Using AI to navigate public Amazon data | Amazon X AI Working Sessions

Good morning, everyone. If this is your first session with us, I'm Andrew. I'm co-founder and CEO of Reason Automation. Thank you for joining us for the final day of our Amazon x AI working sessions. So with that in mind, I'm gonna shut up and say, let Scott introduce himself and introduce his topic.

Awesome. No, thanks, Andrew. Genuinely excited to talk about what I have, uh, today. And, uh, this is something, you know, like everything, it's, it's, it's new, but I have, uh, actually made impact in this with my business. And, um, so I do know, again, there's takeaways, but there's also just a new way of thinking about, um, uh, We're gonna talk about AI visibility. And so there's lots of different ways that, you know, I'm sure, you know, this is like session number 9 that like AI is impacting e-commerce. And the one that I wanna share, I'm certain it's like, it's already here. And I don't think people think about it a lot. It's not the first thought when they think about like, you know, this new era. But just a quick about me, I started out as an Amazon seller, scaled a reseller business. But when I said like, when I say like I sold on Amazon, I was really just a software developer. And so I've turned that into, you know, Smart Scout. So I've been running that for about 5, 6 years. Shoot, I was going to take this out, but this is just my claim to fame that one of my products, we sold tens of thousands of products on Amazon. Um, one of them was called Liquid Fire, and it was the reason that Amazon got fined by the FAA. So, uh, this is, uh, my claim to fame in the FBA world. Uh, uh, fortunately for me, um, very lucky, you know, uh, we were not found culpable. Amazon was, you know. So that's, that's just a funny little story, uh, uh, of all the types of things that I've tried to sell on Amazon. And You know, I, I usually think more about like the different, uh, data levels. Uh, this is what I nerd out on. But, uh, more lately I'm going to talk about, um, AI visibility. Here I have it as consumer product discovery. So let's back up. Uh, what— when ChatGPT was— it was like November 2022. Um, and, uh, there's been so many ideas and so many different ways that, um, AI is impacting, your day-to-day life. I'm gonna— Andrew, help me out here, but I, I assume most people here are managing e-commerce in some way, managing marketplaces.

Maybe they have, uh, a D2C component, uh, maybe it's, uh, Walmart TikTok shop, you know, there's, uh, but yeah, I think we've got the full pantheon represented here of folks who are all the way from, you know, smaller independent sellers who are mostly just solo entrepreneurs all way to folks who are representing larger enterprise e-commerce teams. I recognize several of the attendees, and I know that they're representing, you know, tens or hundreds of millions of dollars in vendor sales or seller sales.

Yeah. And, you know, your role in an organization, whether you're a solo operator or you're part of a big team, you know, some of these may have a bigger impact on you. And I put this together. This is totally my opinion. You guys can fight me on it. I, I could be totally wrong, but I do have at least opinions on the final two. So I'm In terms of content creation, like, it is clear that, uh, creating compelling content, either text or image, it's here and people are able to like do some cool stuff. I'm sure you guys have learned some things. You may be past me on what your capabilities are. Product development, yes, it's here. Market research and analytics, absolutely. The ability to like combine, combine datasets. We're no longer encumbered by a, a, a user interface. We can just You know, take the data, ask a chatbot, and, you know, before we know it, we learn. Now, last year, I feel like 2025, you could not escape this idea that people were talking about, about agentic commerce. That's basically, you know, think of chatbot just buying for you, buying like preemptively. Maybe you never have to see— maybe you have to never enter your credit card information. You never have to click the 4 buttons to actually get it to your house. And it's clear, and from— I mean, people have done this, but it's not here. It is not widespread, um, and we don't know exactly what it's going to look like. I think, um, what is here is this next one: consumer product discovery. I know it's here because, like, it's impacted my life of, like, how I take— if I'm making a serious purchase, whether, uh you know, obviously the high ticket, you know, cars. I'm going to show off a treadmill that I bought and my process of learning how to buy what I wanted to buy. And so this product discovery, the flip side, the e-commerce operator is going to think about how do we make ourselves visible in this product discovery.

So— And sorry, Scott, really quick, if you don't mind going back, I have a quick interjection that I just can't help myself on because I am a I, my last few roles at Amazon were actually in Alexa Shopping. Okay. And, you know, we were trying to build the actual like voice commerce function and whatnot. And the, if, if anybody here in the audience watching it was also at Amazon at that time, then you probably have a lot of PTSD flashbacks to the promise of voice. Uh, you know, every Prime Day had voice deals and yada, yada, yada. And everybody was so hot on the idea that like that was where the last transaction Completion was going to happen and it was a complete wet firecracker in exactly the same way that my gut tells me in Gentec is going to be for the next foreseeable future. But you're 100% right that all of the supporting things, including product discovery, are already here and super, super prominent. So just huge plus one to your point here.

No, no, no. That's, that's funny. Cause like Andrew, like that was one of my first feelings as well. When I see like a Gentec Commerce, I'm like, oh, this feels very close to Alexa voice shopping, which. Interesting and feasible, but like, will people adopt it? Will it become— I mean, you know, you talk to any, uh, uh, you know, most of, uh, many, many people's wives, like, shopping is like the fun part, right? Yeah, seeing it, checking out. Oh my gosh, I don't even want to go into this. I saw a Korean, um, app that basically it's a video game that all you do is just shop. It's all fake. Yeah, but the game is just like, you just choose what you want to buy and you just like, um, so, um, what a gentle commerce becomes, I don't know. But, uh, so, uh, I just wanted to like ground us around some, um, uh, some data points that, um, according to a study from, uh, Packview, uh, 20% of conversations, uh, show shopping intent. And that's kind of true for me, even if like sometimes I'm like totally top of the funnel. I'm just like, just like, you know, curious, like, is creatine impactful? We know if you say something about creatine, like, you know, you're only a few steps away from being at the bottom of the funnel. Um, and so, uh, you know, so much search is happening here. Now I'm gonna just take a story like, uh, this happened to me. I think February this year, our treadmill broke down and I'm trying to show my wife that she could trust me with a big purchase. This is like $3,000 to $6,000 was kind of like our budget range because we're hoping for it to last for 20 years. And I really wanted to get the best that I could with that dollar. Now, So I just ask questions. What's the best treadmill out there that has iFit or Peloton content? I wanted like the, you know, the Run With You video kind of experience. And so already out the gate, I'm asking some very specific questions. And, and a note here, you know, I ask one simple question. Here is all of the citations that ChatGPT ChatGPT is bringing in to make more informed, uh, content. Like, it kind of needs this, like, this refresher, this context to give better answers. If you use ChatGPT in 2022 and 2023, you knew that, like, it's like, hey, I'm not internet aware. If this was, uh, was, um, if you're talking about a current event like, I'm not trained on that data. Um, they kind of like fix that by doing searches. Now, you know, I didn't ask just one question. I asked multiple questions, and I wanted like dynamic, in the weeds, you know, tell me about the durability or firmness of this, compare two versions. I probably, like, I think I asked the same question in, in very different ways because I was like, okay, you know, there's NordicTrack and then there's Matrix. Matrix is what you see in gyms a lot, but like it's a little bit more expensive. I, you know, is it worth the, the next $2,000 or to get to the upgraded model? You know, um, all of this decision process, this learning process, this discoverability is happening inside of, uh, ChatGPT. This is a— it's crazy to think that, like, um, that's why I say this is, this has arrived, this is here. People are doing this constantly in, uh, in categories. I'll be the first to say some categories are more likely than others, and probably the more expensive your product, the more this matters. But, um, then again, you know, there's the Some customers, they just want to get it right. They don't want to, like, they, you know, for me, it's like a, it's a risk to my wife to just like get the wrong treadmill. And even to the point that I'm asking for a coupon code. So, you know, real bottom of the funnel, uh, types of, uh, questions. Now, um, we're going to talk about a little bit about just like, just to refresh everyone so we're all on the same page of how to do LLMs build out recommendations. And I, I saw this and I thought this was really interesting to see that, you know, there's a correlation between AI visibility and SEO. We'll jump into that a little bit. But Claude's not using Google Search to find the citations. They're using Brave's search engine. So it's kind of like, it's very interesting to see that, you know, they have all this training data., you know, billions, almost trillions of data points, but they want that fresh insight. And so they're doing searches. So there's almost like a search engine component. So that's why I feel like SEO is related. It's like, you know, what, how is your product positioned? How's your category positioned? Through search. And I saw even last week, uh, Amazon's Alexa is also bringing in citations because they want to know, you know, the latest, what's like, what's going on. Uh, so the— and, and Claude, which is like totally trending as everyone knows, um, it uses Brave's search engine. They don't use Google. So, uh, very interesting. And so like, you know, SEO for Brave would be more effective than, you know, Google SEO. Again, this is just to illustrate what an LLM is. It's just kind of trained on all of internet data and it uses that to predict, you know, in the underpinnings of AI, it's like, what's the next best, you know, phrase or word? And it's trained on internet data, whether that data is correct or not. I am certain people are having a heyday creating fake internet data only to see that get served up, um, and our responses. I'm sure, um, you know, you've probably seen people like make fake, uh, edits in Wikipedia, and, you know, some of them have stayed there for years. And so that's certainly happened. But like, just know that like the body of internet data is the source. Now, um, some sources are better than others. This is a study from Profound that found the share of top 10 websites, uh, in the citations of— when I say citations, just know that this is just an article or a link that, um, that these LLMs are bringing in to have like the fresh data. So I think you— what, there's a— this is dynamic. It's not always 48% Wikipedia, Reddit. I've seen updates to this, but it's a good way to think of how it ingests it. And most of the time, you know, these are high-trust websites, uh, you know, very nerdy stuff— NerdWallet, TechRadar, G2, um, and Forbes, Reddit. You know, these are not, um, small-time blogs, but, you know, um, news publications that have, you know, a fair amount of authority or credibility. Okay, so I just want to show, um, uh, this is actually using a tool called Surfer SEO. Uh, I've seen this happen a few times. Where overnight— this is the day that an article went live and hit, uh, you know, like the top 10 search results. And overnight, your— the mention rate, you know, um, was in the zeros or tens of percentages and then just popped up to 50%. So, um, this illustrates this perfectly, that like, you know, If you have like a good SEO strategy and you have good content that like it gets, gets, gets brought in, it's going to impact visibility. So it feels like we're— this, this whole thing moves us beyond keywords. This is probably the paradigm pre-ChatGPT. Or even, you know, still going on today. People are there. And, um, here we have a top-selling product. And man, Treasure Box Toys for Classroom Birthday Xmas Gifts Goodie Bag Mini Animals Mochi Squishy Stress Relief Sensory Toys Random. Um, lots of keyword stuffing going on. And, uh, is this stuff gonna be answering specific questions of like— I mean, I, I did this yesterday— gifts for a 2-year-old. I don't know if this gives that specific context that's helping. Um, and, uh, so when it comes to, uh, you know, surfacing visibility, uh, context and specificity, it's king. Best protein powder for sensitive stomachs. Travel stroller that fits the overhead bin. I 100% know that question gets asked. Uh, non-toxic cookware for induction stove. Gift for a 10-year-old who likes science. Dog bed for large senior dogs with arthritis. Those are the types of real questions that customers would ask. Um, not just dog bed, not just gift for 10-year-old.. But you know, the, the, the search, the way that consumers are using it is adding like just a little bit of context and hoping that these LLMs, they know because they're using all the body of data on the internet. So now this, this all builds up to this question. So like, are you visible? Are you, um, what, what, how can you see if you're successful here? And I think, um, you know, you, you can measure this yourself. I probably do this, you know, like on my phone once a week. I ask a specific question, just wondering, did my product make the, uh, the cut? And that's a low precision way Certainly you can learn some things, but the, the thing about these LLMs is they are, they'll give you a different answer every single time you ask. And so to be more accurate, to measure visibility, you have to like pull it 10 times, 20 times, 30 times over a course of a few days.. And from that you can have a statistically, you know, viable, like, you know, hey, we're on average, we're 10%, now we're 50%. Uh, so that's the kind of like framework in which I've, uh, worked and built around. Um, and like, again, like people, you like, you could do this yourself. You don't need anything, but there are tools. Um, I'm happy to share everything that I've used. I've, I've talked about Profound and Surfer SEO. Those ones are very good at Google., uh, LLM visibility, uh, or, or like just the generic, um, but I'm, I'm, I'm always like, I'm Amazon first, so I'm like, how is this working in the Amazon ecosystem? So let's ask specific questions. Best unflavored micronized creatine monohydrate powder with 5 grams of serving. I like creatine give me the best. Or, which creatine powder is recommended for both muscle gains and cognitive function? So, you know, you can be very broad or you can be very detailed in the prompt. And what you want to do is pull that, you know, as often as possible to get an idea who is, who are the winners and losers in this game.. So here's the results. Thorne Creatine naked and this German creatine. And if we're look, you see they all have micronized as a part of their title. So I'm not a creatine expert, I just know a few of the terms. I don't know what micronized means, but these guys are targeting that and because of that they're visible 100% of the time. Every single time that we asked ChatGPT for the Amazon recommendation of micronized creatine, these guys were there 100% of the time. Now there's some people that are here most of the time, you know, in the 90s and the 80s. Then there's some people that are there half the time. So, you know, Obviously, if there's an impression to be made, an upper funnel conversation, or even bottom of a funnel, like higher visibility is better. And I'll jump into the best of my knowledge how to influence that. We've covered, like, I think you already have, you know, as we— some idea. It is the body of the internet, you know, hundreds of different sources, but some are going to be more impactful than others. Then you can do this at a brand level. You know, Optimum Nutrition has 4 products that are getting surfaced, whereas Thorne only has 1. And these are kind of like the different ways to think about how people, whether they're using Alexa, Claude, ChatGPT, how they're discovering and learning and seeing products. So, um, now there's— I, when a year ago, it was actually a little bit less than a year ago, I didn't even think this was possible because I was told, um, Amazon's blocking ChatGPT. And I, I don't know every— you know, this, this is dynamic, but, um, I was able to look inside of like the reasoning that ChatGPT did and it's absolutely doing Amazon searches. So Amazon's blocked Agentic Commerce, but they haven't blocked at least, you know, the same type of scraping that Google does, just learning about the product.

They—

I'm sure— I know they block like the checkout pages and they've sent cease and desists. But like, here's an example right here. I want to buy creatine. I'm not saying anything about Amazon here. I want to buy creatine. What's the top recommended product? Here it is: Thorne Creatine on Amazon, or On Target, uh, Transparent Labs Creatine on Amazon, or you could go to their official website. This is ChatGPT just trying to answer the question the best way that it knows, and it's, you know, it knows everyone— many people want to buy on Amazon, or a lot of the links to creatine products are on Amazon, so it's going to, it's going to send traffic there. So, um, I don't— I mean, this is dynamic. Like, I don't know the exact relationship between, you know, uh, a search and an Amazon purchase, but like, this is, this is reflecting reality today. I, I put this together as kind of like a, a really interesting way to think about um, you know, the estimated monthly revenue of a creatine brand, and— but it's— or contrast that with its visibility. So I ran another, you know, search just on creatine and found that some are more visible than they have revenue, and some have more revenue than they are visible. So, you know, this could be— I kind of think for some of these down here, like Sports Research or Bulk Supplements, this might be a leading indicator that like, hey, they're relatively high visibility to their revenue. And I would expect their revenue to climb as, you know, people keep searching and learning. Right now, you know, you can't listen to a podcast without getting a creatine advertisement at some point. And so, you know, people are taking those ads, they're going to LLMs, they're learning, and ultimately, you know, maybe they'll be like, hey, Bulk Supplements looks pretty trustworthy to me. ChatGPT mentions it, so I'm going to trust it. So I was also curious, like, the same thing works for treadmills, you know, high consideration product. This answers kind of my question, you know, what are the, uh, what are the best treadmills? Who's getting the most visible? NordicTrack. Um, Peloton is down here at number 6. Uh, and this, this really does work for, for any category. Okay, now I spend a lot of time talking about, you know, the, the buildup, the, uh, the problem set. Your visibility. But how do you make a difference? In every single ChatGPT search, in every, uh, uh, Claude search, they come up with citations, um, or sources. And, um, those sources will mention specific products. And this was— this one's done for the treadmills. I can tell that because here we have nordictrack.com, and, um, how many prompts it was. This kind of like organizes that data. How many ASINs did we find there? This shows that usually the top one is amazon.com. ChatGPT is looking at amazon.com. And, um, and then how— and then the frequency of the responses. So this is most— almost like an indicator of like, you know, if it's, if it's pulling up in every single, uh, result, you know, 20 times, 12 times, 8 times. That's a relative strength of that source or how often it's getting checked. So, um, I, you know, runnersworld.com, very interesting. That's just like a, a, a, like a publication, like a, not a newspaper, a magazine, and it's influencing how So whatever runnersworld.com is saying is influencing the products that it gets mentioned. Um, I— this barbend.com, I think if I go back, yeah, it was the third most cited. And here we have this article, the 9 best treadmills on Amazon of 2026. Again, I can't— I do need to emphasize again that like every category will be slightly different. How do people searching, uh, your product— I personally would prefer to buy on Amazon, um, and this one is all treadmills on Amazon, and, uh, and it's got the 9 best ones. You're either here or you're not, and what it says about you is going to impact what it does. And so Um, you know, even has like their recommendations: best overall, uh, best with iFit, best for running, best manual, best smart treadmill. And it gives different recommendations. So, um, the truth is, I think when it comes to like this off-Amazon effort or your visibility, um, it's hard to make overnight success. But, um, in one category that I was working on, we had a product that was like, I think, the 9th most visible. And then we reached out to all these publications, um, and had about 10 to 20 different, uh, you know, content updates on those publications. And like, I'll be honest, like, it was a labor of love. And we went from 9th in the category to 3rd. Most visible. So, um, that is some of the effort required. And, you know, this may be out, um, of like the specific vertical you're in, but there's no question that it does— that it impacts it. Now I'm going to talk a little bit more about what you can do on Amazon. So this, uh, research paper about Amazon Cosmo taught us a lot about How's Amazon approaching this? And Andrew, I, uh, you know that you worked for, uh, Alexa with it. I'm wondering if like the, the term Cosmo itself is going to go the way of Rufus and it's just all going to be called like Alexa. I don't hate that from a branding standpoint. I think you're muted.

Yeah, yeah. If I had to guess, yes, it's all going to go the same way, um, ultimately collapsing into whichever brand Amazon feels like is actually makes the most sense from a consumer perspective. Yeah, I—

yeah, it's kind of like, uh, Google's foray into like Bard and like their different like, uh, AI, and then they're just like, okay, we're just gonna be Gemini. Like, you know, you want to consolidate, um, AI.

Yeah, it's like AI already has a naming problem, you know, like how many different, uh, Claude models are there and which versions of those, and ChatGPT had the same problems. So let's, let's go for simplicity, please.

So yes, so, uh, Cosmo, or now Alexa, uh, they talked about, um, what, how they are going to ingest information about your products that ultimately lets them understand, um, and get context about your product on Amazon. And inside this, they say like, hey, There's 15 different questions that we want to know about every product on Amazon. So clearly you want to make sure that you're answering those questions. Here they are. Like, you could screenshot this, you could actually just Google to get to that Cosmo paper and, you know, find those questions yourself. And they may seem obvious But you'd be surprised how much, you know, how few Amazon products are answering all of these questions. Like, maybe they're asking most of— answering most of them, but, uh, they, they, um, yeah, there's a lot. And you can answer them in your text and your images. So they're looking at, uh, you know, what is the product, who is it for, when do they use it, like, um, what type of, you know, what role does it play in the larger experience, what products is it paired with, um, what lifestyle or interest does it support. And already you can see why I, I feel it's not very hard to understand, um, you know, the specific types of consumers that might be a fit for one product over another. If, you know, you, um, want a softer touch treadmill, or if you want a, a very firm treadmill, uh, you would need to say that, and you need all the content on Amazon and off Amazon to kind of reverberate or, uh, to, to echo, um, what, uh, what, what your product's known for. And so I, you know, just as, as a, as a, as a tester or someone that just likes to— like, I, I've, I've seen this visibility, uh, play out, and I wanted to take some of my products and just like, ah, ask the question, hey, are we answering this? This is a, uh, a SanDisk Ultra Micro memory card. And, um, we ask the 15 different questions both on a written and visual basis, like, are we answering the questions? And, uh, overall across the 15 questions, like, this one gets an 81 out of 100. The good news is, hey, there's room for improvement. I probably don't want to get 100 out of 100 on my products because, um, then it means like there's probably not too much I can do. Um, and so, um, if you want to reach out to me, like, I'm happy to like nerd out on like how to score, uh, products. But ultimately, like, my goal in this, uh, presentation was just to talk about this issue of visibility and how, uh, it's, uh, like, it's already here in your category. In your category, it may actually be very impactful, more than you would like. If you're in treadmills or cars, like high consideration, um, or even like kitchen appliances, I guarantee you, like, it's, it's there. Now, if your product is less than $10, you know, maybe people aren't researching it as much on ChatGPT. But, uh, but unless it's a consumable, of course people, uh, care what they ingest. And so I don't know, uh, very interesting. Uh, I am excited to see where this goes. I think we're kind of the beginning phases of visibility, but I mean, I'm, I'm, I'm convinced, like, it's, it's here, it's impacting lots of, uh transactions today.

Yeah, absolutely. Thank you so much, Scott. I feel like there is so much there that I and everybody else has questions about to follow up on. And accordingly, we've got some questions for you. So the first one, thinking back to the beginning when you were talking more generally about just tracking LLM visibility, you mentioned a bunch of tools, but I just want to make sure that for anybody sitting at home trying to write all these things down, Um, what were some of the tools you mentioned about tracking if or how you're showing up in LLMs versus just kind of going in and checking yourself?

Okay, so I, I think about this, um, so if you're familiar with like the Google SEO tools like Ahrefs, SEMrush, um, this is kind of like, this is like the SEO of like this decade. It's kind of evolved into AI visibility, and there's so many tools. I can tell you some of my favorites, like the ones that I pay for and use. And, um, some of them are going to be better at like, um, just surfacing like your categories visibility. And, um, some are a little bit more geared towards e-commerce, like they might have a bit better— I think Profound has like an e-commerce angle to it. Um, I've worked on some of this a little bit myself, but I'm just in the Amazon Uh, Lane, I use Surfer SEO. Um, it's very simple and it gets you straight to like— I like to surface the most, uh, uh, visible citations. I mean, like, some of the tools that like I was already using, they just built it. Like, HubSpot actually has it, and we were already using it for very different reasons. Um, and if you type in AI visibility like software, there's probably about 50 answers right now. And I just told you some of my favorites.

Yeah, just to add on to that, the Scott's point about like, all your existing tools are probably building it in is really, really key. And this is a fun time because everybody's starting from the same point. And even though there are certain tools that do things a little bit differently than others, the best and most important thing is to use one. And the easiest one to use is one that you're already signed up for, one you're already paying for, and just get started with what it says before you start shopping around for like the quote unquote best tool. Just start using one.

And I agree. Like, I, I think, you know, just answering the question, answering a specific, you know, query or prompt about your category and then measuring it, they're all going to surface similar results. Yeah. Yeah.

Like, for a reason, you know, we use, you know, Semrush and Ahrefs, both of the tools that Scott mentioned earlier for SEO, because I've been an SEO geek my entire life. And to his exact point, you know, they rolled out AI visibility tooling and measurement like almost immediately. And like, is it any good? I don't actually know. I haven't done an exhaustive comprehensive thing versus everybody else, but the answers appear to be mostly the same. So yeah, just get started. One more. And I, you know, I'm really excited to hear your take on this one, but I think a lot of folks here are, you know, now swimming in a ton of information about thinking about how their products show up. And there's 45 different ways that they could kind of approach this. You gave them 50 different 15 different Cosmo questions. They're probably thinking about how do I go and manage all that in my catalog, etc., etc. And so in your perspective, like if you had to prioritize amongst that universe, how would you even get started? Like what 2 or 3 actions should somebody here take this afternoon to go and improve how their top products are being surfaced and recommended?

Well, I like— the first thing I would do is just ask a question yourself in an LLM and look at the citations. Like, be aware this is who's talking about your product in your category. And if there's like a powerful— like if there's a listicle, like, you know, like the best 9 treadmills, it could be worth building, you know, you're reaching out to them and giving them a reason to include your product. And because like you're, you're almost investing in the future so that the next generation of LLMs, it's training off of that. So, um, uh, that's probably like the first thing. Even maybe you don't want to reach out to them yet, that's fine, but just, just like see the types of articles, the types of citations that it's building out. Um, and then I also would look at Here's a great example. There's a grill company in my town and, you know, they— it's a very big category, probably high consideration. They decided, they're like, we're gonna win at tailgating. You know, there's like 10 different ways to use a grill. There's camping, there's this, there's that. Like, we're gonna win at tailgating. And so they made sure that their content across the board is tailgating. And so when, you know, someone's searching like, hey, it's like, you know, it's football season, I want to grill for tailgating, they're going to, they're going to have an easier time being known for something because they planted their flag. So I do think the features of your product, you're going to want to synchronize across your content so you get known for that. If you change the wording every single time, you know, it's not going to know what to make of it. And, uh, so it is a reason to plant your flag in a category and not try to be everything to everyone. If you're creatine for women, if that's like, you know, your niche, um, make sure you say that in your content, in your images, and eventually you'll be known for that in the LLMs. So yeah, those are two different, you know, early ideas that I would just take action on.

Yeah, that makes a lot of sense. And I can't emphasize enough how much I agree with just going and doing the search and asking the questions yourself. I feel like a lot of us have some hesitation about discovering like what the reality might be and are like, do I really want to know? And I see this all the time because I do vendor finance financial data. And a lot of times brands haven't taken the effort to understand what their profitability looks like because they're afraid of the answer. Or they, you know, they know that their stakeholders are going to get pissed off about the answer or what have you. And those are all very, very valid feelings and not valid reasons to not do it. And in the same way, doing that initial AI visibility research and being really honest with yourself about, you know, to Scott's point, who are you targeting? What niches are you going for what is the AI search experience for those niches and those types of questions today, and being really honest about where you rank or don't rank at all.

And if you haven't gone through that exercise, I've got, I've got one thing to add. So what's actually nice about this, specifically the measurability tools, is like, hey, there's a number, you know, you're in 5th place visible, you know, you're somewhere, which lends itself to like something that you can track over time. And improve upon. Like I said, like we had one quarter, our goal was to go from 9th place to 3rd place. And so you can work on it and make measurable progress. Like, you know how many things that we do on e-commerce are like less measurable? This one is measurable. You know, your, your, your product page like score. How are you answering these questions? How visible are you? You know, 30%, 60%? Do you want to like get to somewhere? So, um, uh, if you're like just, uh, let's say you're part of a big team and, uh, you know, sometimes like measuring work and progress is a big deal. It's like, hey, we were here and now we're here. And so that's one thing that I've kind of gravitated towards. If there was no way to, like, measure this, I probably wouldn't have gotten as nerd— nerded out.

Yeah.

Yeah.

Honestly, the parallel that I take is— and this is again very niche— but F1 driving and racing, you know, if you're not familiar with the way F1 works, it's only, you know, 10 teams, 20 drivers in a given year. And so there's not a lot of variability and the top people and the bottom people tend to be known really far in advance. And so oftentimes you might ask if you're not familiar, like, why bother? Right? And they're not trying— if you're in 20th place to go win the next season, like, that's not a realistic goal, but they are trying to win or want place in one race and maybe move from 20th to 19th to 18th. And they celebrate those wins. And the only way that they can do that is by measuring rigorously and being really, really open and honest with themselves about the current position. And what improvement opportunities look like. So I really respect that point. Any last questions, please do get them in now. But one thing I will add is Scott has been very generous about not pitching SmartScout in the context of this presentation. But of course, his own tool is really good at auditing that product and category level visibility. A lot of the example visuals you saw in his presentation today came straight out of SmartScout. So since he won't say it, I will go check it out. It's a good way to get started with Amazon Visibility if you haven't already started with other tools. Scott, thank you so much for your time on a Friday during Prime Day week and with a new child. So I appreciate you, you know, taking a break from your parental duties. I'm sure your family appreciates it as well. And thank you very much, everybody. Have a great weekend.

Thank you.

Thanks, Scott. Be well.