WEBVTT
1
00:00:00.100 --> 00:00:10.120
[upbeat music] This, this is The Billion Dollar Sellers Podcast, your go to source for cutting edge strategies and success stories from the world of Amazon and e-commerce.
2
00:00:10.700 --> 00:00:22.780
Buckle up and get ready to take your Amazon business to new heights. Don't forget to subscribe to the Billion Dollar Sellers newsletter. Welcome your host Welcome your host, Kevin King. Hey everyone.
3
00:00:22.820 --> 00:00:35.150
Welcome to the Billion Dollar Sellers Podcast. I'm your host, Kevin King, and today is April twenty-seventh, twenty twenty-six. Before I jump in, I got a message from a reader named Benja, and this is what he said.
4
00:00:35.589 --> 00:00:46.200
"Since I joined your newsletter, my business has only grown and I've learned countless things I had no idea about. It's given me a new lease of life in Amazon FBA. Thank you." That's what it's all about, man.
5
00:00:46.380 --> 00:00:54.870
I love hearing that kind of stuff. All right, big show today. We're talking about why you should stop tracking your Amazon rank and, uh, what you should be looking at instead.
6
00:00:55.500 --> 00:01:02.440
Rufus just got a wild new feature where shoppers can basically tell Amazon who they are, and that changes everything about how your listings get surfaced.
7
00:01:03.000 --> 00:01:07.880
A twenty-seven-year history of SEO and where we are now with four brand-new disciplines you need to know about.
8
00:01:08.600 --> 00:01:14.740
A really important warning about how your bestseller badge can actually make you a target for hijackers coming in through Mexico.
9
00:01:14.820 --> 00:01:25.260
And Anthropic just ran an eye-opening experiment with AI agents negotiating deals for people, and the results are something every seller needs to hear. But first, here's your stump Bezos question for today.
10
00:01:26.340 --> 00:01:36.090
ChatGPT introduced sponsored suggestions in the conversational product recommendations about six weeks ago. So the question is, how much annualized revenue has it generated so far?
11
00:01:36.140 --> 00:01:43.830
Think about that, and I'll give you the answer at the end of the show. All right, let's get into the big one. Stop tracking rank. Start tracking market share.
12
00:01:44.440 --> 00:01:54.320
So Christian Umlak from Autopilot put out a piece that I think nails something a lot of us have been feeling, and that is that rank tracking made sense back in twenty sixteen, but in twenty twenty-six it's measuring the wrong thing.
13
00:01:54.900 --> 00:02:03.980
And here's why. There isn't just one rank anymore. Amazon personalizes the search results page by browsing history, purchase history, location, device, time of day.
14
00:02:04.420 --> 00:02:10.070
There are millions of ranks, one per shopper, and a scraper running from a data center isn't seeing what your customer sees.
15
00:02:10.720 --> 00:02:20.600
On top of that, over half of Amazon traffic is mobile now, and most rank trackers are still scraping desktop. So your position eight might not even exist where your customers are actually shopping.
16
00:02:21.360 --> 00:02:26.800
And Amazon is constantly shuffling products around, moving them from position four to twelve to six to test placements.
17
00:02:27.200 --> 00:02:35.960
That noise takes about seventy-two to ninety-six hours to stabilize, but teams and tools are reacting to it daily and tanking their own bids in the process.
18
00:02:35.980 --> 00:02:44.700
And the search results page isn't even where all the business lives anymore. You got browse nodes, subscribe and save, deals, DSP, sponsored display, external traffic, Rufus.
19
00:02:45.180 --> 00:02:56.180
Rank tracking is measuring a shrinking slice of the pie, and rank tells you nothing about outcomes. You can rank number one on a zero volume keyword. A competitor can eat your share while you're sitting at position two.
20
00:02:56.700 --> 00:03:08.440
Rank is a proxy for visibility. Visibility is a proxy for traffic. Traffic is a proxy for sales. That's three layers from what actually matters. So what should you be looking at instead? Search query performance.
21
00:03:09.020 --> 00:03:15.540
If you have brand registry, SQP is sitting right there in your brand analytics tab, and most brands barely touch it.
22
00:03:15.680 --> 00:03:25.120
Every week for every ASIN and on its top one hundred queries, Amazon hands you four numbers as first party data. Impression share, which is what percent of the real estate you occupied.
23
00:03:25.700 --> 00:03:36.600
Click share, meaning what percent of the clicks you won. Cart add share, so what percent of cart adds came from your listings. And finally, purchase share, which is the percentage of purchases that were yours.
24
00:03:37.140 --> 00:03:46.580
No scraping, no proxies, no, uh, seventy-two-hour shuffle artifacts. It goes back two years, and it's available through the SP API. Now, here's where it gets powerful.
25
00:03:47.040 --> 00:03:56.320
You can read these four shares as a funnel, and the diagnosis basically tells itself. If you've got high impression share but low click share, that's a search results tile problem.
26
00:03:56.360 --> 00:04:05.640
Your main image, title, price, badges, review count, something on that tile isn't pulling people in. High click share but low cart add share, that's a product detail page problem.
27
00:04:05.700 --> 00:04:14.480
Your A+ content, bullets, images, variation, clarity, social proof, something's not converting once they land. A high cart add share but a low purchase share means it's a closing problem.
28
00:04:15.060 --> 00:04:25.180
Price, shipping speed, stock, your offer isn't sealing the deal. Now, if you've got low impression share but high purchase share, holy cow, that's a hidden goldmine. When you show up, they buy.
29
00:04:25.580 --> 00:04:37.580
Bid harder on that keyword. And high impression share with near zero purchase share, that's a budget black hole. Cut spend immediately. But here's the insight that, uh, I think really buries rank tracking for good.
30
00:04:38.220 --> 00:04:47.700
SQP doesn't just give you share, it gives you total count, which is the market's absolute volume and brand count, which is your absolute volume. So now you can answer the only question that matters.
31
00:04:47.760 --> 00:04:57.520
Is it us or is it the market? If your purchases are up and your share is up, but the market's flat, you're winning. If purchases are up, but share is flat and the market's up, you're just riding a tailwind.
32
00:04:57.530 --> 00:05:04.240
Don't confuse tide for skill. If purchases are flat, but share is down and the market is up, you're losing share in a growing market.
33
00:05:04.340 --> 00:05:14.780
That's the most dangerous state in commerce, and it's completely invisible to a rank tracker. If everything's down, but your share is up, the category is contracting, but you're gaining position. Hold steady.
34
00:05:15.260 --> 00:05:25.800
And if purchases and share in the market are all down, it's macro or seasonal. Manage your cash. A rank dashboard shows you a flat green line in all five of those scenarios. That's the problem.
35
00:05:26.360 --> 00:05:35.700
And there's one more cut you should be making. Pull your queries apart into branded and unbranded. Winning purchase share on, let's say, your brand water bottle is loyalty harvest.
36
00:05:36.220 --> 00:05:44.480
Winning purchase share on stainless steel water bottle is category conquest. Only one of those builds a durable business, and only SQP separates them cleanly.
37
00:05:45.240 --> 00:05:51.640
Rufus is starting to surface conversational queries that fragment into longer, higher intent phrases.
38
00:05:51.680 --> 00:05:59.600
So instead of stainless steel water bottle, people are typing things like, "An insulated bottle that keeps coffee hot for an eight-hour shift and fits a car cup holder."
39
00:05:59.620 --> 00:06:05.620
The queries fragment, but the underlying markets persist. Whoever measures share of demand stays oriented.
40
00:06:05.680 --> 00:06:16.568
And think about seasonal categories as well.Halloween, for example, barely exists for 11 months and then becomes one of the largest categories on Amazon. Back to school compresses months into three weeks.
41
00:06:17.008 --> 00:06:27.168
A rank tracker cannot tell you your category is 10 times your average this week. SQP's total countable and your brand count tells you if you caught the wave. So here's what I want you to do.
42
00:06:27.888 --> 00:06:39.448
Stop opening the rank dashboard tomorrow morning. Instead, pull SQP, pick your top 20 queries, and run the final diagnostic. Separate branded from unbranded. Track share and market size weekly.
43
00:06:40.038 --> 00:06:48.508
You're doing twenty sixteen's work with twenty twenty-six stakes. The data's already sitting in your account. All right. Now let's talk about an interesting stat.
44
00:06:49.028 --> 00:06:58.988
Marketplace Pulse put out their twenty twenty-six seller index, and they asked sellers which single operational area AI or automation has delivered the most measurable positive impact.
45
00:06:59.028 --> 00:07:06.108
The top answer, and this one's kind of surprising, was listing optimization. About a quarter of sellers said that's where AI has moved the needle the most.
46
00:07:06.628 --> 00:07:15.988
Image and video creation came in second, and advertising management was third. But here's the kicker. The number one response overall, especially among distressed sellers, was no area.
47
00:07:16.148 --> 00:07:26.267
Almost fourteen percent of distressed sellers said AI hasn't delivered measurable impact anywhere. So there's still a huge gap between people who are using it well and people who, you know, haven't figured it out yet.
48
00:07:27.448 --> 00:07:34.368
All right. Uh, this next one is a big deal. Rufus knows who you are. Amazon just turned Rufus into a profile matching machine.
49
00:07:34.988 --> 00:07:45.048
So there's a new feature called Tell Us About You, where shoppers can type free text descriptions of themselves into Rufus, style, hobbies, kids, pets, apartment size, whatever they wanna share.
50
00:07:45.628 --> 00:07:57.448
And Amazon saves it as a persistent identity profile tied to their account. And that profile follows the shopper everywhere. Every search, every Alexa request, every Rufus chat, every device.
51
00:07:57.828 --> 00:08:07.828
The query stays the same, but the identity layer underneath it changes the results. So picture this. Two shoppers type storage bins. One has told Rufus she has three kids in a small apartment.
52
00:08:08.198 --> 00:08:14.768
The other said he runs a minimalist home office. They get completely different ASINs, same keyword, different stacks. That's wild.
53
00:08:15.348 --> 00:08:25.328
Vanessa Hung surfaced the rollout on April twenty-second after Ritu Java spotted it first. So why does this matter for sellers? The old game was keywords, reviews, and conversion rates.
54
00:08:25.588 --> 00:08:31.788
Those signals still count, but there's now a new layer sitting between the search bar and your ASIN, and that's who Rufus thinks the shopper is.
55
00:08:32.588 --> 00:08:42.168
Your bullets, your A+ content, your back-end keywords, they now do two jobs at once. They feed the keyword index like always, but they also get read against the saved shopper profile.
56
00:08:42.888 --> 00:08:53.848
And listings that name the person are the ones who win. I'm talking phrases like designed for families with young kids, built for small home offices, perfect for pet owners and apartments. Those are anchors.
57
00:08:54.268 --> 00:09:04.108
Rufus maps them to real profiles on file. Generic spec sheets that are just dimensions, materials, feature lists with no human in them, those are completely invisible to the profile matching layer.
58
00:09:04.788 --> 00:09:12.228
Doesn't matter how well you rank on the keyword. Three hundred million people used Rufus in twenty twenty-five. Twelve billion dollars in incremental sales.
59
00:09:12.408 --> 00:09:21.708
Monthly users up one hundred and forty-nine percent year over year. Interactions up two hundred and ten percent. And shoppers who use Rufus convert sixty percent higher than shoppers who don't.
60
00:09:21.868 --> 00:09:32.708
This is not a side feature. This is the front door. So here's your action step. Audit every listing and ask one question. Does this copy describe a product, or does it describe a product for a specific person?
61
00:09:33.368 --> 00:09:41.168
If the answer is just a product, rewrite it. Build the buyer persona first, then write the bullets, then the A+, then the back end.
62
00:09:41.688 --> 00:09:51.288
Identity language is the new keyword stuffing, except it actually has to be true and specific. And on the bigger picture here, Amazon's ad layer is getting more AI mediated by the month.
63
00:09:51.968 --> 00:10:00.048
AI shopping prompts went paid on March twenty-fifth. You can win the auction and still lose the match if your listing doesn't speak to the shopper Rufus has on file.
64
00:10:00.648 --> 00:10:11.188
Paid placement gets you in the room, and identity language gets you the sale. Rufus is not gonna forget what shoppers told it. Sellers who write to a person beat sellers who write to a search engine.
65
00:10:12.248 --> 00:10:13.488
All right, this is a cool one.
66
00:10:14.368 --> 00:10:23.048
Somebody put together a twenty-seven-year history of SEO going from two thousand all the way to twenty twenty-six, and the twenty twenty-six slide is really the one you need to pay attention to.
67
00:10:23.828 --> 00:10:32.648
We're now in what they're calling the four-discipline future. AI native search is the default as more than fifty percent of Google searches result in zero clicks to external websites.
68
00:10:33.108 --> 00:10:41.568
The twenty-five-year blue link era is functionally over. And personal AI assistants from Google, Apple, Amazon, and startups are now handling research, comparison, and purchasing.
69
00:10:42.108 --> 00:10:50.188
Agent-to-agent commerce is emerging, where AI systems negotiate on behalf of consumers. So instead of just one type of SEO, there are now four permanent disciplines.
70
00:10:50.688 --> 00:11:01.428
The first one is traditional SEO, search engine optimization. That's down to about twenty-five to thirty percent of discovery now. Still important for brand awareness, but not sufficient on its own.
71
00:11:01.468 --> 00:11:12.428
The second one is GEO, generative engine optimization. That's being cited by AI systems such as Google, Perplexity, emerging AI platforms as authoritative and trustworthy sources.
72
00:11:12.948 --> 00:11:20.028
It's a citation-based traffic and a multi-engine optimization. Number three is AEO, answer engine optimization.
73
00:11:20.608 --> 00:11:31.498
This is when getting selected as the definitive answer by AI engines for direct queries, voice assistants, and featured responses. And the last one is ACO, agentic commerce optimization.
74
00:11:32.108 --> 00:11:40.888
Being recommended by AI agents that research, compare, and purchase products autonomously on behalf of consumers. This is a new battleground, especially for e-commerce.
75
00:11:41.468 --> 00:11:48.247
It's about structured data, product metadata, and agent evaluation architecture. There's a link to the full interactive timeline in the show notes.
76
00:11:48.528 --> 00:11:57.648
It's worth checking out if you wanna see how we got from PageRank to AI agents in twenty-six years. All right. Now let me tell you about today's featured event.
77
00:11:57.688 --> 00:12:10.672
If you source from China and you're going to Canton Fair or you've been thinking about it, uh, BBSS Dream one hundred member Kian Golzari is hosting a private master class called Canton Fair Mastery on May second and third in Guangzhou.This isn't a giant seminar.
78
00:12:10.752 --> 00:12:18.612
It's a private room at the Four Seasons with Kian and five elite operators, and they'll be breaking down the playbook that serious seven and eight-figure brands use.
79
00:12:19.232 --> 00:12:30.192
We're talking supplier leverage, price and MOQ strategy, product development, and long-term partnerships. There are only fifty seats available, so if you're interested, check out cantonfairmastery.com.
80
00:12:30.252 --> 00:12:42.532
There's a link in the show notes. All right, this next one is an important warning, and it comes from BDSS Dream one hundred member Vanessa Hung. Your bestseller badge just made you a target. So here's what happened.
81
00:12:43.132 --> 00:12:49.052
A seller hit number one, and then their US listing vanished the same day. And the attack didn't come through the US. It came through Mexico.
82
00:12:49.712 --> 00:13:01.002
What the hijackers did is, is that they uploaded a flat file in the Mexico marketplace flagging the ASIN as an adult product. Amazon's catalog applied it upstream and suppressed the US listing. Gone, just like that.
83
00:13:01.572 --> 00:13:11.452
And here's why you're probably exposed even if you didn't know it. Back in twenty twenty-two, Amazon auto-enrolled every active US FBA seller in NARF, which is the North America Remote Fulfillment.
84
00:13:11.952 --> 00:13:23.172
When that happened, your ASINs got duplicated in Canada and Mexico automatically. Those duplicate listings accept external flat file contributions, and that's the open door. So here's what you need to do.
85
00:13:23.212 --> 00:13:34.052
Don't audit your US back end first because the problem isn't there. Check your NARF status instead right now. Go to Seller Central, then Settings, Account Info, and Remote Fulfillment with FBA.
86
00:13:34.712 --> 00:13:38.452
If you see toggles for Canada, Mexico, and Brazil, it means you're enrolled.
87
00:13:38.592 --> 00:13:48.632
Turn off any marketplace you're not actively managing and treat any bestseller badge or ranking spike as a high-risk window because that's when the catalog becomes a target. Thirty seconds to check.
88
00:13:49.022 --> 00:13:57.852
Worth doing before the next ranking win, not after. Now everybody say, "Thanks, Vanessa." Be sure to check out her website. There's a link in the show notes. Quick break.
89
00:13:58.152 --> 00:14:01.392
Let's talk about reaching page one on Amazon using micro-influencers.
90
00:14:01.972 --> 00:14:10.392
So there's a platform called Stack Influence, and what they do is they automate micro-influencer product seeding at scale, thousands of collabs per month. And you're not paying fees.
91
00:14:10.412 --> 00:14:12.912
You're just sending them free products. That's it.
92
00:14:12.992 --> 00:14:22.672
Top Amazon brands like Magic Spoon, Unilever, and Mary Ruth Organics have used it to get to page one positioning and increase their monthly revenue as high as thirteen times in as little as two months.
93
00:14:23.332 --> 00:14:33.432
You get external traffic sales, full rights image and video UGC, plus it's a hundred percent automated management. You don't have to lift a finger. BDSS readers get ten percent off if you sign up this month.
94
00:14:33.572 --> 00:14:41.992
Link's in the show notes. All right. This next story is, uh, one of those things where you read it and you're like, "Okay, the future is arriving faster than I thought."
95
00:14:42.532 --> 00:14:53.992
Anthropic, the company behind Claude, just ran a one-week experiment called Project Deal. They turned their San Francisco office into a Craigslist-style marketplace, but with a twist. AI agents did all the negotiating.
96
00:14:54.652 --> 00:15:02.872
Sixty-nine employees signed up. Each got a hundred dollar budget. Claude interviewed every participant about what they wanted to sell, what they wanted to buy, and how they wanted their agent to negotiate.
97
00:15:03.472 --> 00:15:10.732
Those answers became custom system prompts, and then the agents went to work posting listings, haggling with each other, and closing deals on behalf of their humans.
98
00:15:11.432 --> 00:15:17.912
At the end of the experiment, a total of a hundred and eighty-six deals were closed, just over four thousand dollars in total transaction value.
99
00:15:18.412 --> 00:15:29.152
Goods exchanged ranged from a snowboard to a plastic bag of ping pong balls. And participants liked it so much they said they'd pay for the service. But here's the part sellers should really pay attention to.
100
00:15:29.162 --> 00:15:39.472
Anthropic ran a secret parallel test. They quietly gave some participants their top model, Claude Opus 4.5, and others their smallest model, Claude Haiku 4.5. The result?
101
00:15:39.592 --> 00:15:48.672
People with the smarter model got measurably better outcomes. People with the weaker model got worse deals and didn't even know. This is the agent-to-agent commerce future everyone's been theorizing about.
102
00:15:49.122 --> 00:15:56.152
The quality of the AI representing the buyer or the seller determines who wins. Cheaper model, worse outcome, and the human on the losing end has no idea.
103
00:15:56.972 --> 00:16:07.572
If shoppers start using agents to buy on Amazon, the seller who's listing and pricing an offer optimized for agent evaluation wins. The seller still optimizing for human eyeballs loses quietly.
104
00:16:07.592 --> 00:16:15.172
Project Deal was a pilot with a self-selected pool of AI nerds, but the signal is loud. Agent-to-agent commerce is closer than it looks, and the model gap is gonna matter.
105
00:16:16.152 --> 00:16:25.872
And if you missed Ecom Mastery AI and BDSS at Nashville, the vault is still open. More than forty sessions, eighty-one thousand dollars in bonuses, AI-powered search across all the replays.
106
00:16:26.412 --> 00:16:34.972
But the price goes up tomorrow, so if you've been on the fence, grab it now. Link's in the show notes. Oh, and you gotta see this video also in the show notes.
107
00:16:35.092 --> 00:16:44.732
A guy ordered a pack of, uh, chopsticks from Amazon and then received thousands of shipments. It's an old Chinese ranking hack that apparently is still going. Um, pretty wild to watch.
108
00:16:45.792 --> 00:16:56.372
Okay, now before we wrap up, here's a few more hot picks for you. Most Amazon sellers are using AI the wrong way. A look inside the AI systems Amazon uses to protect shoppers.
109
00:16:57.172 --> 00:17:09.412
Amazon is wielding its power over AI shopping bots. And sixty percent of Amazon search now involves Rufus engagement. Links to all of those are in the show notes. And here's your parting shot for today.
110
00:17:09.422 --> 00:17:12.652
"No person will make a great business who wants to do it all himself or get all the credit."
111
00:17:13.072 --> 00:17:24.192
That's from Andrew Carnegie, and I think that applies to everything we talked about today, whether it's using SQP data instead of guts, letting AI agents handle negotiation, or learning from people like Vanessa and Tian.
112
00:17:24.832 --> 00:17:36.312
Don't try to do it all yourself. Oh, and remember that Stump Bezos question from the beginning? ChatGPT introduced sponsored suggestions into conversational product recommendations about six weeks ago.
113
00:17:36.932 --> 00:17:48.292
So how much annualized revenue has it generated so far? The answer is a hundred million dollars in annualized revenue from conversational product recommendation in just six weeks. That's a signal you can't ignore.
114
00:17:49.812 --> 00:18:04.432
All right. That's all for today, folks. I'll see you again on Thursday. This is Kevin King signing off from the Billion Dollar Sellers podcast. [upbeat music]
1
00:00:00.100 --> 00:00:10.120
[upbeat music] This, this is The Billion Dollar Sellers Podcast, your go to source for cutting edge strategies and success stories from the world of Amazon and e-commerce.
2
00:00:10.700 --> 00:00:22.780
Buckle up and get ready to take your Amazon business to new heights. Don't forget to subscribe to the Billion Dollar Sellers newsletter. Welcome your host Welcome your host, Kevin King. Hey everyone.
3
00:00:22.820 --> 00:00:35.150
Welcome to the Billion Dollar Sellers Podcast. I'm your host, Kevin King, and today is April twenty-seventh, twenty twenty-six. Before I jump in, I got a message from a reader named Benja, and this is what he said.
4
00:00:35.589 --> 00:00:46.200
"Since I joined your newsletter, my business has only grown and I've learned countless things I had no idea about. It's given me a new lease of life in Amazon FBA. Thank you." That's what it's all about, man.
5
00:00:46.380 --> 00:00:54.870
I love hearing that kind of stuff. All right, big show today. We're talking about why you should stop tracking your Amazon rank and, uh, what you should be looking at instead.
6
00:00:55.500 --> 00:01:02.440
Rufus just got a wild new feature where shoppers can basically tell Amazon who they are, and that changes everything about how your listings get surfaced.
7
00:01:03.000 --> 00:01:07.880
A twenty-seven-year history of SEO and where we are now with four brand-new disciplines you need to know about.
8
00:01:08.600 --> 00:01:14.740
A really important warning about how your bestseller badge can actually make you a target for hijackers coming in through Mexico.
9
00:01:14.820 --> 00:01:25.260
And Anthropic just ran an eye-opening experiment with AI agents negotiating deals for people, and the results are something every seller needs to hear. But first, here's your stump Bezos question for today.
10
00:01:26.340 --> 00:01:36.090
ChatGPT introduced sponsored suggestions in the conversational product recommendations about six weeks ago. So the question is, how much annualized revenue has it generated so far?
11
00:01:36.140 --> 00:01:43.830
Think about that, and I'll give you the answer at the end of the show. All right, let's get into the big one. Stop tracking rank. Start tracking market share.
12
00:01:44.440 --> 00:01:54.320
So Christian Umlak from Autopilot put out a piece that I think nails something a lot of us have been feeling, and that is that rank tracking made sense back in twenty sixteen, but in twenty twenty-six it's measuring the wrong thing.
13
00:01:54.900 --> 00:02:03.980
And here's why. There isn't just one rank anymore. Amazon personalizes the search results page by browsing history, purchase history, location, device, time of day.
14
00:02:04.420 --> 00:02:10.070
There are millions of ranks, one per shopper, and a scraper running from a data center isn't seeing what your customer sees.
15
00:02:10.720 --> 00:02:20.600
On top of that, over half of Amazon traffic is mobile now, and most rank trackers are still scraping desktop. So your position eight might not even exist where your customers are actually shopping.
16
00:02:21.360 --> 00:02:26.800
And Amazon is constantly shuffling products around, moving them from position four to twelve to six to test placements.
17
00:02:27.200 --> 00:02:35.960
That noise takes about seventy-two to ninety-six hours to stabilize, but teams and tools are reacting to it daily and tanking their own bids in the process.
18
00:02:35.980 --> 00:02:44.700
And the search results page isn't even where all the business lives anymore. You got browse nodes, subscribe and save, deals, DSP, sponsored display, external traffic, Rufus.
19
00:02:45.180 --> 00:02:56.180
Rank tracking is measuring a shrinking slice of the pie, and rank tells you nothing about outcomes. You can rank number one on a zero volume keyword. A competitor can eat your share while you're sitting at position two.
20
00:02:56.700 --> 00:03:08.440
Rank is a proxy for visibility. Visibility is a proxy for traffic. Traffic is a proxy for sales. That's three layers from what actually matters. So what should you be looking at instead? Search query performance.
21
00:03:09.020 --> 00:03:15.540
If you have brand registry, SQP is sitting right there in your brand analytics tab, and most brands barely touch it.
22
00:03:15.680 --> 00:03:25.120
Every week for every ASIN and on its top one hundred queries, Amazon hands you four numbers as first party data. Impression share, which is what percent of the real estate you occupied.
23
00:03:25.700 --> 00:03:36.600
Click share, meaning what percent of the clicks you won. Cart add share, so what percent of cart adds came from your listings. And finally, purchase share, which is the percentage of purchases that were yours.
24
00:03:37.140 --> 00:03:46.580
No scraping, no proxies, no, uh, seventy-two-hour shuffle artifacts. It goes back two years, and it's available through the SP API. Now, here's where it gets powerful.
25
00:03:47.040 --> 00:03:56.320
You can read these four shares as a funnel, and the diagnosis basically tells itself. If you've got high impression share but low click share, that's a search results tile problem.
26
00:03:56.360 --> 00:04:05.640
Your main image, title, price, badges, review count, something on that tile isn't pulling people in. High click share but low cart add share, that's a product detail page problem.
27
00:04:05.700 --> 00:04:14.480
Your A+ content, bullets, images, variation, clarity, social proof, something's not converting once they land. A high cart add share but a low purchase share means it's a closing problem.
28
00:04:15.060 --> 00:04:25.180
Price, shipping speed, stock, your offer isn't sealing the deal. Now, if you've got low impression share but high purchase share, holy cow, that's a hidden goldmine. When you show up, they buy.
29
00:04:25.580 --> 00:04:37.580
Bid harder on that keyword. And high impression share with near zero purchase share, that's a budget black hole. Cut spend immediately. But here's the insight that, uh, I think really buries rank tracking for good.
30
00:04:38.220 --> 00:04:47.700
SQP doesn't just give you share, it gives you total count, which is the market's absolute volume and brand count, which is your absolute volume. So now you can answer the only question that matters.
31
00:04:47.760 --> 00:04:57.520
Is it us or is it the market? If your purchases are up and your share is up, but the market's flat, you're winning. If purchases are up, but share is flat and the market's up, you're just riding a tailwind.
32
00:04:57.530 --> 00:05:04.240
Don't confuse tide for skill. If purchases are flat, but share is down and the market is up, you're losing share in a growing market.
33
00:05:04.340 --> 00:05:14.780
That's the most dangerous state in commerce, and it's completely invisible to a rank tracker. If everything's down, but your share is up, the category is contracting, but you're gaining position. Hold steady.
34
00:05:15.260 --> 00:05:25.800
And if purchases and share in the market are all down, it's macro or seasonal. Manage your cash. A rank dashboard shows you a flat green line in all five of those scenarios. That's the problem.
35
00:05:26.360 --> 00:05:35.700
And there's one more cut you should be making. Pull your queries apart into branded and unbranded. Winning purchase share on, let's say, your brand water bottle is loyalty harvest.
36
00:05:36.220 --> 00:05:44.480
Winning purchase share on stainless steel water bottle is category conquest. Only one of those builds a durable business, and only SQP separates them cleanly.
37
00:05:45.240 --> 00:05:51.640
Rufus is starting to surface conversational queries that fragment into longer, higher intent phrases.
38
00:05:51.680 --> 00:05:59.600
So instead of stainless steel water bottle, people are typing things like, "An insulated bottle that keeps coffee hot for an eight-hour shift and fits a car cup holder."
39
00:05:59.620 --> 00:06:05.620
The queries fragment, but the underlying markets persist. Whoever measures share of demand stays oriented.
40
00:06:05.680 --> 00:06:16.568
And think about seasonal categories as well.Halloween, for example, barely exists for 11 months and then becomes one of the largest categories on Amazon. Back to school compresses months into three weeks.
41
00:06:17.008 --> 00:06:27.168
A rank tracker cannot tell you your category is 10 times your average this week. SQP's total countable and your brand count tells you if you caught the wave. So here's what I want you to do.
42
00:06:27.888 --> 00:06:39.448
Stop opening the rank dashboard tomorrow morning. Instead, pull SQP, pick your top 20 queries, and run the final diagnostic. Separate branded from unbranded. Track share and market size weekly.
43
00:06:40.038 --> 00:06:48.508
You're doing twenty sixteen's work with twenty twenty-six stakes. The data's already sitting in your account. All right. Now let's talk about an interesting stat.
44
00:06:49.028 --> 00:06:58.988
Marketplace Pulse put out their twenty twenty-six seller index, and they asked sellers which single operational area AI or automation has delivered the most measurable positive impact.
45
00:06:59.028 --> 00:07:06.108
The top answer, and this one's kind of surprising, was listing optimization. About a quarter of sellers said that's where AI has moved the needle the most.
46
00:07:06.628 --> 00:07:15.988
Image and video creation came in second, and advertising management was third. But here's the kicker. The number one response overall, especially among distressed sellers, was no area.
47
00:07:16.148 --> 00:07:26.267
Almost fourteen percent of distressed sellers said AI hasn't delivered measurable impact anywhere. So there's still a huge gap between people who are using it well and people who, you know, haven't figured it out yet.
48
00:07:27.448 --> 00:07:34.368
All right. Uh, this next one is a big deal. Rufus knows who you are. Amazon just turned Rufus into a profile matching machine.
49
00:07:34.988 --> 00:07:45.048
So there's a new feature called Tell Us About You, where shoppers can type free text descriptions of themselves into Rufus, style, hobbies, kids, pets, apartment size, whatever they wanna share.
50
00:07:45.628 --> 00:07:57.448
And Amazon saves it as a persistent identity profile tied to their account. And that profile follows the shopper everywhere. Every search, every Alexa request, every Rufus chat, every device.
51
00:07:57.828 --> 00:08:07.828
The query stays the same, but the identity layer underneath it changes the results. So picture this. Two shoppers type storage bins. One has told Rufus she has three kids in a small apartment.
52
00:08:08.198 --> 00:08:14.768
The other said he runs a minimalist home office. They get completely different ASINs, same keyword, different stacks. That's wild.
53
00:08:15.348 --> 00:08:25.328
Vanessa Hung surfaced the rollout on April twenty-second after Ritu Java spotted it first. So why does this matter for sellers? The old game was keywords, reviews, and conversion rates.
54
00:08:25.588 --> 00:08:31.788
Those signals still count, but there's now a new layer sitting between the search bar and your ASIN, and that's who Rufus thinks the shopper is.
55
00:08:32.588 --> 00:08:42.168
Your bullets, your A+ content, your back-end keywords, they now do two jobs at once. They feed the keyword index like always, but they also get read against the saved shopper profile.
56
00:08:42.888 --> 00:08:53.848
And listings that name the person are the ones who win. I'm talking phrases like designed for families with young kids, built for small home offices, perfect for pet owners and apartments. Those are anchors.
57
00:08:54.268 --> 00:09:04.108
Rufus maps them to real profiles on file. Generic spec sheets that are just dimensions, materials, feature lists with no human in them, those are completely invisible to the profile matching layer.
58
00:09:04.788 --> 00:09:12.228
Doesn't matter how well you rank on the keyword. Three hundred million people used Rufus in twenty twenty-five. Twelve billion dollars in incremental sales.
59
00:09:12.408 --> 00:09:21.708
Monthly users up one hundred and forty-nine percent year over year. Interactions up two hundred and ten percent. And shoppers who use Rufus convert sixty percent higher than shoppers who don't.
60
00:09:21.868 --> 00:09:32.708
This is not a side feature. This is the front door. So here's your action step. Audit every listing and ask one question. Does this copy describe a product, or does it describe a product for a specific person?
61
00:09:33.368 --> 00:09:41.168
If the answer is just a product, rewrite it. Build the buyer persona first, then write the bullets, then the A+, then the back end.
62
00:09:41.688 --> 00:09:51.288
Identity language is the new keyword stuffing, except it actually has to be true and specific. And on the bigger picture here, Amazon's ad layer is getting more AI mediated by the month.
63
00:09:51.968 --> 00:10:00.048
AI shopping prompts went paid on March twenty-fifth. You can win the auction and still lose the match if your listing doesn't speak to the shopper Rufus has on file.
64
00:10:00.648 --> 00:10:11.188
Paid placement gets you in the room, and identity language gets you the sale. Rufus is not gonna forget what shoppers told it. Sellers who write to a person beat sellers who write to a search engine.
65
00:10:12.248 --> 00:10:13.488
All right, this is a cool one.
66
00:10:14.368 --> 00:10:23.048
Somebody put together a twenty-seven-year history of SEO going from two thousand all the way to twenty twenty-six, and the twenty twenty-six slide is really the one you need to pay attention to.
67
00:10:23.828 --> 00:10:32.648
We're now in what they're calling the four-discipline future. AI native search is the default as more than fifty percent of Google searches result in zero clicks to external websites.
68
00:10:33.108 --> 00:10:41.568
The twenty-five-year blue link era is functionally over. And personal AI assistants from Google, Apple, Amazon, and startups are now handling research, comparison, and purchasing.
69
00:10:42.108 --> 00:10:50.188
Agent-to-agent commerce is emerging, where AI systems negotiate on behalf of consumers. So instead of just one type of SEO, there are now four permanent disciplines.
70
00:10:50.688 --> 00:11:01.428
The first one is traditional SEO, search engine optimization. That's down to about twenty-five to thirty percent of discovery now. Still important for brand awareness, but not sufficient on its own.
71
00:11:01.468 --> 00:11:12.428
The second one is GEO, generative engine optimization. That's being cited by AI systems such as Google, Perplexity, emerging AI platforms as authoritative and trustworthy sources.
72
00:11:12.948 --> 00:11:20.028
It's a citation-based traffic and a multi-engine optimization. Number three is AEO, answer engine optimization.
73
00:11:20.608 --> 00:11:31.498
This is when getting selected as the definitive answer by AI engines for direct queries, voice assistants, and featured responses. And the last one is ACO, agentic commerce optimization.
74
00:11:32.108 --> 00:11:40.888
Being recommended by AI agents that research, compare, and purchase products autonomously on behalf of consumers. This is a new battleground, especially for e-commerce.
75
00:11:41.468 --> 00:11:48.247
It's about structured data, product metadata, and agent evaluation architecture. There's a link to the full interactive timeline in the show notes.
76
00:11:48.528 --> 00:11:57.648
It's worth checking out if you wanna see how we got from PageRank to AI agents in twenty-six years. All right. Now let me tell you about today's featured event.
77
00:11:57.688 --> 00:12:10.672
If you source from China and you're going to Canton Fair or you've been thinking about it, uh, BBSS Dream one hundred member Kian Golzari is hosting a private master class called Canton Fair Mastery on May second and third in Guangzhou.This isn't a giant seminar.
78
00:12:10.752 --> 00:12:18.612
It's a private room at the Four Seasons with Kian and five elite operators, and they'll be breaking down the playbook that serious seven and eight-figure brands use.
79
00:12:19.232 --> 00:12:30.192
We're talking supplier leverage, price and MOQ strategy, product development, and long-term partnerships. There are only fifty seats available, so if you're interested, check out cantonfairmastery.com.
80
00:12:30.252 --> 00:12:42.532
There's a link in the show notes. All right, this next one is an important warning, and it comes from BDSS Dream one hundred member Vanessa Hung. Your bestseller badge just made you a target. So here's what happened.
81
00:12:43.132 --> 00:12:49.052
A seller hit number one, and then their US listing vanished the same day. And the attack didn't come through the US. It came through Mexico.
82
00:12:49.712 --> 00:13:01.002
What the hijackers did is, is that they uploaded a flat file in the Mexico marketplace flagging the ASIN as an adult product. Amazon's catalog applied it upstream and suppressed the US listing. Gone, just like that.
83
00:13:01.572 --> 00:13:11.452
And here's why you're probably exposed even if you didn't know it. Back in twenty twenty-two, Amazon auto-enrolled every active US FBA seller in NARF, which is the North America Remote Fulfillment.
84
00:13:11.952 --> 00:13:23.172
When that happened, your ASINs got duplicated in Canada and Mexico automatically. Those duplicate listings accept external flat file contributions, and that's the open door. So here's what you need to do.
85
00:13:23.212 --> 00:13:34.052
Don't audit your US back end first because the problem isn't there. Check your NARF status instead right now. Go to Seller Central, then Settings, Account Info, and Remote Fulfillment with FBA.
86
00:13:34.712 --> 00:13:38.452
If you see toggles for Canada, Mexico, and Brazil, it means you're enrolled.
87
00:13:38.592 --> 00:13:48.632
Turn off any marketplace you're not actively managing and treat any bestseller badge or ranking spike as a high-risk window because that's when the catalog becomes a target. Thirty seconds to check.
88
00:13:49.022 --> 00:13:57.852
Worth doing before the next ranking win, not after. Now everybody say, "Thanks, Vanessa." Be sure to check out her website. There's a link in the show notes. Quick break.
89
00:13:58.152 --> 00:14:01.392
Let's talk about reaching page one on Amazon using micro-influencers.
90
00:14:01.972 --> 00:14:10.392
So there's a platform called Stack Influence, and what they do is they automate micro-influencer product seeding at scale, thousands of collabs per month. And you're not paying fees.
91
00:14:10.412 --> 00:14:12.912
You're just sending them free products. That's it.
92
00:14:12.992 --> 00:14:22.672
Top Amazon brands like Magic Spoon, Unilever, and Mary Ruth Organics have used it to get to page one positioning and increase their monthly revenue as high as thirteen times in as little as two months.
93
00:14:23.332 --> 00:14:33.432
You get external traffic sales, full rights image and video UGC, plus it's a hundred percent automated management. You don't have to lift a finger. BDSS readers get ten percent off if you sign up this month.
94
00:14:33.572 --> 00:14:41.992
Link's in the show notes. All right. This next story is, uh, one of those things where you read it and you're like, "Okay, the future is arriving faster than I thought."
95
00:14:42.532 --> 00:14:53.992
Anthropic, the company behind Claude, just ran a one-week experiment called Project Deal. They turned their San Francisco office into a Craigslist-style marketplace, but with a twist. AI agents did all the negotiating.
96
00:14:54.652 --> 00:15:02.872
Sixty-nine employees signed up. Each got a hundred dollar budget. Claude interviewed every participant about what they wanted to sell, what they wanted to buy, and how they wanted their agent to negotiate.
97
00:15:03.472 --> 00:15:10.732
Those answers became custom system prompts, and then the agents went to work posting listings, haggling with each other, and closing deals on behalf of their humans.
98
00:15:11.432 --> 00:15:17.912
At the end of the experiment, a total of a hundred and eighty-six deals were closed, just over four thousand dollars in total transaction value.
99
00:15:18.412 --> 00:15:29.152
Goods exchanged ranged from a snowboard to a plastic bag of ping pong balls. And participants liked it so much they said they'd pay for the service. But here's the part sellers should really pay attention to.
100
00:15:29.162 --> 00:15:39.472
Anthropic ran a secret parallel test. They quietly gave some participants their top model, Claude Opus 4.5, and others their smallest model, Claude Haiku 4.5. The result?
101
00:15:39.592 --> 00:15:48.672
People with the smarter model got measurably better outcomes. People with the weaker model got worse deals and didn't even know. This is the agent-to-agent commerce future everyone's been theorizing about.
102
00:15:49.122 --> 00:15:56.152
The quality of the AI representing the buyer or the seller determines who wins. Cheaper model, worse outcome, and the human on the losing end has no idea.
103
00:15:56.972 --> 00:16:07.572
If shoppers start using agents to buy on Amazon, the seller who's listing and pricing an offer optimized for agent evaluation wins. The seller still optimizing for human eyeballs loses quietly.
104
00:16:07.592 --> 00:16:15.172
Project Deal was a pilot with a self-selected pool of AI nerds, but the signal is loud. Agent-to-agent commerce is closer than it looks, and the model gap is gonna matter.
105
00:16:16.152 --> 00:16:25.872
And if you missed Ecom Mastery AI and BDSS at Nashville, the vault is still open. More than forty sessions, eighty-one thousand dollars in bonuses, AI-powered search across all the replays.
106
00:16:26.412 --> 00:16:34.972
But the price goes up tomorrow, so if you've been on the fence, grab it now. Link's in the show notes. Oh, and you gotta see this video also in the show notes.
107
00:16:35.092 --> 00:16:44.732
A guy ordered a pack of, uh, chopsticks from Amazon and then received thousands of shipments. It's an old Chinese ranking hack that apparently is still going. Um, pretty wild to watch.
108
00:16:45.792 --> 00:16:56.372
Okay, now before we wrap up, here's a few more hot picks for you. Most Amazon sellers are using AI the wrong way. A look inside the AI systems Amazon uses to protect shoppers.
109
00:16:57.172 --> 00:17:09.412
Amazon is wielding its power over AI shopping bots. And sixty percent of Amazon search now involves Rufus engagement. Links to all of those are in the show notes. And here's your parting shot for today.
110
00:17:09.422 --> 00:17:12.652
"No person will make a great business who wants to do it all himself or get all the credit."
111
00:17:13.072 --> 00:17:24.192
That's from Andrew Carnegie, and I think that applies to everything we talked about today, whether it's using SQP data instead of guts, letting AI agents handle negotiation, or learning from people like Vanessa and Tian.
112
00:17:24.832 --> 00:17:36.312
Don't try to do it all yourself. Oh, and remember that Stump Bezos question from the beginning? ChatGPT introduced sponsored suggestions into conversational product recommendations about six weeks ago.
113
00:17:36.932 --> 00:17:48.292
So how much annualized revenue has it generated so far? The answer is a hundred million dollars in annualized revenue from conversational product recommendation in just six weeks. That's a signal you can't ignore.
114
00:17:49.812 --> 00:18:04.432
All right. That's all for today, folks. I'll see you again on Thursday. This is Kevin King signing off from the Billion Dollar Sellers podcast. [upbeat music]
