maybe it’s going to be something else. It opens it all up. And it’s a much more sort of expansive back and forth that opens up lots and lots of purchasing possibilities. That’s a big lever. uh Another big lever when you start to put these tools into the merchants and brands and Alibaba is you can start to personalize. Now we’ve seen personalization like on Netflix, hey, here’s a cue for you that we recommend videos you watch. But here we can personalize the content, we can personalize the marketing, we can personalize the promotions. Increasingly, you can personalize products. Very powerful tool, but you kind of need more know, firepower. Well, generative AI lets you do that. And then the third is probably prediction, which is not really generative AI. It’s traditional AI. The more you can understand customer intent, the more you can use those other things, conversational search, personalization, bundling, packaging, alternatives. Those things all sort of go together. Now, there’s other stuff that generative AI is bringing in terms of uh traditional e-commerce like multimodal search. can search with a photo. uh More content creation overall, things like that. But what Alibaba has been doing over the last couple weeks is basically integrating large language models. It has its Quinn series into core search and recommendation engines of Taobao and Tmall for the first time. OK. And as mentioned, that lets you get more conversational, more personalized shopping experience, and so on. They’ve started to do it, or at least the press is saying they’ve released some numbers on what the results of this AI Meet Search plan is. Here’s some numbers. I’m not totally sure how true these are, but they were in the press. I’ll read them. A 20 % improvement in relevant search results for complex queries. A 12 % improvement 12 % increase in return on merchants advertising costs and a 10 % rise in click-through rates for some recommendations. Okay, anyways, the AI story, let’s call those the three biggest levers right now. Now people argue about what the definition of AI agents are. put, mean the simpler version would be to say, look, this is just automation. Generative AI is very good at sort of intelligence, assessing things, and then generating content and making recommendations. Well, this adds the next step is it can also make decisions, and it can execute those decisions principally by giving it a big suite of tools it can draw on, like it can go into Expedia and book your plane ticket, like it can move a robot around, like it can drive your car. So, it can understand, study, learn, and then agents bring in, it can decide and execute. Now in e-commerce, does that mean it can negotiate with sellers? Does that mean it can hunt around for bargains for you? um So automation would be one way of thinking about it. The other way to think about it is like, look, this is just a digital worker. This is a non-human workforce. And you can customize them to do all sorts of some more general, some very specialized, sort of vertically integrated. OK, depends on where it is. We’ll see how it goes. Now, McKinsey had a pretty decent article on AI meets e-commerce. I’ll put the link in the show notes. And they basically said it’s three things, which kind of sounds like my ABC. The first is they said, look. There will be merchant agents uh that are kind of on the website or so on. people can deal with the agent, the concierge, directly, conversational search, that sort of thing. Fine. Agent to agent. This is helping one agent talk to another agent. OK, if I have a personal assistant, I can send it to a web page and it can deal with the concierge. That’s agent to agent. The broker, which is an agent, can deal with the concierge at a various company. That’s agent to agent. But it’s a different type of interface and you need a different protocol for that. The merchant agent, if you’re going to the site, that’s an MCP. If you’re an agent to agent, it’s a protocol for agents. And the third one they say is like, the merchant has an agent, the concierge, it can deal with a broker agent. Now, I’m putting my words in there, broker and things like that. But it’s basically those three scenarios I already talked about. Fine. So, we get three different interactions that could be happening more, I think. But then you start to look at other things. OK, if that’s how commerce is going. Do you give these things payment capabilities? And the answer is yeah. mean, Visa and OpenAI, or is it MasterCard? Visa, I think it’s Visa. They already have a deal for an MCP for payment. So, these agents will begin to be able to make payments, which is obviously huge. That creates all sorts of fraud issues. That’s a big deal. Trust, fraud becomes a very big topic again. We can look at things like loyalty programs. What does that mean to your membership program? What about if you’re trying to get not just sales, but engagement? We want the customer to spend more time on our website, more interactions, because that generates a lot of data for us. Well, it may not be the customer. It may be the agent representing the customer, the assistant. What does that do for your strategy? Fulfillment, what about returns? What about coordinating in-store service? Remember everyone talks about the omni-channel? Well, is the human going into the store, but the agent is doing everything online now because that just got more complicated? Can I put the assistant in a robot and have the robot walk down to the store? We’re kind of one move away from that. So, when you start to go through the typical activities of e-commerce, omni-channel, offline, online, and you put this idea of assistant broker concierge as players in the system, it all gets really complicated. And we kind of realize, yeah, this is going to change a lot. It might change the majority of the industry in some cases. Think about it like when e-commerce first emerged, what it did to traditional commerce, websites, travel sites, platform business models. Now for certain businesses like Starbucks, e-commerce has never been that big a deal. But if you were a travel agent or a big bookstore, e-commerce changed everything. This could be like that, and I think it will be like that in a lot of specific situations. let’s say you’re a merchant or a brand. You have a website. You’re selling makeup. Maybe you’re a retailer. You’re selling lots of other people’s products. really anything doesn’t have to just be physical products can be services it can be people creating content it can be a lot of things you need to get into the idea that you’re no longer just serving a human being that shows up at your website your store your app you are also now having to serve their assistance and also brokers they have contracted you’re going to have to serve all three and you’ve got to have a concierge of your own So the game just got more complicated. Think about a human uh coming into a hotel to deal with the hotel concierge. OK, it works for the hotel. He or she, now it works for the hotel. You might ask them things about the hotel, although often you ask the front desk for that. Can I get my suit cleaned? Where do I get? uh some breakfast like this, I’ve got an allergy, mean lots of things related to the hotel staying experience. Well, that sounds a lot like conversational search. It’s just a more interactive form of dealing with the core service. But you might also ask, hey I’m looking to go uh see a show in New York. Do you have any recommendations? Oh yeah, you should see Les Misérables or whatever Jersey Boys information. Can you get those tickets for me? Yeah, yeah, I can get the tickets for you. Sure, not a problem. So now your concierge that you’re dealing with representing the hotel, they’re going out to other businesses to do transactions for you. They’re giving you advice. the scope, this is no longer just a hotel providing hotel services. Once you put the concierge there, they are starting to provide lots of other service outside of what we would define as the hotel business. They can go out and make you restaurant reservations. They can get you theater tickets. Maybe they can book you a driver to the airport. Who knows? Now, the physical concierge in a hotel is quite limited in what they can do. A digital concierge may not be limited at all. Maybe they take care of health care. Maybe they get you movie tickets. Maybe even who knows what they can do. So, they’re going to have an ability to move out into other places. ah They can solve your problem. So, we can sort of see these concierge agents, they could be very vertical specific, like let’s say healthcare, or they could be pretty general. And the other thing to keep in mind is these agents, concierges, yes they represent the hotel. They might have some self-interest of their own. Does the hotel concierge get kickbacks personally by recommending certain, I don’t know, certain restaurants? Maybe. Okay, that’s a little bit dark. What about, hey, the hotel concierge is good friends with someone who owns a restaurant and he likes to refer to his friends. They can operate with their own sort of interests me in the way that many employees of businesses can do. Now, when you look at that same scenario, let’s say the buyer scenario, we’re talking about the assistant. This is now my own personal assistant. I’m telling them to go buy things, go do things, get my groceries. OK. They’re going out to deal with merchants, brands, retailers of various types. They may constitute a very small part of the shopping experience. Get me a pizza. OK, I’ll get you a pizza. It’s ordered. It’s on the way. Pretty surgical, pretty short. They may be going out and having lots of extensive interactions with businesses on your behalf that go on and on. They could be involved in every stage of the shopping and usage experience. So, it can be narrow or it can be pretty broad. and anyone who wants to deal with me and sell me anything, they’ve got to go through my assistant. Well, as I said earlier, personal assistants actually, especially of powerful people, they’re very important. They can whisper in the boss’s ear, yeah, that guy’s an idiot. Don’t work with them. OK, boss isn’t going to care. Yeah, OK, I won’t deal with that business anymore. I won’t deal with that supermarket. Where should I get my haircut? Oh, use this company. They’re pretty good. Don’t use that other one. Yeah, they’re not good. OK, personal assistants that are digital can do the same thing. They can shape and curate what you see and what you think. And now, did you build that agent yourself, and it’s yours on your own server? Or are you contracting it from another firm, let’s say OpenAI, that may have its own self-interest? So, my example of this has always been Wealthy people usually have a team around them that handle different parts of their life and different parts of their business. They have an accountant, they have a lawyer, they have a personal shopper, they have someone who takes care of their kids in terms of getting them to school and things, they have a health advisor, they have an advisor that plans their vacations. Usually, six to ten people. Well, we could all have six to ten personal assistants just like that, but they’ll be digital. Now the third one, the eight, I’m sorry the broker. You know, this is like health insurance. This could be like a travel advisor, a travel broker. I gave a talk five or seven years ago, something like that, in Beijing where it was at a conference for luxury travel brokers. And I didn’t even really know about this business. But high net worth individuals, when they go on vacation, they have sort of, they’re really almost like personal assistants. but they’re brokers in that they sit on both sides and they will put together and recommend various you know travel plans let’s go to the Antarctic and rent and get on a rubber boat and go up on the Antarctic and play around there and we’ll put it all together or let’s go to the you know a safari in Masai Mara and they’ll do that and they’re kind of you know can go between the buyers and the sellers in this way and I was at this conference speaking and The audience was 50 % luxury properties, like the Four Seasons in Dubai had representatives there. And 50 % were these sorts of luxury travel brokers. And we were sort of speaking to all of them and how they work together. And there were so many side arrangements between these brokers and the merchants and the Four Seasons and then the high net worth. That sort of intermediary thing was really kind of interesting. and they do have their own self-interest. A simpler example would be health insurance broker or really just health insurance company. Insurance sits between you and the hospital and they go and negotiate with the hospital on your behalf to get a discount and get lower fees and then they sell you a service and you buy the insurance and they deal with them. Fine. When I was running a hospital I used to have a trick we would do. Because the insurance companies would come in every year and say, we want a discount. We want a discount, or we’re taking you out of our network. And if you’re out of our network, none of our patients will go to you. And they have a scale advantage, typically. But they’re in the business of squeezing me for money. So, what we used to do is we would identify the top 20, 30, 50 companies in town, like British Aerospace, things like this. And we would invite the wives of the CEOs to come out for the day and have coffee and brunch with our staff. And we had a luxury hospital. And we’d sit them down and show them around the luxury hotel. We had really top tier, like a four- or five-star hotel. And they loved it. And then when the insurance company would come back and say, you’ve got to give me a discount or we’re cutting you out of the network, I would say no. No. uh Yeah, you but you should let British aerospace and I’d list all the companies you should let them know we’re not available in the network because the CEOs and their families come here and They would never kick us out of the network Now you can do the same thing as a merchant as a brand well you have to do the same thing if you’re a merchant in a brand or a brand and these Broker agents are sitting between you and your customer with their own self-interest trying to squeeze you for a discount, have to go around them. So, pop-up stores. Create great pop-up stores that people love. They come into your store and then when the AI agent tells them, don’t use that coffee shop, they’re not good, they will say back to them the same thing the CEO’s wife said back. We like that store, it’s our favorite. So, you can kind of go around them to pressure the intermediary in that way. I’ll talk about that in a bit. But yeah. OK, so those concierge broker assistance. Let me finish up this point one. Point two and point three are short. OK. So, you can see this is kind of a big deal. It raises questions for merchants and brands and retailers. Question number one, how do I get past the assistant? How do I get past the broker? How do I get the customer? Because ultimately, that’s what’s most important here. And this is a real problem. People have spent decades optimizing their websites for search engines. And now if ChatGPT doesn’t mention you, the traffic is not going from the search engine to your website anymore. It’s going to ChatGPT, and that’s where it stops. So, you can basically disappear as a website. Well, it’s the same as a merchant or brand. You can disappear to your customers if you are not dealing with the assistants or the brokers. How do I get past them? I got to work with them. but I also got to get around them. How do I get data on my customers? If I’m dealing with the agent, am I cut off from the data about what my customers care about and do? Google Search has an interesting question that their algorithm, which helps someone find what they’re looking for, is based on looking what people search for. But now the people doing the searches aren’t humans. So, the data is getting bad, or at least more complicated. Are we going to become dumber because we can’t get data and engagement from our customers anymore? Given one and two, how do I add value to my customers? oh My primary strategy has always been you’ve got to continually add value to your customers every year, every year, every year. More services, more products, more personalization, better content. How do you do any of that if you’re cut off from your customers to a large degree and the data? And not only am I getting a hard time reaching my customers, but I also have to continually add value to the broker or the assistant the same way I try to do it for the customer itself. I’ve now basically got three parties I have to deal with. How am I going to improve my experience year after year if I’m cut off and data poor? So those are big questions. Here’s a couple of conclusions. I’ll give you my working conclusions. this is going to wipe out a lot of brand equity. You have a pretty basic product. Most of what your value is coming from is the emotion or the legacy brand or the feeling someone gets. Well, if you’re dealing with the assistant and not the customer, a lot of companies, that’s going to just go away. There’s no brand equity. Well, there is some, but a lot of that emotion and stuff, the end user, the core customer is not going to see it. So, a lot of people are going to get their brand equity wiped out. uh Advertising revenue on e-commerce might be really wrecked. You know, if you look at Amazon’s revenue, one of the biggest sources of revenue was uh advertising. Is advertising still valuable if it’s an agent doing the searching? If it’s an agent browsing your webpage and looking at your products, is that the same type of advertising revenue as a human? Advertising revenue, going to get hit. As mentioned, merchants and brands, retailers, are going to now have to find out how to provide value to customers, brokers, and assistants. Going to have to do all three. And customers is going to be a lot harder. It’s going to change the game. And they all have interests. You’re going to have to convince them all. That’s complicated. That’s a big deal. Last one. entertainment is going to become more important. What’s the strongest way I can create a connection with the end consumer, not the broker, not the assistant? Watching videos. The broker’s not watching videos, but the end customer is still going to watch videos. They are still going to play games. They are still going to go into my pop-up store for the great experience. They’re still going to go onto my online store if it’s a wonderful experience. So… The fight for eyeballs is going to become much more important because a lot of the shopping might be done by the agent now. So TikTok, TikTok shop, they look to me like they’re in good shape. They already have the eyeballs of the customer because they’re going there to be entertained. And if they buy stuff along the way, they may throw that to their agent, but a pure Walmart type shopping experience, which is not fun. Why wouldn’t you just have your agent do that? So, entertainment is going to be super important. I think that’s it. OK, that’s point one. I know that was longer than I thought it was going to be. All right, point two. I’m going to do these fast. Point two, we’re going to see a new ecosystem emerge. Lots of new parties. People are going to change what they’re doing. And there’s going to be new parties and new players we didn’t anticipate. How do you deal with fraud? How do you do marketing and advertising in this world? Do I have to market and advertise to agents now? But how do I do that? Do I write articles and show them ads? How do I convince an agent to recommend my site to their client, their end user? Is this going to emerge into offline? And then we got all these new protocols, MCP, agent to agent protocols. Expect an entire, a significantly new commerce ecosystem to emerge. That’s point two. Point three, we’re going to see new business models. I always start with the user experience, right? That’s the beginning of everything. But then we’ve got to think, OK, what business model is going to be built against this? One of the reasons e-commerce is such a good business model is because I’m sorry, such a good business is because it enables marketplace platforms which are a very powerful business model. Do I even need marketplace platforms in an agentic commerce world? Do these agents need to go to a marketplace platform and scroll through hundreds of listings? Does it even need to do that? uh Unclear. I mean, we’re really talking about human and non-human commerce getting mixed together. Is it still a good idea as a marketplace to aggregate a lot of sellers into one location because that makes it, you why do we do aggregation in one place? Because it’s convenient, right? Do agents care about convenience or do they care about efficiency? Maybe they can just crawl sites like web crawlers and find the best deals and maybe they don’t need a marketplace or an aggregated source of, you know, sellers at all. Do you still need to aggregate buyers? Now, this I think is true. I think people that are launching brokers, now, OpenAI, they will say, we’re going to give you a personal assistant. And they will go to merchants and say, we’re going to give you a concierge tool. I tend to think those are going to function like brokers, where they’re going to sort of realize, look, I’ve got all these customers I’m representing now. Once I aggregate demand, I can turn around to the sellers and demand a discount, just like the health insurance companies. And if I can get a 10 % discount on something by purchasing economies of scale, that will make my agents more valuable than smaller sites. So, I think the health insurance model is going to be a thing. That’s not a marketplace. And that’s not a network effect. That’s different. Yes, we’re going to aggregate buyers. Is it brokers? Is it personal assistants? I don’t know. Do you need platform market business models at all? Can’t we just operate by protocol? Like every computer is connected by every computer by ethernet. Can’t we just have everything connect to everything and the agents can web crawl basically? I don’t know. Do network effects matter? It’s not clear to me that that’s true. Marketplace platforms, Amazon, benefited from network effects. Everyone wanted network effects for 25 years. A lot of what I talked about doesn’t involve network effects at all. Does Google still have a network effect if its mostly just agents doing the searches on behalf of humans? Doesn’t that wreck the algorithm? Doesn’t that decrease a lot of the value? So, there’s a lot of really big business model questions, and that’s number three. So that’s kind of my three points. Point one, the user experience is going to change dramatically. We’re going to see at least three parties now, user, uh advisor, and broker. And then on the merchant side, we’ll have concierges. So, three players. But yeah, the user experience is going to be different on the buyer side and also probably on the seller side. And there’s going to be a lot of intermediations. Point number two, we’re going to see a new agentic commerce ecosystem with lots of new players. Number three, these business models are going to change. In some sectors it won’t be much, but in others it could be pretty transformative. A lot of the big strengths we are used to may be going away. mean, Elon Musk had a comment a week ago where he thought smartphones were not even maybe going to be a thing. The whole idea that you’re going to have a smartphone with an operating system and a lot of apps That may not make sense at all. Maybe you’ll just have an edge computing device in your pocket that you talk to. And if that’s the case, or maybe you look at the screen, if that’s the case, why do I need to look at app versus app versus app? Why do I need an app store? Those are all built for human convenience. But if I’m mostly talking with the generative AI, the agent, it’s dealing with all that stuff. It may deal with it in an entirely different way. So, a lot of big business model questions. Now there was another article, I’ll finish up here, by BCG talking about this. I’ll put the link to that article. It’s called When Brands Meet AI Bots, Customer Experience in the Era of Agents. I’ll just summarize a couple points. It’s pretty consistent. Here’s the point. This is a quote. As more customers rely on third party AI agents to help them shop, now third-party AI agents would be brokers and assistants, not the concierge. Quote, brands risk losing their influence unless they reinvent the customer experience. That’s kind of my point. You got to get them to stare at the screen. You got to get them to come into your store. Something. It says they can spark that reinvention by understanding and tapping into three sources of power. Now this I kind of agree with, but kind of don’t. Understand, quote, how agents make recommendations. How each agent’s algorithm operates. You can ensure that they access, ingest, and understand the right information about your brand. Work with brokers. Work with assistants. Point two, what makes your brand different. The stronger your brand differentiation, the better your chances to offer a customer experience. I don’t actually agree with that. I think. I think the idea of a brand is something for the human brain. We need to be able to distinguish things in our brain. if one thing is very different, we understand it easier. I think experience and keeping eyeballs on screen is what’s going to matter more than having a distinguished brand. And then three, what makes your human team special? That’s just whatever. I didn’t think that point was very good. Here’s a last quote from the same article. The more that these agents affect how customers shop, what they purchase, and how they use what they buy, the more brands risk losing control over the levers that have traditionally shaped their customer experience. Bam, I agree with that entirely. Brands that become too dependent on third-party generative AI tools will see a thinner inflow of customer data, limited options to deliver personalized messages and interactions, and fewer opportunities to capitalize on their expertise. That’s pretty good. OK, that’s kind of it for today. yeah, this is kind of a huge topic. And I think we’re going to get there in a year. uh AI comes before AI agents. So, AI is being rolled out at scale in e-commerce right now. Alibaba is leading the charge. AI agents will follow that quickly. 12 months. if you’re a merchant or a brand, you got to move right now. I mean, I need to do this too. You have got; we’re all going to get intermediated in some form. Especially if you don’t have content or eyeballs on screen. Okay, that is it for me. I hope that’s helpful. I think this is kind of exciting and scary. But yeah, I would hit the panic button on this one. As for me, I’m here in Shenzhen. Been a fantastic week. amazing bouncing around to different companies you know sort of the who’s who of digital China flying around beijing Hangzhou shanghai i always like to go to shanghai and then go out on the bund sit and have a drink it’s kind of my favorite thing in China is to sit on the bund and you have a drink in a really nice bar yeah so I did that a couple nights ago and flew back down to Shenzhen uh I did some hunting around here, meeting some companies, things like that. yeah, it’s been a pretty spectacular week. I’m in a good mood. I’m a little wiped out. So, I need a couple of days off, which I’m going to do. I also, I missed, uh I’m a week behind on podcast and subscriber articles. I’ll catch up. this week. So, I didn’t do last week because I was running around. So, I’m going to play catch up this week. It’ll all be fine by the end of the week. And all this AI, agentic, e-commerce stuff, I’m going to send out in pretty detailed articles. I think that’s worth going further into depth on. So that’s on the way. OK, that is it for me for today. I hope that was helpful. I actually feel pretty good about this podcast. Sometimes I finish up and I’m like, ah, that wasn’t so good. This one kind of like, yeah, is important. So yeah, I feel pretty good about this one. Anyways, that’s it for me. Talk to you next week. Bye bye.
——-
I am a consultant and keynote speaker on how to accelerate growth with improving customer experiences (CX) and digital moats.
I am a partner at TechMoat Consulting, a consulting firm specialized in how to increase growth with improved customer experiences (CX), personalization and other types of customer value. Get in touch here.
I am also author of the Moats and Marathons book series, a framework for building and measuring competitive advantages in digital businesses.
Note: This content (articles, podcasts, website info) is not investment advice. The information and opinions from me and any guests may be incorrect. The numbers and information may be wrong. The views expressed may no longer be relevant or accurate. Investing is risky. Do your own research.
