Amazon now tells you if your product is a winner before launching

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, welcome to the Billiondoll Sellers podcast. I'm your host, Kevin King, and today is Monday, August 3rd, 2026. We've got a loaded one for you today. So, let's jump right in. So, on today's show, I'm going to show you how Amazon will now literally tell you whether your product's a winner or not before you ever launch it. And it's using its own first party data to do it. We'll also break down Amazon's Q2 earnings and how the company basically turned itself into one giant AI factory that you're now selling inside of. We've got what Amazon's new AI image rule actually says versus what everybody thinks it says. why the Tik Tok algorithm just got a whole lot harder to beat and a social listening tool that tells you what people really say about your brand and a whole lot more. Oh, and one quick heads up before we get going. So, Norm and I were running a free webinar this Thursday, August 6, at 2 p.m. Eastern, and we're going to show you exactly how to build an email list of uh buyers who actually want your product. And the crazy part is they never even have to come to your website. And registration opens Tuesday, and it's capped at 500 attendees. So, it's going to be a really, really quick sellout. Keep an eye on your inbox tomorrow for that registration link. All right, first up, today's Stump Bezos question. So, Google's Gemini AI now has about 900 million monthly active users. Here's the question. What percent of those conversations actually have consumer purchase intent behind them? So, somebody looking to buy something, think about it, and I'll give you the answer at the end of the show. All right, let's get into it. And I want to start with the story. this whole episode is named after because honestly this is one of the more useful things Amazon has quietly shipped in a while. So, Amazon just added a new feature to Product Opportunity Explorer which is already uh one of the most underrated tools sitting inside seller central and it's called validate a product idea. And here's how it works. So, you type in a product title, a short description, a few bullets, and your target price and Amazon just tells you straight up whether you'd have an advantage or a disadvantage in that market. And that's before you spend a dime on inventory. So Isaac Gross over at IGPPPC, he ran a test with this. He plugged in sugar-free electrolyte powder packets at $30. And just look at what Amazon handed back to him. So he got a full feature gap analysis showing exactly where his concept was trailing the competition. So things like mineral breath, added vitamins, certifications, and sweetener transparency. And he got the top brands owning that segment with Flav City sitting at like 27% of it. He got customer demographics broken down by income bracket. a full 12 months of seasonal click trends and then specific listing recommendations like the fact that seven out of 10 of the benchmark titles explicitly name at least two electrolyte minerals right there in the title. And it even confirmed for him that his $30 price was well positioned against the category benchmark which was sitting at about $29. And here's the kicker and this is really why you should care. This is first party data. So this is coming straight out of Amazon's own search purchase and review data. It's It's not scraped third party estimates from some tool that's just guessing. This is Amazon telling you what Amazon actually knows. So, you can find it under seller central then growth. Go to product opportunity explorer and then validate a new product idea. And honestly, it's worth running on your next idea, but I'd also run it on the products you're already selling just to see where you stand. All right, so next up, and this one's the big picture story because Amazon's Q2 2026 earnings dropped last week, and there's a lot more going on here than the headlines let on. So, if you just skim the news, you probably saw the Wall Street version. Net sales up 20% to just over $200 billion, operating income up 43% to 27.5 billion, and AWS growing 37%, which is its uh fastest clip in 18 quarters. But buried in all of that is a much bigger story for us as sellers. So Jarro Kuano over at the business engineer, he put it really, really well in his deep dive. He said Amazon has basically reorganized itself into, and I love this phrase, the everything AI factory. So it's one giant interconnected machine where the store, the cloud, the ads, the robots, and the delivery network all run on the same engine. And you, the seller, you're operating inside that factory. So let me show you the proof that it's it's really an AI company. now and then what it actually means for your business. So, first follow the money. Amazon spent $173 billion and that's up 64% on capital expenditures over the trailing 12 months. And their free cash flow actually went negative like a $7.6 billion outflow because of it. And Amazon flatout says that increase primarily reflects investments in artificial intelligence. So, they nearly doubled their long-term debt on purpose just to build faster. And on top of that, AWS's AI business and its custom chip business. So that's Tranium and uh Graviton. They each blew past a 25 billion dollar annual run rate and both are growing triple digits. And Anthropic and OpenAI, the two leading AI labs on the planet, they both made multi-year multi-gawatt commitments to Amazon's chips. And Amazon stake in Anthropic delivered a 53.4 billion pre-tax gain in a single quarter, which pushed net income up to 62.6 6 billion. So when a company redirects that much capital into one thing, that thing becomes the company. And the retail side, it isn't separate from all of this. It's the showcase for it. So here's where it actually hits your business. And I've got six things for you. So first up, AI shopping assistants are going mainstream and they're going fast. So Amazon merged Rufus and Alexa Plus into one thing called Alexa for shopping. And it's an agentic assistant that compares products, tracks price history, and can automatically buy stuff for you through price alerts and auto buy. Active users nearly doubled. Interactions are up 5x year-over-year. And here's the one that really matters. Customers who shop through Alexa spend over 40% more per order. So translation and increasing share of your sales is going to be decided by an AI reading your listing, not a human scrolling past your main image. Which means structured, specific, benefit-rich listing content. It isn't optional anymore. It's how you get recommended by the machine. Now, the second one, auto buy completely changes the repeat purchase game. See, if a customer sets auto buy on your competitor's electrolyte powder, you you don't get a second shot at that buy box. So, winning that first AI assisted purchase now locks in a subscription-like revenue stream, which means price competitiveness, and review velocity matter more than ever because the agent, it's watching both of those. Then, here's a big one. Ads are getting cheaper for the sellers who use Amazon's AI and pricier for the ones who don't. So, advertising grew 26% year-over-year, and it's now a $70 billion plus business, which is bigger than any software company on Earth. And Amazon expanded its ads agent to 11 new countries this year. And advertisers who use it, they're seeing 8% lower cost per impression and 6% lower cost per acquisition. So, when your competitor's AI is optimizing campaigns in minutes, running everything manually is basically a tax you're choosing to pay. Moving on, speed is the new table stakes. Prime members got over 40% more items same day or overnight in the first half of the year. And Amazon Now, which is that 30inut delivery, it added 80 US cities with 80% quarter overquarter sales growth. And grocery and everyday essentials are growing meaningfully faster than the rest of the business. So, if you sell replenishable everyday use products, this is your tailwind. But if your inventory placement can't support those fast delivery promises, you're getting more and more invisible. And then this one's kind of wild. The robots are coming for your fees and maybe in a good way. So the nextG Proteus robot moves 1300B loads and now takes plain language voice commands from the warehouse workers. And Amazon opened up its entire logistics network to outside companies through something called Amazon Supply Chain Services. So PNG, 3M, and American Eagle, they're already in. So, Amazon is turning fulfillment itself into a product powered by that same AI stack. And the last one, the big brands, they just keep flooding in. So, Amazon added more than 700,000 products from names like Rabban, Bobby Brown, and Ted Baker. So, the AI factory attracts these premium brands. And that means more competition and higher customer expectations in just about every category. So, where's all this headed? Well, Kofono's framing really nails it. Amazon is the only company that owns both the wafer and the doorstep. So when a customer askked Alexa to reorder detergent, that request runs on Amazon's models, on Amazon's chips, in Amazon's data centers. It ships through Amazon's trucks and it monetizes through Amazon's ads. So every single step of that loop belongs to Amazon. So the strategic takeaway for you is pretty simple. The customer journey is being rebuilt around AI agents. And every Amazon tool you touch, your listings, your ads, your fulfillment, your pricing, all of it's being rebuilt right along with it. So the sellers who win the next 24 months, they're going to be the ones who optimize for how machines shop, not just how humans browse. The old Amazon was two companies. The new Amazon is one factory. So make sure your products are built for the assembly line. All right, so let's take a quick look at some interesting stats. And this one's all about buy now pay later. So those uh installment payment providers. So the first thing that jumped out at me is just how tight the top of that market is. So a firm, Clara, PayPal pay later, and Afterpay. They're all basically neck andneck. Each one used by somewhere in the low to mid 40% range of buy now pay later shoppers. So there's really no runaway leader up top. It's it's a four-way race. And the second thing, there's a massive cliff right after those four because the very next name on the list drops all the way down to about 18%. So it's four heavyweights and then a long long tale of everybody else fighting over the scraps. And if you're offering installment payments on your own store, that tells you pretty clearly which handful of names your customers already know and trust. Okay, so next up, and this is an important one, Amazon's new AI image rule is not what most sellers think it is. And by the way, this breakdown is based on some analysis from the team at Incrementum Digital. So, credit to them. So, Amazon just started requiring sellers to label AI generated people in their listings. And a lot of the e-commerce world, they read that as a crackdown on AI imagery in general. But if you actually look at what the rule covers, a much more useful picture shows up. It targets photorealistic synthetic humans and nothing else. Most of the imagery that actually helps a shopper buy, it doesn't need a person in it at all. So, here's what the rule actually says. Amazon now wants a metadata tag on any photorealistic AI generated person that shows up in your images, your videos, or your A+ content. And shoppers, they see an indicator when one is present. And this isn't Amazon just being difficult. It follows New York's new synthetic performer disclosure law, and similar rules are already spreading to other states. So, this is really compliance, and it's it's probably just the beginning. So, here's the line they're drawing. If there's no person in the image at all, so it's just your product sitting in a setting, an object, a benefit shown visually, that's totally fine. Nothing happens to it. It's only when you put a synthetic person in there that it gets flagged and labeled. So, pretty much everything that actually moves the needle for a shopper is already sitting on the safe side of that line. People buy what they can picture owning. And that takes a scene, the product in a believable moment that's close to the shoers's own life. Now, a real lifestyle photo shoot is still the gold standard, but a lot of brands, they can only afford like one or two of those. But you can render your product on a gym floor, on a work desk, in a stroller cup holder, on a nightstand, without booking a shoot, and without a single synthetic person on screen. So, no person, no label, nothing to disclose. And the same logic works for benefits a camera can't easily capture, like quiet enough for a nursery or packs flat in a carry-on or fits a standard cup holder. And Amazon's new 125 character item highlights field is built for exactly those kinds of claims. So pair each one with an image that shows it. Now, here's the one move to avoid. So the rule is aimed squarely at generating fake people. And I get it. It's tempting if you can only afford one or two models to just generate a whole diverse cast so every shopper sees someone like themselves. But a synthetic person on your listing, it implies a real customer and manufacturing that is staging social proof that doesn't actually exist. And no metadata tag makes an invented customer real. And shoppers, they can kind of feel the pretense. But there is a clean exception here. So an AI generated person is totally fine when it works as information, not as testimony. So a synthetic hand holding the product for scale, a body showing proportion or fit. A figure demonstrating how something assembles those read as specs, not as customers. And the key is you're presenting them as a demonstration, never as a gallery of happy buyers. And one more wrinkle worth knowing. So, the rule actually exempts images of real people even when AI was used to alter them. So, AI retouching, lighting fixes, background cleanup on real photography, none of that triggers the tag. So, shoot a consenting model for real and then let AI handle the polish. Just don't generate that real person into scenes they never actually shot because that's fabrication. It requires their consent and it can land you right back on the labeled side. So, here's how I'd sum it up. Anything without a person in it generate away because those extra use scenes and benefit shots don't need a label and they give a small catalog a reach it just couldn't get any other way. But the spots where a person actually belongs. So your models, your testimonials, your customer content, that's where the real humans go. And whatever you do, don't invent customers because a fake face pretending to be a real buyer, that's the one and only use that's both labeled and flatout deceptive. So yeah, it sounds like a rule against AI imagery, but really Amazon just drew a line around one single use of it, and that's people. And with Q4 bearing down on us, this slow stretch right now is exactly when you want to figure out what you're going to generate and what you're going to actually shoot. All right, so let's talk about today's software tool of the day. And this one answers a question I think every seller has kind of wondered about, which is what are people actually saying about your brand or your competitor's brand when they're not leaving an Amazon review? So, it's called Meltwater, and it's an enterprisegrade social listening platform. And the scale of it is is kind of wild. It monitors something like 1.2 trillion conversations across more than 300,000 news sources, 15 plus social networks with full access to X, over 25,000 podcasts, plus Reddit, Tik Tok, YouTube, Discord, blogs, and forums. And for sellers, the use cases go way beyond vanity metrics. So, you can spot emerging product trends before they even hit Amazon search data. You can track sentiment on your brand and your competitors. You can find influencers who are already talking about your niche. And you get an early warning when a PR problem starts brewing. And their AI assistant, which is called Mirror, it writes the boolean searches for you and turns all that raw conversation data into actual insights. And here's the really timely one. So Meltwater now monitors what the large language models. So Chat, GPT, Gemini, Perplexity, and Claude, what they're saying about brands. And that matters more every single month because shoppers, they're increasingly starting their product research with an AI instead of with Amazon's search bar. Now, heads up, this is enterprise software with enterprise pricing. So, there's no public price list, but typical contracts run somewhere between 15 and $30,000 a year for small teams with the median around 25,800 according to vendor. So, this one's really for the 7 to9 figure brands, not somebody launching their first product. You can check it out over at meltwater.com and there's a link in the show notes. Okay, so next up, if you're doing anything on Tik Tok, listen up because the algorithm bar just got noticeably higher. So you ever post a Tik Tok video that gets like a few hundred views and then just dies? Well, according to Stuart Badley over at Optimize Your Marketing, that's not bad luck. That's a test you failed. And in 2026, that test has gotten a whole lot harder to pass. So here's the thing. Every video starts with a test. So when you post something new, Tik Tok doesn't show it to the whole world. It shows it to a small sample of your existing followers first as a trial. And how that little group reacts, it decides everything. Do they watch to the end? Do they share it or save it? If the answer is no, the video just stalls out and it it almost never recovers later, no matter how good the idea actually was. And here's the biggest change. So to have a real shot at going viral now, a video needs a completion rate above 70%. Meaning 70% of viewers watch all the way to the end. Now, back in 2024, roughly 50% was enough. So, that's a major major tightening. Videos that would have gone wide two years ago, they now just die in the test phase. So, every second of footage that doesn't need to be there is now a liability. And there's a strong secondary signal, too, which is rewatches. So, if like 15 to 20% of your viewers watch the video a second time, Tik Tok reads that as a mark of real quality. And the way Stuart puts it, Tik Tok isn't asking whether people watched your video once. that's asking whether they watched it to the end and then watched it again. And that's a much higher bar than most content is built for. And then there's the other big shift, which is consistency beats viral moments. So posting three to five times a week now outperforms chasing one big viral hit and then going quiet for weeks. Because Tik Tok's system, it rewards the accounts it can test and learn from regularly. So an account that posts sporadically gives the algorithm less data to work with, and that quietly caps its reach even when a video does perform well. So, look, knowing the rules, that's the easy part. The hard part is actually putting in the work. So, cutting your videos tight enough to clear that 70% bar, showing up several times a week without missing, and fighting off that itch to swing for one giant viral moment. And in Stuart's 18 years doing this, the accounts that actually grow on Tik Tok, they're never the ones sitting on the single biggest hit. They're the ones that show up so consistently, the algorithm just learns to trust them. So, if your stuff keeps dying off after a few hundred views, the real bottleneck, it might have nothing to do with how good your content is. It might just come down to your completion rate. And here's something worth a look from our friends over at Stack Influence. And it's all about reaching page one on Amazon just by sending free product to micro influencers. So, Stack Influence is a platform that automates micro influencer product seating collaborations at scale. So, we're talking thousands of collabs a month. And it helps you bump your Amazon ranking, generate user generated content, and boost your recurring revenue. And the nice part is you pay the influencers with product, so you stop negotiating fees. You get full rights to the image and video content, and it's 100% automated, so you're not lifting a finger to get collabs at scale. And top brands like Magic Spoon, Unilver, and Mary Ruth Organics, they've used it to get to number one page positioning. And Blue Land, the ones from Shark Tank, they ran a campaign that generated a 13x return on investment. So if you sign up this month, you can get 10% off. and I'll put a link in the show notes. All right, so before we wrap up, here are a few more hot picks for you. So, Amazon just collected $600 million in Trump tariff refunds. Whatnot is absolutely crushing at selling products online and just raised $20 billion. Shopify messaging now supports WhatsApp marketing. Amazon product highlights are going to start showing up on desktop on August 10th. And there's a really good read on how millennials are changing product discovery and commerce. And you can find the links to all of those stories in our written newsletter over at billiondollarellers.com. And here's your parting shot for today. And it comes from Charlie Mer. He said, "Assume life will be really tough and then ask if you can handle it. If the answer is yes, you've won." And finally, remember that Stump Bezos question from the top of the show. What percent of Gemini's conversations actually have purchase intent behind them? Well, the answer is 6%. So out of 900 million monthly users, only about 6% of those chats are somebody looking to buy. So, something to keep in mind before you go betting the whole business on AIdriven discovery just yet. All right, so that's all for today, folks. Thank you as always for riding along and I'll see you again on Thursday. This is Kevin King signing off from the Billiondoll Sellers