The Main Image Fix That Made $19K | Matt Kostan

The Main Image Fix That Made $19K | Matt Kostan Speaker 2: Most Amazon sellers think that AI is just for optimizing listings or improving your ads. But that's barely scratching the surface. In this episode, Matt Kostan reveals how Amazon sellers are using AI for what really matters, building better brands, making smarter decisions, and providing creative breakthroughs outside the Amazon norm. If you think you're maximizing AI inside your business, you're wrong. You're leaving a massive amount of value on the table. Let's get into our guest. He's created seven figure brands and launched eight successful Kickstarter projects, all based on the power of shopper feedback. His work has been featured in Forbes, Business Insider, and Entrepreneur. Please welcome, in a second, Matt Kostan. All right, let's go to a sponsor first. Are you spending more on Amazon ads and getting less return? There's a smarter way to grow, and without increasing your budget. Levanta's Affiliate Shift Calculator shows what you could do if you reallocate just a portion of your ad spend into affiliate marketing. If you're an Amazon seller doing over $5 million a year, This is for you. And guess what? It's free. All you have to do is click on the link in the description to go to your custom forecast today. So let's get to it. If you have any comments or questions, throw it in the comment section so you can sit back, relax, grab a cup of coffee, enjoy this show. Uh, let's bring on Matt. Speaker 1: Hey, hey, how's it going? Speaker 2: Haven't seen you in a little while. Well, actually I saw you in Vegas. Speaker 1: Yeah. Yeah. Speaker 2: We had a good time there, but what are you up to? We've got a topic today. I had no idea that we were going to be talking about AI. So what's new with you, sir? Speaker 1: Yeah, fair enough. So I'm obviously working hard on product opinion as well. I think what we're doing in that department is trying to figure out Basically a good way to put our content together. So putting together SOPs and because a lot of people want to do testing, but they also want to kind of be able to just follow, Hey, step one, step two, step three. So the most recent thing we've done is put together like a five-step main image SOP for, for your main image on Amazon. And the results are pretty crazy. Like we have one case study where they spent like 400, $500 US on a series of tests. The result was getting an extra $19,000 per month. And the only change was that main image. So pretty cool stuff that we're working on with that. Speaker 2: I want to talk a little bit about product opinion just for a second. Usually I save this to the end, but I've seen you, even if I'm not at the event, you're out there. You guys are pushing it and it's paying off because there's so many times I'd always hear about different platforms that would do a little bit of the same thing. You're different in a lot of ways and all of a sudden your name is coming up in all sorts of either webinars or podcasts, in posts. So whatever you're doing is paying off and that just goes to show you Sometimes you got to spend money to, you know, to make money, to get those impressions. And I want to shoot that back. There are so many, we've got a bunch of people on right now, but there are so many people that do not spend any money on services like yours. And like you just said, the guy spent what, 400 bucks, 500 bucks? Speaker 1: Under 500 bucks US and the ROI is insane. Speaker 2: And this is where with anything, there's some things you got to focus on and spend good money. You know, there's a lot of things people focus on and spend bad money. But I just wanted to give that shout out. And if you haven't used Product Opinion, or if you're afraid to use it, like, oh, I'm not going to, I'm not going to use, I'm not going to buy insurance on product coming in from wherever. It's the same thing. You need it. You really do need it. Speaker 1: Yeah, there's a mindset of like, hey, I'm going to spend more money on PPC, right? Like, and I get it. I would, I'd do the same. But I'm going to spend, um, but I feel like doing the consumer insights, that research is kind of like your foundation. So instead of spending more money to drive traffic to your listing, why not make sure that your listing is as good as possible first. So you have that solid foundation and then all of a sudden your PPC costs go down and you're actually converting the traffic that you're spending so much money on. And I've always said this too, like everyone knows they should be testing, but it's kind of like the broccoli, right? We're sales paying for PPC is kind of like the chocolate like, oh, yeah, I got to pay on PPC. I get immediate traffic. I get those immediate results where you know what testing can take a little bit of a little bit of time. And I guess what I'm what I'm trying to say with the broccoli part is, yeah, we're trying to make, you know, testing as sexy as easy as possible so that it's even fun. Right. So we put a lot of energy and effort into making it so that it's a no brainer. And it's actually like very just easy as like another kind of way to get your product out there. And the kicker is, you know, you get results. One way or another, you're going to learn something. So it's always worth to be getting those incremental gains on your listing there. Speaker 2: Yeah. And I just noticed Simon's got this comment up here. So Matt did an awesome job with my Kitimoto London bus. So Simon, if you don't know Simon, he's one of our, I think he's our only listener. Unknown Speaker: No, no, we've got more. Speaker 2: But Simon is one of those guys that has an excellent, really great product. And I guess he's doing it right. He's coming over to you and doing that research. So let's get into AI. So that's the rage, right, over the last three years. It's actually been much longer than that. Everybody kind of jumped on the bandwagon a few years ago. I remember the first time I actually used it. My buddy Jeff Sass were at my other buddy's office, Colin Campbell, and we were going to go to a Christmas dinner and he goes, Norm, have you seen this? This is awesome. This thing called ChatGPT or OpenAI. And I went, no, I haven't. He goes, it makes poems. So yeah, that was my first experience making poems on AI. But it does a little bit more than that. And let's get into some of these tools. I'm really interested to hear what you have to say about them. Speaker 1: Yeah, totally. So I figured like, obviously, we talk about AI and eCommerce all the time. And what I've been trying to think about is also like, hey, how can I make my daily life easier, like, and beyond just like productivity apps and that sort of thing? Like, I wanted to, yeah, I basically wanted to just see if I can get some practicality Out of it too. So I figured, yeah, we'll chat about some interesting ways of using AI and some tools that you might not have heard of just yet. One of the kind of random ways I guess I was using AI was, so I'm taking apart, so I really enjoy doing manual labor. I spend like a lot of time in front of the screen and manual labor is kind of like my escape, I guess, from it. So for fun, I'm like tearing out the, uh, the deck in my backyard. It's close to the ground, that sort of thing, tearing it out. And, um, uh, you know, I had gloves on, I rip out one of the, uh, one of the boards or whatever. And then there's the spider and it's like a gnarly looking spider that I haven't kind of seen before, right? It's different. It was actually on my hand too. It was on, on the glove on my hand. I was like, Oh, I shake it off. And I'm like, hey, I'm curious, like what type is that, right? So I pick it up more or less with the hammer and I take a picture, ask ChatGPT, like, hey, what is this? And you know what? It's like, hey, it's a black widow. Yeah. And it's like, don't get bit by it. It's very venomous, blah, blah, blah. It looks like a female based on this. It gives me the whole thing, right? And I'm like, holy shit. We have black widow spiders in our backyard. Note to self, you know, maybe, uh, and they usually come in bunches. Oh, there was a few too. Yeah. Like after is because once I noticed that one, then I started noticing like the five, six or seven more of them. So like, oh, that's great. Yeah. Speaker 2: You know what's funny? Because I hate spiders. So I do a lot of research on spiders. But there's a brown widow that is taking over and eating up and killing all these black widows. And they're more venomous. Speaker 1: Oh, nice. Maybe. Yeah, maybe. I didn't even realize we had them in Canada. It was more of a like a thing. I don't know. I just, I didn't, I didn't even realize it. So I'm like, Hey, it just, you know, my knowledge plus one. Speaker 2: Fun, fun. Now this leads me to something else because you could have a picture on Amazon and it's not what the algorithm picks up. It's not what it sees. So it's a really good idea probably to take a picture And figure out or go and just put it into ChatGPT and say, what do you see? Speaker 1: Yeah, it's a good segue. We have a tool. It's free. You don't have to opt in or anything for it. It's imagecheck.com, C-H-E-K. And you can upload your images on it and it'll actually use Amazon's own recognition one-to-one to tell you what the AI sees. And what's interesting with that is you get some instances where, you know, it, it, I think one example was, um, someone had a spatula, but it was kind of turned too much on an angle. So the AI thought it was a toothbrush or like the back of a toothbrush. And a super easy fix is like, okay, so I'm going to turn my spatula to make it so you can actually see the outline. And then the AI is reading it as a spatula. So I mean, not something earth shattering that you have to do, but as Rufus is picking up on images, it's reading it. It makes sense to make sure that you're putting your best foot forward for those results. So yeah, that's a good segue to that. Speaker 2: And it might, you know, tell you something else that it sees in the picture or it could give you some other keywords that you're not associating with the picture, but might as well. It is an AI world right now. Speaker 1: And the other aspect of which I think is actually going to be even more important is the AI can also read your lifestyle images and the demographics of the people that appear. So the whole point of Rufus obviously is to like, well, it's to make Amazon more money, but it's to, and to do that, you want to give the most relevant results, obviously. So if you ask Rufus like, hey, I'm looking for a gift for my brother who's this age, whatever, whatever. So because Amazon can actually read, if you have lifestyle pictures of, you know, a man in his thirties, whatever, camping and that sort of stuff, Then it can actually use that as a data point of making that potentially more relevant. So that part I would actually not skimp on. If you have your target demographic, then definitely put those people in the lifestyle images. Speaker 2: This leads me to something else. When people are using lifestyle images, sometimes, a lot of times, they're outside of their demographic. So they might be targeting, or you might be seeing some younger people that are walking with their kids, but they're really targeting older age, me, with some type of supplement, right? And this is another thing we had. I forget who it was. I'm not sure if it was Colin Raja or The person just before them, I forget I'm old, but talking about the way that you should be writing your bullet point points now, more humanistic, more towards your main demographic and making sure they didn't talk about it during the podcast, but making sure that those demographics also show up on the slide deck. Yeah. So if you're not doing it, yeah, you could be missing out, especially if, You got some old guy, me, and it's targeting younger people. You're not going to get the avatar match. Speaker 1: It turns people off. Yeah. Or for example, some people will have, they'll just, they'll do the lifestyle images, but the head is cut off, right? The head's out of frame, that sort of thing. And yeah, it's really easy. I mean, what, I mean, it doesn't take a lot of research to understand that, you know, shoppers are finicky. If they have an excuse to hit that back button, they're going to do it because they know that there's 10,000 other results that they have to get through. So your goal as a seller is just don't give them a reason to hit back. Don't give them a reason to bounce. So yeah, no, absolutely. I definitely put people, make sure that there's that avatar match. Speaker 2: Now, what about just listing? What are you using now to make sure that your listing is optimized for compliance like right now? There was a lot of changes made a little while back. So the title, the bullets, the AI, the way it reflects on your listing. Do you plan to launch a product on Amazon? Colin Rajah breaks down the exact steps to do it right, and you can watch it right here. Speaker 1: I always like running things through the policies as well too, just as another kind of data point. But yeah, right now I'm using, there's a few agencies that I'll work with and then they're on top of the rules, regulations to make sure that we're not going to cross any lines. But if we do, so for example, one part that I'm still bullish on is obviously the main image, like testing out things that may or may not get approved. I actually got a cool tip from someone at Seller Sessions. What they were saying was what they'll do is try to, so obviously the main image is like the 80%, right? Like you gotta get the click to get the sale, that sort of thing. And we're always trying to push the limits with that. In certain categories, you can push more than others, but if it gets suppressed, it's like a 10-minute fix. You put the old image back, you're good to go. But what he'll do is actually create kind of a dummy listing with a new kind of provocative main image and actually see if that'll take. And sometimes it'll get rejected immediately. So that means they know that they have to kind of work on that image. But if it takes, then it's a good data point that maybe it'll take for their main listing that's actually doing the sales as well too. And then the chances of suppression are less. Speaker 2: So if you're doing a dummy listing though, you're going to have to create something with a new UPC. So he goes right, a complete listing. Speaker 1: So Complete has enough to get it so that it can be listed, but basically what he's doing is trying to see if he can actually make that image. So eventually he deletes the listing, I'm assuming. It's not anything. So probably for the UPCs, probably using a reseller one or whatever it might be. But the idea is would it pass that first gate of Amazon being like, hey, yeah, this main image is good to go or not, right? Because sometimes it gets rejected right away upon creation. And it's just one extra check that you can do before you actually put that more provocative main image on your main listing. Speaker 2: That's kind of cool. Speaker 1: I've never heard of that. I've never heard of someone doing that before. I thought about it like, yeah, you're almost like sandboxing a listing to see if it'll stick. Speaker 2: I did that in 2000 and I lost a lot of money in trading stock. Unknown Speaker: But all right. Speaker 2: So just I see that there's a few questions coming in. And just so you know, especially if you're new to the podcast, we answer all of these questions up towards the top of the hour. So don't think we're ignoring you. We're getting to them. Now, what other type of AI are you using when you are either creating, launching, or just using it for your listing to optimize it now? Speaker 1: So one that I know is really popular or it's getting popular, I don't know if it's been talked about. It probably hasn't been talked about on the show before, but GenSpark. Having a lot of fun with GenSpark in terms of doing deep research. So you're able to, because it can read and skim like Amazon reviews, it can give you a whole kind of synopsis of a category or product, that sort of thing. Like, what are the common objections for it? But I started using GenSpark outside of the eCommerce space. So one of the things that it can do is actually make phone calls on your behalf. And I was thinking like, oh man, this is cool. So I tested it out a few times. You can actually get it to say, call, make dinner reservations, ask any business kind of like any questions and it'll do the thing and then it gives you a written summary and it also gives you a recording of the phone call to make sure it actually worked. It'll give you a summary that the task was complete. So it's almost like having a personal VA kind of doing the stuff that you don't really have to pay for. Speaker 2: Well, you're a Canadian. Do you remember Wildfire? Oh, you probably weren't born then. Speaker 1: Wildfire? Unknown Speaker: No, that's Wildfire. Speaker 1: I mean, I know wildfires, but yeah, that's out in the West. Speaker 2: No, my buddy held the license for this and eventually it went through Bell Canada. But Wildfire was this application. It was very expensive, but it was very useful. So I had it and it was for these first generation phones that you know the big bricks your phone would ring and wildfire would come on and say you know who it is or what number it is and then would you like me to tell them you're not here would you like me to schedule them and it was incredible even to today's standard it was wild but I think they just ate the technology and they just buried it. Yeah I don't know why. Speaker 1: Yeah, that's cool too, right? Speaker 2: Yeah. Speaker 1: And playing back these recordings, you can tell the person answering the phone is kind of weirded out, but I guess I mean, AI is so common now, like they know they're talking to an AI, but they go with it. Like I haven't had an issue where they just like, Oh, what is this crap? And they hang up, right? Like it actually, it actually works. So, so that was kind of cool, but there's some limitations to that. Like you have to, uh, for that particular genspark.ai phone thing, you have to phone a business. So it got me kind of searching, like, is there another one that can kind of, uh, like do more? And is it more, cause now what's happening? It makes sense. Like, There's a lot of obviously general AIs. Now there's like specific things, right? So one of the ones you might not have heard of is 19pine.ai. Speaker 2: No, I haven't heard of that one. Speaker 1: Okay. So this one's cool. So, and to be fair, I haven't used it to the point of like seeing how well it'll accomplish the tasks, but if it can, oh my God, it's a game changer. What this one does is basically, it's kind of like your personal assistant, but to save you money. And I'm all about, I'm kind of cheap. I'm all about saving money, right? So for example, say you buy something online from Costco or whatever, the price goes down in a week or two. It can actually call Costco on your behalf. It has your membership number. It has your address, whatever stuff that's needed to verify you on the phone. And it'll ask for that, you know, the price difference. I'm like, okay, like. Sign me up. That's cool. And the way they get paid is only on positive results. So if, for example, they get you that $10 off, that extra $100 off, you actually tip the AI in advance and you only pay. It has a week to accomplish it. I don't know why it's a week, but it has a week to accomplish the thing. And if it accomplishes the thing, then and only then does it get paid. So there's no downside, really. Speaker 2: That's crazy. There's so many tools that are coming out. By the way, Parsippany, Steve Simonson's app. So he's got an app. It is a chat app that you can easily program with all the questions. So somebody comes to your website, you know, it's a customer service thing. But he's got an outbound app and that inbound outbound will call in a human's voice and just start a sales conversation with the person. Like I've been on the phone, he's been demonstrating this and I'm going, oh my God, this really It really sounds like a human. If you haven't checked it out, you might want to contact Steve, but it's one of his apps in Parsimony. Want more unfiltered tips from top eCommerce experts? Well, hit subscribe and you'll never miss a Lunch With Norm episode. Speaker 1: That's cool, yeah. I can speak to, there's something else we're doing at Product Pinion. It might be applicable to some brands as well, but it is around cold email outreach, but it's super smart, it's super intelligent. Yeah, I'm trying to think how you... Speaker 2: What are you doing with cold email outreach? We're always looking for that. Speaker 1: Okay, so this is pretty interesting, or at least it got me super excited. I mean, there's a bunch of apps that'll do this, but there's an app called RB2B. And what it'll do, rb2b.com, it'll actually, if anyone is more or less logged into LinkedIn on the browser at some point, it can actually identify you. To the website owner. So basically, you install the little pixel rb2b on your site. And if you visit that site, and you have rb2b installed, I can see who's actually visited, say, for example, the product opinion site. So if they visited, I mean, they're not really cold, right? They're kind of warm. So, you know, maybe they didn't hit free trial, maybe and I can see which page that sort of thing. So what you can then do is once you have their information, you can, I mean, you can, you also get their email by the way too, right? Like on LinkedIn, you can get the email. It's pretty, pretty. Speaker 2: Does it give you the address? Speaker 1: Or like a mailing address? I don't think so. But definitely email address, definitely LinkedIn profile, all that jazz. You can add them to LinkedIn. But what you can do and another way to get the address, there's another app that's really popular for sales funnels called Clay, clay.com. So Clay is interesting. It's basically a spreadsheet, but it's like super powered by AI. So it'll import the leads that the visitors that are on your site with clay you can then enrich them and Enrichment is just like hey, can we find more information about these guys? So maybe we could find their shipping address right which actually is also a very cool way to get attention to people like ship them physical writing. Exactly. Nobody does that anymore. I love that stuff. Pattern interrupts, right? But you can enrich them. But not only can you enrich them, once you have the enrichment, You can get the AI. It's built in. You can use any of the tools, Claude, OpenAI, whatever. You can get them to get Clay to write a specific subject line and a specific email per contact. So you can get enrichment. You can even have, for example, they call it a Claygent, which is just an AI agent, do some like deep research on that visitor who just came to your site. And then use that deep research to actually craft a specific email to them. Then you hook that up to an email provider like instantly.ai and it'll actually cold, cold, warm outreach to the visitor on your page. So basically it's hyper personalization at scale. So all of this is happening behind the scenes. You're sending like a custom, hopefully thoughtful, Email to these visitors and obviously long-term play is getting them to convert, getting them to do the trial, getting them to sign up. So yeah, this is something that we're working on with ProductPinion to be able to kind of just give better, because everyone hates a generic cold email, right? It's kind of like, you didn't do your research. I'd rather not send a crappy email. So in this way, it's being able to actually like personalize that cold outreach with a few AI tools. Speaker 2: And hopefully it doesn't say, Dear Sir, or, you know, Norman Farrar, Dear Norman Farrar. Okay, that's going in the trash. Speaker 1: The amount of those that you get is just insane, right? Like, how do you expect me to, like, I'm in marketing and you took zero effort to personalize, you know, whatever, like, ugh, in the trash. Speaker 2: We're at the bottom of the hour and in case this is your first podcast, listening to Lunch With Norm, we do something a little different. At the top of the hour, we do something called the Wheel of Kelsey and that's a giveaway that our guests provide. So Matt, what's the giveaway today? Speaker 1: Yeah. So what we'll do is if you don't have a product opinion account yet, we'll set you up with one. We'll give you a hundred dollars worth of credits. And then on top of that, we'll actually go through, we talked about our five-step main image SOP. We'll actually give you a personalized blueprint on how to implement that to get a main image. That's going to get you more clicks and conversions. Speaker 2: Oh, very good. Okay. So if you're interested in this, I hope you are. It's hashtag WheelOfKelsey. Tag two people, you get a second entry. Now, there is a third way. When you go to LWN.news or if you have our newsletter, you'll see an entry for this on Monday for the Wednesday drawing. So this is actually a little bit different now. We used to do it so Matt's would have been at the top of the hour. It's going to be at the top of the hour next week. So it gives more people a chance to join in. All right. And the other thing, don't forget to hit that subscribe button or share this episode. And last thing, let's have a word from our sponsor. Hey, Amazon sellers. If you've listened to the Lunch With Norm podcast over the years, you've probably heard me talk about Flatworld Network and some of the services that we use to help brands find and maintain serious growth. Well, if your sales have plateaued or you feel like your listings just aren't converting, you're not alone. I've built a proven system over the years to help sellers increase conversion and unlock more sales. I'm going to be giving away a free audit to show you exactly what's wrong with your listing and how you can improve it. In fact, I'm going to be completely transparent and I'm going to let you know the good, the bad, and the ugly. So if you can handle the truth and you want to scale, hit the link in the description or go to flatworldnetwork.com and grab your free audit today. That was me by the way. Hey, I just noticed that Simon put in a good point. In the UK and the EU, you need explicit permission. And generally, you can pick out if the IP is coming from Europe. Speaker 1: To speak to that, yeah. So what I was talking about is US only to identify actual individuals. However, you can identify companies in the EU, and it's still above board. And when you identify the companies, then you can still throw that into Clay, and then Clay can work its magic to find the people that work in those companies. You still can work with it. But yeah, what I mentioned was like US only, you obviously have to have the privacy policy, the acceptance of cookies, all that fun stuff in there to do it so that to make sure that it's above board. But yeah, really good, really good point. Speaker 2: Yeah. But the other thing, if you're working on LinkedIn, there's no problem sending out individuals messages to connect. So if you know who the people are and you're, you're finding out, you know, uh, uh, potential clients or potential businesses that you could work with, then that would be fine. The other thing, uh, what's interesting, this came out about, I don't know, maybe four or five years ago. We've used it. And this is more for if you're crafting that, uh, email. Oh my God. Have you ever used crystal? Not magic. Speaker 1: Neither. I haven't used them. Speaker 2: Okay. Crystal is an app for LinkedIn. This is going to blow your ways. Simon, Steve, anybody who's doing it, check this out. Go download Crystal. It's an app that will tie into your LinkedIn. And if you put in anybody, try yourself, You're going to go, holy crap, this thing knows me. Now, this was years ago that this used to happen, but we still use that to identify how to talk to people. So if we know that it's more of a hard sell or more of an analytic or more of a somebody who needs more creative, it goes into a deep dive into the person's personality. Speaker 1: That's cool. So one of the examples that they gave with clay.com was, say for example, you wanted to do deep research on a specific person and you already have their LinkedIn profile. You can actually get it to also find any awards, any blog posts they've ever written, anything basically that they've signed off on, and similar idea where it'll then take that information and then craft the email based on what they find. So, hey, I saw your blog post on blah, blah, blah, blah, blah, and that sort of thing. So you can actually do some like deep dive, like get the agent basically to do the deep dive on an individual basis too, which is Which is pretty wild, but you can do that now. Speaker 2: I just found the page. It's a Chrome extension. Speaker 1: Cool. I'm going to check that out. Yeah. Speaker 2: All right. So let's talk about more further development. Any other types of AI that you're using right now or just in general, how you're using it, maybe even outside of Amazon? Speaker 1: Yeah, totally. So, I mean, there's a few things that, I mean, obviously I use OpenAI quite a bit. We use it for product opinion and then I'll use it personally. So, Norm, I'm sure you've had flights that had cancellations, delays, that sort of thing, right? Speaker 2: Too many times. Speaker 1: We're in Toronto. Yeah, I know. Do you fight them though? Do you actually look for how much compensation you can get? Speaker 2: This is probably the wrong person to ask because, well, I think you met John or Melissa anyways. So Melissa Vong, her boyfriend has a company that will fight for you as long as you don't cash in any coupons. So like, for example, we had a problem in North Carolina coming back. And they said, okay, you're going to have a $700 coupon. Well, we waited a year and a half to settle for 300 bucks because the government, or they bought it. He goes and he fights for you. And I think it's going to be along the same lines. Speaker 1: Yeah. Speaker 2: Why don't you explain? Yeah. Speaker 1: I was going to give John a shout out too. So it's airfairness.com is John's site. And oh my God, I didn't realize Just how big of an issue, I mean, okay, so more or less my story was, I tried to fight it on my own. I thought it'd be easy. I thought it would be like, hey, Air Transat, you canceled the flight, you rebooked it in 24 hours. I'm out of pocket all these expenses, plus because you're a big airline, whatever, whatever. It's supposed to be $1,000 just compensation. Never even mind your expenses. It's compensation plus expenses, plus this, plus that. These airlines are, they do it in such a sneaky way that they're almost, I think their first line of defense is just deny and just mechanical shit. Yeah, they'll come up with the nonsense. I did the application myself thinking everything would be okay. Month later, I get the customer service reply saying like, oh, it's not eligible because of some nonsense with code share. Some people would just give up. Some people would just be like, okay, whatever, I tried. But it turns out anyway, it was BS. And they knew that it was BS when they were telling me. So I went through this painful process. AI did make it like slightly easier, but it was like, Every email that I would send, they had a way around of denying it and it took a lot of energy and effort to get it. And at that point, I'm like, I get it. I get what John is trying to do. I get how he's trying to fight back because they have, and I almost guarantee if I had John do it and I messaged him, I'm like, dude, like I get it now. Like I should have, I should have just had you take care of this. It's wild. It's wild that they'll go, these airlines will go out of their way. And what John's company does, he knows all the in and outs. I'm assuming some of these airlines actually know John's company now too, so they don't mess around. Because there is also like another layer, like if the airline's giving a hard time, you can escalate it to another kind of agency or whatever it is. So these airlines probably know that if air fairness is involved, they're not going to mess around. They're not going to delay it 30 days, 60 days, whatever it is. John will go in there and get you your money that you're supposed to get right away. So yeah, big, big shout out to John. But yeah, I get it. Speaker 2: You know, I remember coming back from Iceland. I was talking to Gracie, you know, Gracie Ryback, and she had a problem. And it, I think nothing was really happening. I believe she contacted John and all of a sudden this pretty much zero compensation or just a few bucks turned into a thousand bucks. So I could go on. We could have a talk about this, but if you want to hear more about this, the Marketing Misfits podcast has John on with his partner talking about all of these issues that come up with travel. And it's a shock. What they do, they just say, oh, it's mechanical, then they don't have to pay anything except, you know, here's a few $10 worth of snacks. Speaker 1: Yeah, exactly, exactly. And it's just crazy how they literally go out of their way to deny everything. But it was an eye-opener and now it's like, okay, I have any issues, air fairness is going to take care of it going forward. Speaker 2: Where do you see the biggest trend right now? Speaker 1: For AI, I think, yeah, so I mean, I see things. So the hyper personalization, I mean, it's already kind of happening as well. But, you know, we're going to get to the point where Amazon knows Everything about you, your shopping habits and everything. So it's going to get to the point where it's going to know you really, really well. And the products that are going to be recommended are going to be recommended obviously based on data and based on your shopping behavior and habits. So we're already seeing that, but I feel like that's definitely going to be like more mainstream. It's like your personal shopping assistant gets to know you just like any other, you know, someone you hire, whatever it might be. And it'll make those super personal recommendations for you. Speaker 2: I had a, well, you know, Kevin King's my partner and he contacted Amazon. He got a report. He couldn't believe it. 780 pages of everything he's ever touched on Amazon. Everywhere he's gone, every page, every, you know, click, they know him inside and out. So if you think, oh yeah, You know, they, they know a little bit about me. They know everything about you and you can suggest or ask them for a report on yourself. So that's pretty scary. And you know what? Facebook does the same, you know, so any of these are just tracking the hell out of you. Let's talk about the future. Where do you see this going? And I know this is a hard question to answer. Speaker 1: Yeah, I mean, I'm also really excited about shopping actually coming up right in OpenAI, so right in ChatGPT. So I think they're doing it right where they're not being, or like, as far as I know, and as far as they're claiming, they're not being paid the affiliate. And I'm here to talk to you about how you can make money to recommend these products. Perplexity is, right? So all of a sudden it's like, hey, I'm going to, you know, like that change. I used, I was using perplexity for a lot of shopping. I stopped. Me too. Yeah. So knowing that, so if, if OpenAI champions that where they're, they're, they're like legitimately not taking the referral money, whatever it might be, then I can see that, you know, that's where I want to shop. That's where I want to, I want to see. So I think that's an untapped, I wouldn't say untapped, and obviously you're gonna show up in OpenAI right now if you have the Shopify store and you do all the right things, but that's just gonna grow. That's gonna be big. And there's ways of making sure that the OpenAI bot can see your Shopify store and all that fun stuff. So I feel like that's an interesting thing where, yeah, people are just gonna go to, OpenAI to get shopping recommendations. And maybe that's a shift away from Amazon too. It's like, hey, Amazon knows everything about me. That's kind of creepy. I want a third party to kind of make recommendations that aren't necessarily biased. So maybe people will shift to do more shopping on ChatGPT with OpenAI. Speaker 2: Yeah in a billion dollars sellers the newsletter two weeks or last week Kevin put in an article about the different ways to get found on AI and one is the LLM s.txt file like the robot.txt file. So just these little tips, you know, get you that much further. So you have to be on top of it or have to have somebody who understands AI to be on top of it. Okay. I got one last question. I just thought of it when you were speaking. Where are there, and there's tons of them, I know, but are there any tools that we haven't discussed that's kind of flying under the radar and we should give it a shout out? Speaker 1: Ooh, that's a good one. I know, okay, so there's one, I mean, I'm on the inside of it, but Dorian is putting together, so Dorian Gorski, like I think he's one of the best in the space in terms of like understanding how to get clicks, how to get conversion. So he's working with the OpenAI APIs and other ones to actually create a, so his app basically, you give it your ASIN, it'll spit out like 10 variations that you could potentially make for your main image. Some of them are a little wonky, but some of them are like actually almost perfect to upload to Amazon. And the reason why that's important is because it's based on not just the AI, but it's based on Dorian's experience of like 10, 15 years of working with Amazon sellers and creating new images that get the click. So he's working on that. It's in beta. We can put a link somewhere. And yeah, obviously it's something to test and play with. I've played with it and it's kind of crazy how good it is. Speaker 2: Yeah, so maybe if you could get us a link, we'll send that over or put it into the show notes. Speaker 1: Totally, yeah. Will do. Speaker 2: Okay, so if you haven't entered yet, it's hashtag wheel of Kelsey, tag two people, you get a second entry and you'll be entered into our wheel of Kelsey. This is going to be drawn next week. And that's going to be for a hundred dollar credit on product opinion, as well as their SEO, their blueprint on optimizing images. That's correct, right? Speaker 1: Yeah, yeah, well, and it'll be personalized too. So we'll actually create your main image and we'll put it through our blueprint that's been doing some crazy stuff as of late. Speaker 2: Perfect. Okay. So Kels, let's go do a sponsor and we'll be right back. Unknown Speaker: We do have a few questions if we can. Speaker 2: Oh, oh, yeah. Okay. Unknown Speaker: So first one's from Rashid. As artificial intelligence continues to evolve, what are some of the most promising ways it can be applied across industries beyond eCommerce? And to follow up, how can organizations prepare to embrace these changes? Speaker 1: So for example, when I was talking about the flow of like, you know, cold email outreach or warm email outreach, identifying visitors, that sort of thing. That same process can be used for any sales process. So even if you're in eCommerce, you're still in the sales business of say, maybe, you know, you have the retail webinar coming up, but say, for example, you want to contact buyers and that sort of thing. You could have a flow specific to buyers. So, I mean, it's kind of like every part of the organization is touched by AI. At Productpinion, we're working on patents. So we actually have two patents in the works. We've used AI, obviously, to help do the deep research of what's prior, what's there, and then also help, obviously, in writing it alongside the lawyer, too. So, I mean, every aspect is kind of touched with it. There's also a very interesting podcast episode. It's Diary of a CEO, right? And he had on there, Jeffrey, what's his name? Jeffrey Hinton. You guys should watch it as well too. So the whole podcast is about, he basically, he's called the Godfather of AI. And this whole podcast is about how like, we're in trouble. Like AI is going to be like the extinction of humanity. So as much as it like helps and everything, there's also the school of thought of like, What happens when we make the AI too smart? It renders humans useless and then we become a slave. Yeah, yeah. It's Terminator, right? So, really interesting watch. I'd recommend checking that. Speaker 2: Watch Humans. I don't know if you ever watched that show. It came out years ago and you'll be surprised, you know, how close this is. I mean, just watch it. It's called Humans. I believe it's on Amazon. Unknown Speaker: Okay, and our last question is from Simon. He asks, so we have our persona and hooks, etc. What is the best image creation tool to use? Speaker 1: Well, I was going to say too, Jo on LinkedIn, she has a ton of these SOPs on how to use... So I mean, there's the tool, but then there's also the prompt, right? So in terms of tools, sure, use OpenAI, ChatGPT, their image gen is phenomenal, but it comes with having to master the prompt of it. We have a tool called Variationizer.com. It actually generates AI stuff in layers. So you can actually manipulate it. So it's not just like a solid image and you're stuck with it and you have to kind of chat with the AI to change things. This will actually take your main image. You can generate like eye candy around it. And because it's a layer, you can move it around and that's fully free. So you can check that out at Variationizer.com. Otherwise, if you wanted to use OpenAI, yeah, I would, we'll put a link to Jo's stuff. She has, she actually won in Iceland, the hat contest with BDSS with how she does the prompting and her prompts have went viral. 15,000 comments on LinkedIn and that sort of thing. Speaker 2: So that would be like one for 23 million. Speaker 1: Was it? Yeah, right on. Speaker 2: So, and if you haven't subscribed to our newsletter, she's got a good newsletter. It's AI for eCommerce. Speaker 1: Absolutely. Yeah. Speaker 2: All right. Let's go over to the sponsor and then we'll head right over to the Wheel of Kelsey. Unknown Speaker: Start, scale, exit, repeat. I'm Colin C. Campbell and I've started over a dozen multimillion dollar companies in the last 30 years. I spent the last 10 years writing the book Start, Scale, Exit, Repeat to figure out what it is that these serial entrepreneurs do over and over again. We interviewed over 200 people. We created 58 chapters, over 30 illustrations, 180 call-outs, and we quite frankly made this book for the ADHD entrepreneur. It's been number one on Amazon in 15 categories and has won 12 awards globally. Get your book today either on ebook, paperback, hardcover, or audible on Amazon or your favorite bookstore. Speaker 2: Do you know Colin? Speaker 1: Yeah. I've spoken with him a few times. Speaker 2: Yeah. So when he was 28 years old, we were in the same EO forum. He had a company valuation at $1.2 billion. And when he says, Oh, I sold, you know, multimillion dollar companies, like he's talking about multiple, multiple, multiple, and then eight figure companies. Like he's a, yeah, he's got the Midas touch. He's a smart, smart guy. Speaker 1: Hot damn, that's incredible. Speaker 2: All right, Kels, let's go over, well first, contact information. Matt, then we'll go over. Speaker 1: Yeah, so easiest is matt at productpinion.com and yeah, happy to answer any questions about clicks, conversion, AI, identifying black widow spiders. I'm your guy. Unknown Speaker: So we have our wheel. This is for Colin Raja's giveaway of his hour consultation for the Amazon product launch. So we'll spin this up and let's see who the winner is. If you are the winner, please email us k at lunch with norm.com. Speaker 2: Ooh, Marina, congrats. Another Canadian from Toronto. All right. We ran one minute over, but we started one minute late. So Matt, we kept you to the hour. Speaker 1: No problem. Speaker 2: All right, buddy. Well, thank you so much for coming on the podcast today. Speaker 1: Thank you for having me. Always a good time. Speaker 2: Alright, we'll see you later. Thank you so much for joining today. We couldn't do this podcast without you. If you'd like to see more or see some of our other communities, there's always the Polish podcast. You've got the YouTube clips that YouTube or at Lunch With Norm shorts, which are like three minute clips and under. There's the newsletter. There's their subreddit. There's the WhatsApp group. Facebook group, you name it. We've pretty much got the span of all social media. So check it out and either be part of one, be part of everything. That's up to you. All right, everybody, we'll see you next Wednesday and enjoy the day.