From A9 to AI: How Amazon AI Is Changing EVERYTHING For Sellers

From A9 to AI: How Amazon AI Is Changing EVERYTHING For Sellers Speaker 2: Amazon search is no longer ruled by keywords. Like it or not, it's being rewritten by AI. Rufus and COSMO aren't just buzzwords. They're reshaping how your products are found, ranked, and bought. But here's the part most sellers are missing. These AI changes demand a whole new content strategy. In this episode, Max Sinclair reveals how Amazon algorithm is evolving, what you need to change in your listing, and why ignoring this shift could cost you everything. All right, we're going to be talking about and talking to one of my favorite guests. He's the founder of Ecomtent.ai, a leading content generation platform built for the future of AI powered search like Cosmos and Rufus. He helps brands celebrate high performing visual and written content tailored for emerging AI discovery tools. He's previously held global leadership roles at Amazon and also hosts the New Frontier AI for eCommerce podcast. And this is our returning guest, Max Sinclair. But before we get to Max, we'll have a word from our sponsor. Are you spending more on Amazon ads and getting less return? There's a smarter way to grow and without increasing your budget. Levanta's Affiliate Shift Calculator shows what you could do if you really allocate just a portion of your ad spend into affiliate marketing. If you're an Amazon seller doing over $5 million a year, This is for you. And guess what? It's free. All you have to do is click on the link in the description to go to your custom forecast today. So sit back, relax, grab a cup of coffee and welcome to the show, Max. Speaker 1: Great to be back. It's been about a year or so since I was last on, so it's good to return. Speaker 2: It's been a while. What's new, Max? What's going on over at eComtent? Speaker 1: Well, we just launched our freemium model. I guess that's a big exciting news. We'll be talking about that. Yeah, like sellers can log on for free and they can get a audit, the most comprehensive audit of their Amazon catalog, I think available anywhere. Yeah, like we, as you may know, we raised some, some venture money and people wondering what we are planning to do with this. And basically we're going to be spending a lot of GPUs for free on Amazon sellers and hopefully giving back to the community that's been really great to us. And so yeah, if you want to see how your catalog is optimized for A9 and RUFUS and COSMO, you can have a look for free on eComport. Speaker 2: You just hit a nerve. So, last podcast or maybe two podcasts ago, we had a question and I said I would do some digging on this and I did about an A10 algorithm and my answer was there is no A10. So, can you just tell us a little bit about that? Speaker 1: This is a classic Amazon guru speak of, oh, there's a new algorithm, it's A10. Of course, the Amazon algorithm is shifting all of the time, but it's called A9. Because Amazon acquired a company called A9.com. They were a machine learning company. I think they acquired it in the early 2000s. And that's why we have the A9 algorithm as it's called in Amazon, because it's just an external purchase. They never had an A8 algorithm. And there is not an A10 algorithm. I think some people may be confusing COSMO for A10. So, you know, maybe people talk about it in that language to kind of make it sound like there's been an evolution, which of course it has. But yeah, A9 is called A9 because that's the company that Amazon acquired. Speaker 2: Very good. So thanks for that. I know that a lot of people, I, a matter of fact, I wrote a blog article two years ago about A10 and I, it was completely, I misunderstood and you know, I, I, I deleted it after I found out. I'm not sure who it was. It might've been Danny McMillan that said, Hey, You know what? This isn't quite right. So very politely. And it's true. It's the name of the company. So I just wanted to get that straight. So there is no, there won't be an A9, there won't be an A20. It's just, sorry, there won't be an A10, 11, 12, 13. The name of the company is A9. That's not changing. There's going to be updates. Speaker 1: Exactly. I think to be fair to the people who talk about Some people talk about A10 to try and get the clickbait and a YouTube video watch or whatever, but I think those who make the mistake in earnest, you could say that there's so much changes that are happening with AI, which we're going to talk about right into the Amazon algorithm, In many ways, you could misrepresent it as an A10 because it's such a step change in what's happening, but the name doesn't really make sense. Speaker 2: Right. Okay. Well, why don't we just get right into it and let's talk about how is Amazon employing AI to change search and discovery right now? Don't forget to subscribe and leave me a comment saying, I subscribe, and I'll personally reply to your comment. Speaker 1: Sure. So there's two key things. Which Amazon are doing in the AI space and these are, I'm sure many listeners are familiar, Rufus and COSMO. So Rufus is this ChatGPT style search that appears on the Amazon app and also In the search bar, sometimes you can see a lot of experimentation about where COSMO is cropping up, but that's kind of like a large language model, LLM-based interface that customers can interact with and search. And we have some numbers, you know, the million-dollar question that is debated in all these WhatsApps, and I'm sure also on your WhatsApp norm at certain points, is how many customers I'm actually using this kind of AI interface to search and discover products. And the answer, according to our calculations, is that it's about 14% of searches coming through RUFUS. And I can tell you how we came to that number. Amazon AWS put out a blog. So Amazon have not released numbers on how many people are using RUFUS. Obviously, that's highly sensitive information. But the AWS team did put out a blog explaining how much infra they had stood up to support RUFUS. And obviously AWS is selling tranium chips, kind of like GPUs. And they're kind of used Amazon RUFUS as a case study on the AWS website. And we have this on our website that you can have a look at afterwards as well of how much GPUs basically they use. And they use 80,000 AWS chips. So that gives us a bit of an insight into how much people are using RUFUS. Why? Because the whole AWS business model is scaling up compute to fit the needs of the customer and that's kind of the big innovation that AWS had whereas previously you'd buy a server and you'd have a fixed amount of bandwidth on that server that you could use. With AWS you can kind of scale up and down compute and therefore You know, on a usage basis. So the fact that they use 80,000 chips means that they needed 80,000 chips, right? Like that's why they use 80,000 chips. And 80,000 chips gives you about 3 million tokens you can process per minute. And if we work backwards from that and assume that the average query length on RUFUS is similar to the query length on Perplexity, which is about 10 to 11 words, and each word comprises about 1.5 tokens, because the relationship between words and tokens, the large language model we use, which is numbers, isn't 1 to 1. It's about 1.5 to 1 on average. So anyway, we assume it's 1.5 on average. That gives you about 15.75 tokens per query. And if you divide that by the 3 million capacity, You get 275 million queries a day that the AWS infrastructure can support. And 275 million queries a day is about 13.7% of 2 billion Amazon searches per day. So that's how we got to that 14% number, working backwards from the actual infrastructure that AWS use. I mean, there's a lot of numbers I've thrown out there, but yeah, I don't know what your thoughts are on those mass norms. Speaker 2: I thought it would probably be higher, and I was thinking around 20-ish percent, but I'm sure once sellers get used to RUFUS, that's gonna skyrocket very quickly. Speaker 1: I agree, and I'll caveat this in saying this article was from October 24, so the number could be higher now, and I think you're right, because if you think about this, a logical comparison would be Google gets 8 billion searches a day. That's kind of like traditional Amazon search. And ChatGPT gets about 2 billion messages a day, which is kind of like the AI search. So that's 25%, right? So that would directionally feel like the right number. I mean, the maths that I laid out puts it about 14%. So I think you're safe to assume we're somewhere in the region of 14 to 25, roughly, people using RUFUS. Speaker 2: All right. And why is Amazon leaning so heavily on AI right now? Speaker 1: Well, you know, as an ex-Amazonian of six years, as you, you know, kindly put in, announced at the beginning of the podcast, It's within the culture of the company. Have you ever read a shareholder letter from Amazon? Speaker 2: No, I haven't. Speaker 1: They're fantastic and I highly recommend you and everyone else listening to read them because it's a way that Jeff Bezos and now Andy Jassy Don't just communicate with their shareholders, but also they use it as a mechanism to communicate to the entire company. And they write these brilliantly insightful letters about how they're thinking and, you know, at a super high level where the company's going. And there's two quotes that I would like to extract from these letters, which I think show why Amazon is going AI first, all in on AI. So in 2016, Jeff wrote, and this is a classic Jeff quote, customers are always beautifully, wonderfully dissatisfied, even when they report being happy and business is great. Even when they don't know it yet, customers want something better. No customer ever asked Amazon to create the Prime membership program, but it sure turns out they wanted it. And I think that kind of speaks to the Customer's obsession that is a key tenant of Amazon. I mean, as a seller or sellers listening, you probably don't feel that, but like for the end customer, there's a huge focus on it and they want to invent and create the best experiences for customers. And that is AI. Like I think it's unarguable that a ChatGPT conversation is way better than, you know, doing 10 Google searches. And that's part of why they're leaning on it. Speaker 2: It's funny and I've just found myself just over the last week or two just kind of honing in on where I'm searching. I might do the odd because I forget to go over to Perplexity or go over to ChatGPT, but I've got so many conversations now that The AI is open all the time, no matter what I'm doing, and I'm always asking it for questions. When I'm watching TV at 12 at night or one in the morning, I've got it open asking questions, even to Siri, which I lean on probably the least, but I'm always asking just stupid questions I would never even think of. It was maybe, when was this song recorded? What were the, I forget this person's face, or wherever she was in other shows, tell me a list. Or, you know what I did the other day? This is so stupid. We're going down a stupid rabbit hole, I know. But back in the day, you're way too young for this. But back in the day, there was these albums that came out, and they were by KTEL, okay? And the one album I had was 22 explosive hits. And there was a few songs in there that came back to mind, and I said, I want to play, I want to create a playlist of all KTEL's. Speaker 1: Yeah, there's another quote I'd like just to throw in there on this AMA topic from the most recent shareholder letter. If you take one thing away from this podcast, try out eComtent freemium. If you take a second thing away, read these shareholder letters. They're fantastic and they give you an insight into Amazon. And in this year, Andy Jassy said, generative AI is going to reinvent virtually every customer experience we know. You'll see changes to the norm in search, shopping, personal assistants. Some of these areas are already seeing rapid progress. If your customer experiences aren't planning to leverage these intelligent models, their ability to query giant corpuses of data and quickly find your needle in the haystack, a bit like what you're just saying, their ability to keep getting smarter with more feedback and data, and their future agentic capabilities, you will not be competitive. I think him stating that is almost talking internally. He's saying to the teams, as an ex-Amazonian, I can tell you we'd read these and you would think about this as if you want to get promoted or you want to do well in the company, you've got to embody what they're talking about. In the few times that you interact directly with the CEO like you would from these letters that they write. And yeah, AI is going to reinvent the customer experience at Amazon. And that's just part of the DNA of the company. They love inventing. Invent and Simplify is another one of the key leadership principles alongside customer obsession. And it's going to be even more, we're going to see even more changes than what we've already seen. Speaker 2: You know, for all the listeners, if you've ever listened to Max talk before on the other podcast he's been on, he's been on a few of them. He speaks as an Amazonian, but you were actually in the AI department, weren't you? Speaker 1: I worked, yes. So I worked in search. One of my specific roles was I'm here to talk about ensuring browse quality and customer experience for Amazon's launch in Singapore. So basically ensuring the customers could discover good products and the catalog work well. I know very well the A9 algorithm. I know its benefits and its drawbacks, and I can tell you how the A9 algorithm works. I mean, I think this... Speaker 2: That was my next question. Speaker 1: Go on. I think this is common knowledge, but I can kind of confirm it, which is you would stack rank the results based on keywords. So you would look at various signals and the actual secret source of The actual formula was secret. They wouldn't tell you like what the weightings for each of these were, but broadly you had price, sales velocity, in stock, conversion, you know, 20 signals maybe of what a good product is. And you'd have some secret formula that would work out on each keyword, you know, like office chair, what was the best product and what the ranking should be based on the various weightings of these signals. And that's how the algorithm worked. This is totally, totally different now, what's happened with COSMO. Totally different algorithm. Speaker 2: So based on that being totally different, tell us some of the changes. So we had the, I don't want to say the traditional, but we've had the algorithm from the past, now moving forward. Speaker 1: I would call it deterministic AI, right? So you'd had deterministic AI where you would you know the stuff that we've all we were all used to where you would crunch numbers and and you know computer crunches losing numbers looks all these signals across billions of data points. Here's your output. It's not it's not doing anything generative or creative in there. It's just like there you go. Speaker 2: How personalized are your Amazon listings now? Max explains the pregnant woman slipper test and how COSMO changes everything about search results. How does the How does Amazon Seller really adapt to all of these changes? Speaker 1: The key change is the introduction to COSMO. As we were saying at the beginning, maybe some people confuse COSMO with A10, but COSMO is Amazon's common sense algorithm. Basically, they launched this, by the way, before RUFUS. Related but separate thing, RUFUS is kind of this front-end experience that people can talk with and get product recommendations. COSMO is they've put a large language model also in the back end. Why did they do this? They did this to increase the personalization fundamentally. So each customer now gets a super personalized and in my view increasingly more personalized you know view of Amazon and the famous example that they have in the COSMO paper is kind of the pregnant woman searching for slippers. They will show her slip resistant shoes because they know that that woman is you know, they'd have the date of that woman's pregnant and you know, the common sense LLM AI algorithm will just say okay, if you have a pregnant woman, we're going to give a slip resistant shoes. So in that paper that came out in Q124, they say that they did a test on 10% of US traffic I mean, I spoke about this before in the podcast, but maybe if you missed it, they did a test of 10% of US traffic. It saw a 0.7% increase in product sales. And they say it in the paper, the rollout of this is going to be worth billions of dollars. And from everything we've seen in eComtent, you know, working with hundreds, if not thousands of sellers across all, you know, every locale, It's rolled out everywhere now. Now we're getting into, you know, May 2025. And obviously, Amazon has grown, right? So you can see like that, that slight tweak of increase in conversion, and I'm sure they've made it better and better and better, as the algorithm has, you know, they've refined it, it is valuable. And the reason to answer your question that this is different, is because these, returning to what I was saying before, these large language models work on a token basis. They don't work on a word basis. So they take a prompt, they convert the prompt into tokens, which are numbers. They then compress the numbers because this stuff is super expensive to run. So Amazon has a huge, you know, why everybody is the most valuable company and one of them in the world, they compress that into the numbers, you know, they compress them, then it goes into the AI engine, then the AI engine sees all of you know, all of its training data, all of the data on the customer and outcomes of personalization. So there's been a fundamental technology shift in how the Amazon algorithm works from deterministic this kind of large language model generative based AI with COSMO. And that's, yeah, that's a big shift that that's happened in the last year or two. Speaker 2: I had a this week, had a new client came on, we were doing an audit. One of the things I looked at was When did you ever optimize your listing last? And it was years ago. This was a long, long, long title, over 200 characters, 223 characters. The bullets went on forever. Uh, so many duplicated keywords, uh, or even a primary keyword over and over and over again. It was old, old school, you know, listing optimization. So forget about the images. Forget all about that. But I went to him and I just told him, have you heard about, you know, not having these long titles anymore, not duplicating, should be 150 characters or less. And he came back and he was like, well, Don't do that. How are people going to find my listing? My keywords are in there. I said, well, yeah, put your main keyword phrase into the title, but Amazon is going to change. Whoever's looking at that, they're going to change that title anyways. He couldn't believe it. I was saying, yeah, your title is just as important. Give what it needs. Just give a highlight. And then with your, uh, the same thing with your bullet points, you can optimize your backend still. And, but you don't want to do that. First of all, it's a pain in the ass when you're on mobile, but second of all, you don't need it anymore. The algorithm is going to pick up and choose what it's going to show to the, it's almost on an individual basis. And so that was just like a lightning bulb, you know, just kind of clicked and it was, Oh, I don't have to have this ugly listing anymore. So it's, and that's what's happening. Speaker 1: A hundred percent. Amazon not only changed the title links and they rolled out some new stuff at the back end of last year, right? Like links with titles, links with bullet points, basically making it shorter, more compressed, but also fundamentally this large language model understands the context of your listing. And, and, and, and yeah, like they they penalize you and they said this, you know, actually in that in that term change. They will penalize you for repeated keywords. They don't want a catalog which is keyword stuff. I mean, they actually said that in the thing they rolled out. So, I mean, obviously, you know what you're talking about, Norman. You're 100% correct. I mean, I can talk about some ways that we, you know, that we would suggest people optimize. Speaker 2: Why don't we talk about four different ways that people can optimize their listing nowadays? Amazon's old ranking algorithm is dead. Max breaks down what sellers need to know about the new common sense AI model. Speaker 1: It's about intent. It's not about keywords anymore. And when we talk about intent, what you need to talk about on your listing is What is the usage? What is the functional usage of the product? What are the event that people or activity that people are using it for? What is the time or the seasonality when they're using it? What is the body part or like room location that they're using it in? And then critically, what activities are used for and who is the audience? The audience probably being one of the most important ones. Optimize for that intent and those, I'd say, subjective descriptors of who the product's for and why and when. And that is going to help the AI algorithm, which is now far smarter than any human, to surface your listing to the right customer when they're searching for it. Speaker 2: And now let's have a word from our sponsor. Tired of negative reviews dragging down your star rating in sales? Traceuse has your back. Traceview specializes in removing non-compliant Amazon reviews the right way. I'm talking 100% compliant with Amazon terms of service. And with over 11,000 reviews removed for 400 plus brands, they know what it takes to protect your reputation and boost conversions. And here's the best part. You only pay for performance. That means you only pay for reviews they successfully remove. No contracts, no monthly fees, just results. Plus, as a Lunch with Norm listener, you get two reviews removed for free. Ready to clean up your reviews? Visit TraceFuse.ai. That's TraceFuse, T-R-A-C-E F-U-S-E dot A-I. We're back. Hey Max, I have a question for you about the next step forward. I'm seeing like people are saying, oh AI is gonna do this two to three years out. Things are moving so fast, it's hard to predict what's gonna happen tomorrow. But to you, what does the future look like on Amazon? Speaker 1: So I'll talk about three things which I'm excited about. So the first one is agentic shopping. And we've seen with Google I.O., which was a few days ago, and actually eBay have just launched something similar yesterday, this agentic shopping experience. What on Google this looks like is it will basically, you give it a query, it will go search the web, you know, go look at all these different shops. And then you can put in a target price. So you can say, you know, I'm going to wait until that's $50 and then it will check out and you can pay by Google Pay. And I'm here to execute that transaction for you once it hits that price. So the customer will basically set up agents and they'll scour the web, find the stuff and purchase them for them if it hits a price that they're happy with. eBay doing the same. And my experience or my observation rather I'd say in this kind of AI race that we've seen in the last three years When one leader goes, the rest follow. So I would not be surprised to see some kind of agentic shopping experience on Amazon, maybe through Rufus, maybe it's part of the Rufus experience somehow, where they, you know, you can say, oh, I don't like this now, but I'm willing to buy this when it's, you know, hits $30 or whatever. And I think that's going to change the game almost for these pricing tools, like how, you know, how are you going to manage that? Like suddenly you change your price and loads of orders are triggered. Maybe Amazon will show you what orders are waiting and you can maybe decide to execute them or not. Who knows? But I think that's certainly one that I would expect. What do you think on that one? Speaker 2: Yeah, I'm not sure what to expect. You know, it just comes out so fast. I think there's going to be definite new advances. Today, we're going to be talking about how we're going to shift towards the AI and technology. I think it's going to shift so quick that over the next six months, it's going to be completely different than what we're looking at right now. Because first of all, like we started off with, we might be at 14% of people using RUFUS. Well, I think that's going to expand quite a bit. I think it's going to be a hockey stick statistic once people really catch on to it. They're always leading, you know, with doing things, being the first to do things, and the followers, the other 80%, will come in very shortly. Once that happens, I don't know what Amazon's gonna do, but they're on the leading edge of AI and this technology, so I'm sure they're gonna fire back with something really cool. Speaker 1: I think another one which was actually announced last year at Accelerate, but then we haven't seen anything since, but I imagine they're, it's funny they do these, it's very classic like tech company these days, they do this big announcement and then There's not actually a product, it's just like a declaration of like a vision. But they announced and accelerate these different search experience by category. So one of the examples they gave was pillow. So if you're searching for pillow, the whole bio experience is totally different and it will like educate you and say like, these are the things you need to consider like thread count, Maybe linen type, type of, you know, and they basically would change the search to be like some very category specific thing. So, you know, I think I wouldn't be surprised if something like that, you know, does come to fruition. I mean, they already said they will do it and it's probably a matter of time before they start to, you know, expand on that. Speaker 2: Well, yeah, buying patterns too. So, Based on, let's say that you're more of an upper income and you're looking for that 3,000, whatever thread is, you know, Egyptian cotton pillowcase, they're not gonna waste their time with anything that, you know, might be a thousand thread count or, you know, they're gonna be able to target you based on the demographic a lot easier. Now, they might be doing some of that right now, but I don't think they're doing it As well as what you'll start to see later on. Another one, this is gonna be a tougher question, I think. Maybe not. You're a smart guy. You're a lot smarter than me. So I'm talking about product launch. So right now there's a variety of different ways that people are launching. Forget about PPC for a sec. Can you see how people can use AI to help with their new launch and rank strategies? Amazon's algorithm works completely different now. Max explains how AI is changing what shows up when an individual customer searches. Speaker 1: Yeah, so I think like I know the traditional launch approach was like let's find some keyword gap and let's you know, let's say that I We found this on this keyword, this competitor's low. I don't necessarily think that's a winning strategy anymore. I think the whole keyword basis is fast evaporating and that won't be the case. Innovation is still going to be important and solving customer needs of new technology and new invention is going to be important. We talked about Danny McMillan. I read his whole article on the honeymoon period and how it's a bit of a myth. I found it very convincing personally reading that and his dissection of it. I think it's going to be about innovation. It's going to be about identifying your target audience, you know, and solving a real problem for them. So spending time with that target audience, speaking to them, understanding why the competition, you know, why it didn't work for them, whatever that competing product is, and then launching something that is truly innovative. And when you do that, I would highly recommend that you follow what we talked about in terms of optimizing for COSMO, talk about the audience, talk about the event they're using it for, talk about the problem they have, all those things, right? Like just good solid marketing things and you're a far more experienced marketer than I am. And I think that's what's gonna help people succeed with new product launches. Speaker 2: So just really old school marketing, going back, reaching out to the customer, finding out what they want, what they liked, what they didn't like. And implementing it. Speaker 1: I think another like a meta takeaway point to think about these algorithms is that they're far smarter than humans now. So like the old way of like, oh, let's like hack this kind of thing with like this like missing keyword, you know, like forget it, right? Like you're not going to kind of like out hack an Amazon algorithm now with like a bit of Helium 10 or keyword, you know. I won't name names, a bit of a keyword tool, kind of like gap analysis. These algorithms are super smart and therefore, yeah, if you solve real customer needs with new and innovative ways that don't exist, they will for sure reward that. Speaker 2: All right, so let's jump over to PPC now. So how's that going to be impacted by AI-based search? Speaker 1: Everything we focus on at eComtent is all about organic ranking and ranking for the algorithm. So I don't claim to be a PPC expert. And at the same time, there's a reason we focus on organic is because I have a very controversial view on PPC, which is I think PPC is going to look like a very dated I think like we'll look back on like pay-per-click specifically blue links pay-per-click marketing as something which was very like 2000s to 2020. I just really believe that the future will be you talk to an AI engine and the AI engine gives you the best answer and the monetization comes from much higher conversion rates that we saw in the COSMO paper. And I think there's a bit of a game here, which is like ChatGPT has launched ChatGPT Shopping, Perplexify has got a shopping integration with Shopify. So Amazon have to build the best AI You know, chatbot and agent to help you find the best product. They have to do that. Otherwise, people will move to, you know, Google as well. Launch Google Shopping, right? So there's going to be this many competing companies for the best AI agent to help you shop. And I think anyone that delivers ads as part of that experience is not going to be the best one and will lose, in my view. And I think that people will be used to asking a question and getting the answer. And that's just what's going to happen. I think what we have seen Amazon experiment with, which I think is very, very clever and credit to them, is sponsored questions. Have you seen this, Norm? So brands, you know, it's obviously in beta, but it's live. Brands can bid on a sponsored question. So if I ask, what's the best headphone? And obviously, you know, Amazon goes through COSMO, it looks at my age and my spending habits and all service to me, like, you know, headphones, which it thinks I'm most likely to buy based on my past history, blah, blah, blah. Brands can then bid on the follow-up question. So you have a follow-up question bidding saying, why should I, you know, why are Sony the best headphone brand? And Amazon launching these, and I think it's a very, interesting way for, you know, there's still to be some paid activities and brands still to pay to be in searches, but the experience for the customer to be non-intrusive, and you still ask your question and get the best answer. Speaker 2: That's brilliant. You know, as a old guy, I think of that as, I'm Canadian. So the hockey game. I used to go out and back in the 60s and 70s, we'd go to a hockey game and there'd be white ice, white boards. And then all of a sudden board advertisement, you know, came and you start to see it and oh, they're wrecking the game. Then they'd have it in center eyes or just between the blue lines and then in the circles and then starting to, it's not like in Europe with soccer, but, or football, but all the advertisements, you know, now they, I think they can have one advertisement on their, their shirt, but it's just exploding. And then you go, Where the hell can they go and add more advertising? And then all of a sudden on the steps going up, you know, and this is what Amazon's doing. It's actually very, it's brilliant finding the gaps where they can make more money. Now, I hate it with sponsored ads. Like, everything is sponsored ads when, you know, up front you get two rows. I mean, it takes away from, you know, good products. It's just people who have more money. But, and conversion. But this, that's something I didn't know about. I'm going to check into. Speaker 1: I wrote a LinkedIn post about it. So it's on my LinkedIn. But I do think advertising will change. I think it will become, you know, personalization has been a big theme of this podcast. I think stuff will be a lot more personalized. Amazon obviously have Prime Video. We've seen ads in that. I imagine they're super personalized. I think we're gonna have ads be reinvented in the kind of chat experience in a way where like a brand new ad will be created And today we're going to be talking about off the fly based on the customer and their prompts. So like something totally different. And I don't know how this will how this will happen. But I can imagine, you know, me as Max in London in his 30s searching for headphones. This is the best, you know, like it just gives you something which is like a paid thing, which is like completely like an AI generated ad, which is like spot on somehow. So I like, you know, obviously, I don't know how this is going to work. I imagine there'll be some new innovations that happen. I think sponsored questions is another non-intrusive way. So maybe you would have like, And on, you know, the organic answer and then like the AI generated, you know, like ad answer and it's done in some way, which is quite clean and nice for the end user. But I think ultimately, like we're moving now into a world which is very healthy with a lot of competition, with a lot of different providers, you know, new players like Perplexity, the old guard, like Google having to reinvent themselves and do new things and, you know, ChatGPT also being a new one. So I think we're moving into a world where this business model will change and I'm convinced of that. And that's why we only focus on organic at eComtent. Speaker 2: Great. Two quick questions. One, if brands are making a mistake, What is it? What is it right now? Speaker 1: Brands are making a mistake if they have not changed their strategy on how they're generating listings in the last two years. Because the technology has changed so much, if you haven't changed your strategy, you're making a mistake. Speaker 2: All right, last question. If there's one thing brands are missing out on, what is it? Speaker 1: Can I plug myself again? Otherwise, I'm going to have to think for longer. Speaker 2: You do it. Go ahead. Speaker 1: I'm trying to think. Yeah, you're missing out on our e-content freemium. I'm trying to think what else. I don't know. If I'm not trying to plug myself. Maybe your podcast, Norman, your podcast is full of pearls of wisdom. Thank you, sir. Speaker 2: All right. I know we got to cut out of here a little bit early today, so why don't you just give us your contact information, then we'll head over to the Wheel of Kelsey. Speaker 1: Absolutely. It's Max Sinclair. I'm most active on LinkedIn. I'm in Norm's WhatsApp group, so you can tag me if you have a question about this. I'm happy to answer. If you disagree with me, tag me and we can fight it out in the WhatsApp. And yeah, ecomtent.ai is the business, so check it out. It's free, as I said. Speaker 2: Once again, Max Sinclair. Thank you. I can't wait till you're on again, because you always come just with all this great new information. Speaker 1: Look forward to it. I'll see you in six months or so, Norm. Speaker 2: All right. We'll see you later, Max. All right, everybody. Thanks for taking part in today's podcast, and we'll see you next Monday. Unknown Speaker: Lunch With Norm.