How to rank #1 on Amazon's AI engine

This This is the Billiondoll Sellers podcast. Your go-to source for cutting edge strategies and success stories from the world of Amazon and e-commerce. Buckle up and get ready to take your Amazon business to new heights. Don't forget to subscribe to the Billiondoll Sellers Newsletter. Welcome your host. >> Welcome your host, Kevin King. >> Hey everyone, and welcome to the Billiondoll Sellers podcast. I'm your host, Kevin King, and today is Thursday, July 2nd, 2026. We've got a loaded show today, so let's get into it. Today, we're going to talk about Alexa for shopping because Andrew Bell just dropped, I think, the best breakdown yet on how to optimize your listings for Amazon's AI layer, and it's a game changer. Then, Amazon just made your entire pricing history public to shoppers, which, you know, changes the game on promotions and prime based strategy. We got a great video from Gloria Chow on how to rank on AI search using earn media, some interesting stats on Amazon's growth projections through 2030, uh, research on why AI generated images might actually be hurting your conversions, a cool inventory management tool, and more, plus the usual hot picks at a parting shot. All right, and here's your stuff Bezos question for today. So, Amazon is expected to grow North American revenue about 31% over the next four years. But the question is, which continent do they expect the highest growth at 70%. Think about that and I'll give you the answer at the end of the show. First quick shout out, Market Masters 4 is happening August 20th through 24th here in Austin, Texas. It's 3 and 1/2 days inside an 18,000t mansion on Lake Travis. Uh where the top 1% of e-commerce sellers tear real businesses apart and put them back together. Most attendees call it their favorite event in all of e-commerce. And I got to say it, it really is something special. And right now, for the July 4th weekend only, you can save $1,000 with the code M A L U M. Only three hot seats left, and there's 30 audience seats available. When they're gone, they're gone. There's a link in the show notes to grab your spot. All right, let's get into it. So, Andrew Bel just put out one of the best breakdowns I've seen on optimizing for Alexa for shopping, which is the AI layer that's now reaching uh roughly 100 million shoppers. and he built this from Amazon patents, science papers, and hands-on optimization of more than 4,000 asens. So, this isn't somebody speculating. This is real data. I'm going to break down the key points for you, but the full piece is linked in the show notes, and uh I'd highly recommend you go read the whole thing. So, the big shift here is from keywords to what he calls missions. A still decides if your product can be found. That hasn't changed, but Alexa is the layer that decides if it gets understood, trusted, and actually selected. So, a shopper says one thing, right? Like, mail wall art from my living room. And Alexa fans that out into a whole family of searches by room, by style, by budget, material, recipient, and use case. Products that show up across multiple of those quarter branches get read as broadly relevant. Products that only rank for one keyword miss the mission entirely. And one stat that really jumped out to me, uh, Belle looked at more than 15,000 product cards and found that if you want to show up in positions 2 through eight, you basically need at least a 4.4 star rating. That's the floor. Uh, so reviews aren't just a signal anymore. They're a gate. If you're below that threshold, you're not even in a conversation. And the median review count for position one products was about 7,700 compared to about 4,000 for positions 2 and beyond. So trust depth clearly plays into slot quality. Now, here's the really interesting part. Position one is a fit decision, not just a numbers game. The position one product isn't mechanically the highest rated or the cheapest or the most reviewed. It's the one Alexa judges as the best overall fits for that specific shopper's mission. Only about 21% of position one products were the cheapest option and about 30% were the most reviewed. So, it's not just one factor, it's the whole picture. All right, so what do you actually do about it? Belle lays out seven moves and I'm going to walk you through all of them. Number one is what he calls noun phrase optimization. So stop stuffing isolated keywords. Instead, stack your highest volume terms into natural phrases. So metal wall art becomes large modern metal wall art for contemporary bedroom. You keep the search science, but Alexa gets human language it can actually reason over. And he argues that 75 characters is plenty for a title now. You don't need to stuff 200 characters in there. Number two is semantic bridging. And this is about connecting your product to meanings the shopper never types. So rooms it fits, occasions it serves like housewarming or anniversary, who it's for, what style it matches. Every legitimate connection is another door into the catalog and another query branch you can answer. Number three is inference optimization. And this one's really cool. You got to map features to outcomes. Don't just say genuine leather. Build the pathway. Premium material leads to luxurious feel, leads to elevates the room, leads to ages with character. Because Alexa isn't matching words anymore. It's inferring which product fits a query like large black metal wall art under $100 for a living room. Your listing has to be inferable into those situations. It works on four levels. Lexical, which is the actual words. Syntactical, the structure they implied. Semantic, the meaning they intended, and contextual, the situation they're in. And the goal is surfacing the right few products clearly explained, ready to act on. Number four, and this is important, keep doing your A9 keyword work. This new Alexa optimization or acco as he calls it doesn't replace SEO. It actually makes it more important because A9 builds the candidate pool that Alexa selects from. A product no query retrieves can never be chosen by Alexa. So you got to do both. And by the way, he confirms there is no A10 algorithm. Never was. It's A9. So if anybody's out there selling you some A10 course uh to heck with that. Number five is query planning optimization. This is a new tactical layer. For every as you got to ask, of all the searches Alexa might generate from my shopper's mission, how many can my product be found by and how many can I truthfully win? He recommends building what he calls a mission map per product. Who's buying for what occasion with what constraints and in what context? You think about the core product, the primary shopper mission, secondary missions, but also who the recipient is, the context, the hard constraints, plus the soft preferences, proof points, and comparison angles. It's basically a cheat sheet for how your product fits into a shopper's life. And the more of those boxes you can fill, the more query branches you're going to show up in. Number six, and this one's huge, fill every attribute field. All of them. Structured attributes are ground truth for Alexa. If you're missing size, material, room, or compatibility fields, you don't get penalized in scoring. You get excluded before scoring even starts. And Bel has this great line. He says, "The fields you leave blank become the silence Alexa hears when the shopper asks the question those fields would have answered." I love that. It just makes it so clear. And make sure your brand register so you control your own product truth. That's critical. And number seven is product page coverage. Uh Alexa indexes nearly everything on your PDP. Your title, all bullets, and one P brands can actually index through bullet 10, your description, native A+ text, and even lifestyle images with and without text. Your page needs to answer the questions buyers actually ask. Durability, dimensions, installation, gifting, cleaning, compatibility, value, all of it. And here's one more thing that's really important. Alexa also researches the open web. So if Tech Radar or Cosmopolitan describes your product better than you do, they win the citation, not you. Get your brand into third party publications and make the off Amazon story match the on Amazon one. So bottom line, the winning as the most keywordrich one. It's the one Amazon can retrieve, verify, compare, personalize, and confidently select for a specific shopper at a specific moment. Go read Andrew's full piece. It's linked in the show notes, and it's the most complete Alexa for shopping playbook out there right now. All right. Now, this next one you really need to pay attention to. Amazon just made your pricing history public. One tap, that's all it takes. Shoppers can now see how your price has moved over the past month, 3 months, or a full year. And you know, it looks like a small feature, but it isn't. Consumers don't just see today's price anymore. They see today's price in context. So, here's an example. There's one item that was bouncing between roughly $19 and $25 for nearly a year. It never once sold at list price. And that pattern is now sitting right there for every shopper to see. And here's the part that matters for you. Price history doesn't just expose your past promotions. It sets future price expectations. So, think about this. How do you run a compelling Prime Day without training shoppers to wait for the next discount? How do you promote hard without teaching people that your real price is always lower? Amazon didn't create that challenge, but they just made it transparent. Every promotion you run is not part of your product's permanent pricing record. Every markdown, every event, all of it visible. The bar for pricing strategy just went up. Credit to Lord Patterson from BTR for flagging this one. All right, we've got a great video this week that you just have to watch. PR expert Gloria Child breaks down how small brands are using earned media to dominate AI search results and she's basically saying that you know the old paytoplay tactics are dead. If you want Chad GBT or Claude or Alexa to recommend your brand to shoppers, you guys show up in gift guides, listicles, and third party press. Those are the trust signals that AI storefronts are looking for. And whether you're a solo Amazon seller or scaling a six-figure Shopify brand, she walks you through how to use free AI tools to craft the perfect pitch and land massive media coverage. The video is linked in the show notes. Definitely worth a watch. And if you're not subscribed to the Marketing Misfits newsletter, is a link for that in the show notes, too. Now, let's look at some interesting stats real quick. So, according to Flywheel Retail Insights, Amazon is on track to become a $1.3 trillion retail machine by 2030. And the market bymarket breakdown is uh it's pretty eye opening. The US goes from $536 billion to $87 billion about a 8.5% compound annual growth rate. Germany goes from $70 billion to $105 billion at 8.4%. UK from 52 to 82 billion at 9.4%. Japan from 39 to 55 billion at 7.1%. But the one that really jumps out is India. 30 billion to54 billion uh which represents a 12.2% compound annual growth rate. That's the fastest growing market of all these major countries. So if you're thinking about international expansion, those numbers are definitely worth paying attention to. And when it comes to Tik Tok shop, there's another stat floating around that really tells you the story of that platform. Out of about 100,000 US sellers on Tik Tok shop, the top 1% which is fewer than 900 sellers accounts for about 60% of all GMV. The top 10% does about 93% of total GMV and they bottom 50% that's 50,000 sellers does 2%. I mean, holy cow. So, Tik Tok shop very much rewards the top performers and the longtail barrier registers. Keep that in mind if you're thinking about jumping in there. Now, today's software tool of the day is an automated inventory management system for Amazon FBA sellers called Inventory Hero. So, you connect your seller central account with one click and it syncs your full order history, then runs real forecasting math, demand baselines, seasonality profiles, safety stock, reorder points, uh, the works. But the killer feature is the AI angle. You can plug its MCP server into Claude and your AI can read live inventory, flag reorders before they're urgent, draft purchase orders with deposits mapped to supplier terms, and even write the expedited email to your factory. And everything gets saved to a shared AI employee handbook, so your VA's clawed and yours give the same answers. The numbers come from a deterministic forecasting engine, not AI guesses, and nothing gets ordered without your approval. It also prices out your stockout losses per skew, which is the revenue leak that never shows up on any Amazon report. It's built by Andrew Ericson, who's a sevenf figureure private label seller that's coached more than 1,000 sellers. The web app works standalone without AI, too. It's covering inventory tracking across FBA, AWD, and 3PL, but also PO planning and gant or comban, IPI, health, aging, and perscute profitability. Free 30-day trial, no credit card. There's a link in the show notes. All right. Now, this next one is, I think, really important for anyone using AI to create customerf facing content. So, there's research out from the journal of consumer research, uh, published in May 2026, and it looks at what happens when people know content was made with AI. And across eight experiments and an analysis of about a million Tik Tok posts, they found that AI labeled posts got 7 to 8% viewer likes and 7% lower combined engagement at even when the quality was identical. creators were seen as putting about 15% less effort in and people felt about 14 and a.5% less connected to them. And the reason is you know we value things more when we believe someone put time and care into them. That effort is what makes us feel connected to the creator and connection drives engagement. AI signals the opposite. We assume the shortcut got taken so we tune out. But here's the thing. The penalty disappears when that AI looks hard to use. Like Photoshop mural filters versus one-click generation. So people run a double standard on AI. It's fine when we use it, but suspect when others do. So the recommendation is be careful using AI on customerf facing content. And if you do use it, show the human work behind it. Tell people what you actually did. You know, something like we shot the product ourselves and used AI to clean up the background. Buyers engage more when they believe real effort went in and lead with usefulness. Content that solves a real buyer problem gets shared and that sharing offsets the AI penalty. All right, before we wrap up, a few more hot picks for you. Tik Tok is introducing something called Symphony Agent to create campaigns at scale. Amazon bought Chad GPT ads to promote Prime Day last week, which uh you know, that's interesting if you think about it. Turkey and Bulgaria are the e-commerce growth leaders in Europe right now. And more marketers are turning to AI to manage their e-commerce ads. Links to all those are in the show notes. And here's your parting shot for today. This one's from Tony Robbins. It is in your moments of decision that your destiny is shaped. I really like that one because you know every day as a seller you're making decisions, big ones, small ones. And it's those decisions, not the markets or the algorithm that shape where you end up. Finally, about that Stan Bezos question from the beginning, I asked which continent Amazon expects the highest growth at 70%. The answer is Africa. Amazon expects to grow to $33 billion in Africa by 2030, which is 70% growth. How cool is that? All right, that's all for today, folks. I'll see you again on Monday and have a great Fourth of July holiday weekend if you're in the USA. This is Kevin King signing off from the Billiondollar Sellers podcast.