#115 - Seller Labs MCP Pulls Amazon Data So You Stop Living In Spreadsheets

Seller Central can make you feel like the answer is always three exports away. I sit down with Tempest from Seller Labs to talk about a smarter path: using AI integrations and large language models to turn messy Amazon seller data into clear, usable decisions without living in spreadsheets.

We get into why old-school dashboards start to break down when you need real-time visibility across thousands of SKUs, shifting fees, and fast-moving performance signals. Tempest explains how Seller Labs is working on an MCP that connects directly to Seller Labs and Amazon data, then plugs into Claude so you can ask plain-English questions like “What changed?” and “Where is profit leaking?” and get a direct, data-backed answer. One story hit hard: a seller used a few prompts to uncover a better shipping solution and reportedly saved around $200,000 per year while staying competitive on price.

We also talk Amazon PPC strategy and how ACOS and TACoS can quietly drift while you’re busy running the business. Instead of manual checks, MCP-style automation can monitor campaigns and send alerts when ad performance starts dragging total profitability down. Then we widen the lens to customer experience: reviews, feedback loops, and the hidden margin damage from returns and return fees that can make a “winning” product look far better than it really is.

If you’re searching for Amazon AI tools, profitability analytics, advertising optimization, and a realistic view of where humans still add value, this conversation will give you practical angles to test immediately. Subscribe for more, share this with a seller friend, and leave a review so more brands can find it. What would you want AI to monitor in your account first?

Learn more about Velocity Sellers!

Check us out on YouTube!

Book a FREE Amazon Audit! 

Transcript

Welcome And Guest Background

SPEAKER_00 0:10

Hey everybody, welcome back to the Velocity Sellers Podcast. I'm your host, Velocity Pete. Today we are joined by Tempest from Seller Labs to talk to us a little bit about the business and AI integrations. It's a hot topic right now. Everyone is always wondering about that, what the newest AI feature is and how you can use it better to help your business be a little bit more profitable. So, Tempest, thank you so much for joining me.

SPEAKER_01 0:34

Thanks for having me.

SPEAKER_00 0:35

Yeah. I'd like for you to start a quick elevator pitch about yourself and maybe a little bit about Stellar Labs as well and kind of give us an idea of who we're speaking with today.

SPEAKER_01 0:44

Absolutely. So for 11 years, uh I managed an Amazon account. Um, it did about seven to 10 million dollars a year. And this was back in the almost, I want to say stone ages of Amazon, but not quite. Um, but I've basically been around for the majority of Amazon, uh, the reports, the fees, the profit, the nightmares and all that stuff. And then since then, uh I've come on to Seller Labs. And the thing about Seller Labs that I love so much, especially as an Amazon seller previously, is that um it's just constantly uh making new tools for sellers really to be able to understand and be able to clearly see what's really going on behind the black box essentially that is Amazon.

SPEAKER_00 1:32

Yeah. And it is definitely a beast. Um, that is absolutely for certain. So I'm excited to dive into some ways that maybe uh sellers can understand it a little bit better using AI and again using these different integrations. So

Why Dashboards Are Becoming Obsolete

SPEAKER_00 1:45

um basically to start off, you know, with that topic in mind, we are definitely seeing a lot of uh shift or uh what's a better way to say it, a lot of transitioning towards these large language models, um, you know, with uploading your raw business data, asking it how to improve and doing all these different things. Uh do you believe that traditional data dashboards will eventually maybe become obsolete? Uh, and do you think that Seller Labs is working on something, or is Seller Labs, I guess should I say, working on something to sort of integrate this even smoother?

SPEAKER_01 2:22

Absolutely. Um, I absolutely believe that the traditional dashboards are becoming obsolete, especially with traditional um dashboards, you end up getting a lot of like older data, right? And what Seller Labs has been working on is a MCP that works directly with the data that you have in Seller Labs and the data that Amazon has. So essentially instead of looking manually through spreadsheets and SKUs and all these things, it's a tool where you can ask it, hey, you know, directly compare this to me, Amazon data that they have versus like what I have. And um what you can really do from there becomes really powerful because essentially what's happening is that people are able to automate what they're doing. So instead of manually doing these things constantly, manually looking through spreadsheets, you know, you can set up so many things that, hey, you know, tell me what's changing here, changing there, and then send me notifications directly to tell me anything I should be paying attention to instead of going into a dashboard and manually digging through all of that as I did for 10 years.

SPEAKER_00 3:36

Right. Yeah, no, it's it's it's exciting to not have to deal with I always call Amazon's um seller central dashboards a little clunky. I think it feels like you're kind of bumping around with like uh trying to find the right thing and just not figuring out your answer that is as easily as you would maybe like to. Um so that's really good to hear that it's improving a little bit. Um and so you guys would describe to me kind of what seller or seller labs is dashboard, or maybe that's not even the right word anymore, but describe to me the solution for this, these issues.

MCP With Claude And Big Savings

SPEAKER_01 4:12

So the solution for the issues right now that we're seeing it's especially becoming popular that we've had for uh about four or five months now is our MCP uh that you can you plug in basically directly into Claude. And with that, you can also pull, you know, through Seller Labs with Seller Labs and Claude, it's pulling the information from you know your reports in Amazon without you having to go in manually to get all these reports and digging through it for you. So, an example of this is that one of our sellers, when you know they first start really testing out our MCP with Claude, is that they found that they were basically giving just a ton of free money away that they had no idea before. And they were able to find very quickly without spreadsheets and this and that, all this stuff, a new shipping solution that worked. And then they were able to compare it to big box retail, you know, Target Best Buy, all that stuff, and still um make a price point for their customers that was still competitive with all of that and save themselves about $200,000 a year. And that was just with a couple of, you know, hey Claude, can you show me this? And then pulled the data, the Amazon data from Seller Labs, brought it all together and showed them that almost immediately.

SPEAKER_00 5:37

Yeah. Wow. Wow. Now I'm excited to to kind of dive into a few other pieces. Um, that's already very, you know, exciting to to hear. Again, like I think the seller central interface is a little clunky. Um, so being able to deal with data in real time and have real answers for it sounds it's very exciting.

SPEAKER_01 5:55

So especially without digging into spreadsheets for hours and hours and hours. Exactly. Um, this was the MCP is something that I wish I would have had for the 10 years that I was running that Amazon account. Uh the account that I ran, we had thousands of SKUs. So looking through that was a full-time job, not just for me, but for like me and another assistant.

SPEAKER_00 6:18

Right.

SPEAKER_01 6:18

And this would have saved because people look at the cost of things, you know, a lot of hidden fees, hidden costs when it comes to Amazon. Because with Amazon, they always just want to sprinkle in a little fee here and there. Right. Um, you're also not looking at your labor time into that.

SPEAKER_00 6:34

Yeah. No, that's how I got started as well was a data scraper slash sorter for uh an under another agency where I was doing that for all the brands was downloading, organizing, sending it back to like the brand manager. I was basically what cloud is doing, except that was my whole job. So uh I'm a I'm an AI Doomist, so I think that we're gonna all lose our jobs to AI, which could be good, could be fine, you know. Maybe we all sit on a beach and drink my ties. Um, but also it's like, well, now I don't have a job. But it's is better, I can admit this part, it is better to let AI do that and the computer do that than for me, for someone, anybody, to sit there and just sort through sheets all day. Um, so that makes it a l a little bit easier to digest.

PPC Alerts That Protect Profit

SPEAKER_00 7:18

Um but but one of the other things kind of on that too is uh a big spreadsheet generator is also like the looking at the PPC campaigns and looking at uh your your ACOS, your tacos, your your metrics that are looking at how well your ads are performing. Um I know on the website I noticed a couple different genius tools, one of them being the ad genius. How do you think the ad genius compares to the MCP? Um, and and what does the future of that look like? So do you think um to give it to make that question a little bit more concrete, how do you think advertising strategies would need to um maybe evolve depending on how but good this information is that you're giving the brands and and how can um maybe immediate customer acquisition costs balance with overall brand profitability, if that makes sense?

SPEAKER_01 8:13

Yeah, so just to to start from the beginning there, um I think where this is starting to balance is um anybody that's had to run ad campaigns will see them like wildly, you know, move up and down, things like that. Or maybe you don't notice it for a few days and it's moving up and down all around. So one of the things that really gets lost in the sauce there when you come to, you know, spreadsheets and then like a database is that, or you look in Amazon, is that sometimes the right hand doesn't really get stalking to the left hand, right?

SPEAKER_00 8:46

Right.

SPEAKER_01 8:47

Right. So the beautiful thing about the MCP is that's another thing where you can set alerts for yourself and it can look at those, you know, say like, hey, look at these campaigns, look at these ASNs. And essentially what it can do again, it can flag you, say, in a Slack message that says, Hey, you know, this ACOS is going here and it's actually bringing down your profitability by like two percent. And it's instantaneously telling you this is going on without you having to dig into it. But that's again the power of the MCP talking directly into Seller Labs, pulling all your data all together instead of somebody manually watching it or, you know, depending on an automated program that's not as smart or as fast.

SPEAKER_00 9:37

Right. Right. And it's very interesting. Again, there's a bunch of different geniuses that um I like. Like the feedback genius is also great for positioning yourself as a brand, you know, getting real-time re uh feedback on your reviews and all that kind of good stuff. And I think that's a very interesting one as well. Um, as far as um to the question of like how do you as or how should a business maybe be engineering a customer experience that that drives positive feedback? Um and I think the feedback genius can can provide a lot of from what I looked at on the website, um, can provide a lot of good, for lack of a better word, feedback um on how your brand is is looking. So yeah, um, it's a very interesting assortment of of geniuses, uh of tools that you guys have built. It's very unique.

Reviews Returns And Hidden Amazon Costs

SPEAKER_01 10:25

Well, we really try to drive like client experience and customer experience. And there's one thing that's always gonna be true about Amazon, um, you know, as and Amazon sellers in general, is that the customer experience is always gonna come first, regardless, right? And a lot of that, you know, is feedback, being able to talk to your customers. And another thing that really, really goes in there that the MCP can also help with is your returns and those hidden costs of, you know, especially if maybe you're an FBA seller, even an FBM seller, those return fees eating into your profit overall. Because if you have something that's, you know, getting constant returns and things like that, it helps the seller either improve on the product to improve the customer experience, or, you know, in some cases, maybe not sell the product at all anymore because really you're seeing that with it looks like you're making a profit on the front, right? And that's the problem with the data being everywhere. Um, and that's something that Seller Labs really helps solve, is that it can tell you directly, hey, you know, this looks profitable, but with your return fees, it's not. Or you think you're making a 20% margin, you're making a 5% margin.

Where Humans Still Beat Automation

SPEAKER_00 11:37

Right. And that's that kind of leads me into a big question that always comes up with anything related to AI and and business integration. And I kind of covered it a little bit earlier, um, talking about how I shouldn't, you know, a person shouldn't be doing this, maybe it should be left to a computer with this level of integration and with this level of um automation is maybe a better word, the level of automation that's happening in seller labs and other places in the AI-driven industry right now, where do you see the human intuition and creativity being directed, or where should it be directed for a modern brand to work or for it to, let's say, stand out? Like where should the balance be of AI integration versus you need a human looking at this? Like, does seller labs have an answer for that? Do you have an answer for that? Is it something that we just kind of are figuring out as we go along? Um, I know it's always, of course, a big question, but where do you think that should be?

SPEAKER_01 12:35

It's definitely a big question, but the where where humans still have a role in that absolutely is that even with any kind of AI you're working with, you still need a human hand to guide it so that the AI, you know, you kind of have to train it and tell it right, this is where I want you to go. Because otherwise you're just getting information and it's just grabbing information and it's not quite sure what's going on, right? Right. And essentially the tool as AI is to help the human decision, but the humans still have to interact with that. And as a brand, I think that with you know, as much automation you can do, you still are making time for human creativity, you know, expanding your brand, being more creative, improving upon your products and just having more time in general to keep expanding, keep creating, have that time to keep making new ideas, along with, you know, train the AI to tell you what it what you want it to do so that you're not spending all your time with your head in spreadsheets. I think most uh people that create brands and entrepreneurs, their strength really is in their ideas and their creativity. And this is really helping them have more time to expand all on all that instead of all the time it takes for managing a team and the spreadsheets, fighting with Amazon, all of those things.

unknown 14:02

Yeah.

SPEAKER_00 14:03

That's a great answer as someone that is again worried about the AI taking over. Um, that makes it feel a little bit, a little bit better. So yeah, no, I hope hopefully it always is an input-output situation and not um AI is gonna rise up and take over your household appliances and stuff like that. Don't watch too many movies.

SPEAKER_01 14:21

I don't think so. I don't think we're quite there yet.

SPEAKER_00 14:24

Well, they have all the smart they got like the smart fridge, the smart toaster now I've seen. So it's gonna like shoot bagels at you, you know, if they take over the toaster. I don't know. But in any case, yes, that's a great answer for related to this type of AI. That's more, I guess, hopefully again, another 20 or so years. Maybe not. We'll see. Um, but in any case, great, great answer. Tempest, I really appreciate you joining me for an episode. This has been wonderful to learn about Stellar Labs. I'll have to have you back on to talk um through the next kind of batch of different things that Seller Labs is gonna release. Again, I like the the sound of it. I like not having to go through spreadsheets ever again. So um we'll have to maybe be in touch as well for for uh for my own brand or for anything else. So but yeah, no. Um it's been a pleasure. Any final thoughts on the integration of AI on Seller Labs in general? Um I will have your link, uh, the link to the homepage as well as your LinkedIn down below. So the people out there watching can go and click. But any final thoughts on on the next steps maybe for Seller Labs?

AI Course Resources And Closing

SPEAKER_01 15:28

Absolutely. For Seller Labs, you know, for our clients, we really want to make sure that, you know, we're all heading into the future together. So we're always also offering resources. We just released a course on AI, on the MCP, on how to get, you know, the best data quality you can get out of these tools. Um and we're always just trying to innovate, listen to our clients. And this really was born from that overall. So we're really excited to work with you. Thank you so much for having me on. Yeah. And I'm excited to do more with you in the future.

SPEAKER_00 16:07

Yeah, that might be the quote. We're gonna take everybody into the future with us. I love it. That might be the quote for the episode. Together. Or together. Yes, much better. A little more condensed. Uh oh, Tempest, thank you so much for joining. Like I said, I will have Tempest's LinkedIn as well as Teller Lab's homepage down below in the description. Make sure to check out both of those. And Tempest, again, thank you for being a guest. Thank you so much. Thank you all for watching. I am your host, Velocity Pete, and we will see you next week.