EP381: Amazon seller Meta ads: stop resetting the learning phase is the Amazon FBA problem, and the takeaway is let it run

This is the High-Voltage Business Builders Podcast, daily intelligence for serious e-commerce…

This is the High-Voltage Business Builders Podcast, daily intelligence for serious e-commerce portfolio builders across Amazon, TikTok Shop, Shopify, Walmart, and every channel that moves the needle. Neil Twa and his Voltage team all day, every day since 2012. Let's get into it. All right, folks, welcome back. I have a wonderful guest here today, and we're gonna talk about ads, media, creatives, everything marketing. So if you wanna learn about marketing, you're in the right seat today. So I'm gonna bring Krista on here, and I'm gonna start off with a question that just came to me, okay. Meta ads, love them, hate them? I would say I love them. Okay. Always been a big fan of meta ads. And why is that? Well, I feel like it was kind of my first love with marketing, but the reason that I love them so much is because I truly feel that the algorithm there is unmatched, you know, despite all the changes we've been through on the platform over the years with, you know, the iOS changes and the craziness that ensued with that. Yeah. It still really is the best platform, in my opinion, to most effectively reach your target audience. You know, they got a lot of data on all of us, they probably know more about me and you than we know. If it's free, we're the commodity, right? Yeah, and they're very effective at being able to serve the ads to the person most likely to interact with them, and on all the other, you know, social advertising platforms, just haven't found that same result compared to meta. Yeah. A lot of users, a lot of algorithmic data, a lot of changes. I think last month or the month before last, we exposed, experienced probably the most meta outages we've ever seen on the ads platform as they change things. They just messed with the algo not too long ago that's impacting people different. What are some of the things, give me three things right now that people should be not doing with their meta ads on their marketing, and give me three things they should be doing. Three things they shouldn't be doing, I would say, one would be hyper-targeting, just doesn't work anymore. Interest-based targeting, okay. Yeah, interest-based targeting, or even location targeting. We still have clients who come to us and they're like, well, we see better, you know, traction in these areas. And we've even seen where like on Google, there's certain, you know, areas where we may like bid higher and try to translate that same strategy to meta. It just doesn't work. Keeping your targeting super broad, that'll be point one for what you should be doing. Is that country-wide to be specific, or are you just talking about location? Like don't eliminate states, don't eliminate certain things, or is it country-side targeting? What are you talking about? We're typically seeing the best results by targeting the entire United States or whatever country that you're going after is going to typically yield the best results. Okay. Another thing that I would say you should not be doing, and this is, I think, you know, overall time is probably tinkering with your ads too much. You want to let it run. You want to let it learn. You want to let meta kind of do its thing. And if you're in there too much turning things off or adjusting things too much, you're going to reset that learning phase and not allow meta to effectively run those ads. That's kind of always been the case. And then I guess the third thing not to do, this is kind of a big one, is have a disjointed strategy between the backend management and the media buying and your creative strategy. We speak with a lot of clients where maybe they're having an agency do the backend, but then they're doing the creative. Meta is highly reliant on the creative strategy. So if those two things are disjointed, you're not going to see, you're not going to see results in the account. Okay. What different campaign types, right? We obviously have lead and objective and purchase, and this is kind of a standard across the industry, but what do you see being the most effective type to target? And what would a, you know, an ABO or a CBO campaign structure look like, and which one do you prefer? So, yeah, seeing as we work predominantly with e-commerce brands, we pretty much optimize all of our campaigns for what we want, the end result, the sales conversion. So I would say 90 to 95% of our campaigns are going to be optimized for sales. We may have a traffic campaign that we run in some of our clients' accounts. However, we always make sure to optimize those for landing page views and not link clicks in order to help drive more quality traffic. A lot of times what we find with those traffic campaigns is yes, we're getting more traffic to the site, but our conversion rates are dropping because it's not the most high quality traffic. And I'll go on the opposite spectrum here of where our sales campaigns, we wanna keep that targeting super broad and target the entire United States. We've actually found with traffic campaigns being a little bit more specific, an old trustee 1% lookalike audience of your purchasers helps to, again, drive a little bit more quality traffic to the site. And we may do that for our clients who need a little bit more of a traffic boost to help rebuild up that remarketing funnel. But typically it is going to be sales. We typically always start with, yeah, CBO will set the budget at the campaign level. However, we have very few campaigns running in a lot of our clients' accounts. We really believe in account consolidation. So within those campaigns, we may have several different ad sets running where we're testing variations of broad audiences or segmenting the creative in a certain way. Sometimes it makes sense to move the budget to the ad set level in order to give those ad sets fair spend, but we'll typically start at the campaign level. So CBO level campaign budget, you break down ad sets by targeting, variations of targeting. And on the ads level, now with this algo update, there's a lot of scuttlebutt around creative need versus stagnant creatives versus duplicating creatives and unique creatives. Maybe you can kind of unpack some of this myths I'm hearing online. I hear both sides of the story. One guy is saying, hey, you have to have all these new creatives. And the other guy's like, no, you don't. You just have to duplicate the winners and this kind of stuff. And I'm like, okay, which one is actually true? It's the Andromeda update that's kind of screwed everybody up, right? I know for sure. It's a crazy time because what Andromeda truly wants is volume, but also I think a lot of people get that wrong. When they think of volume, they think, okay, what's the easiest way that I can create volume? It's by doing a slight iteration off of an ad. So maybe I have the same ad with five different backgrounds or the same ad where I change out one word. That's actually not what Andromeda wants because if you do that, it's going to assign all of those ads the same ad ID. It's essentially going to serve those ads to the same audience. What Andromeda wants is true creative diversity. So yeah, it puts a lot of work on us creative strategists because it wants different concepts in order to unlock those different customer personas. For example, back in the day, we would say, yeah, we want to target somebody interested in an engagement ring or interested in fashion, whatever it was. We don't do that now. Now we set the targeting super broad. And instead what we're doing is we're unlocking those different customer segments with the creative. So you're speaking to a certain mom audience with this creative. You're speaking to a Gen Z with this creative. So that's really how you are targeting those different segments and how meta is unlocking those audiences is going to be through that creative strategy. I will say though, I think a lot of people get really caught up in the creative volume side of it. And yes, while technically that is what Andromeda wants, it wants that creative diversity, it wants that volume. For a lot of our clients accounts, we do still see solid results by we're not blasting the account with a ton of ads. And I think a big part of that comes down to the budget. So a lot of times when meta comes up with these updates or you're speaking to some of these agencies, yeah, Andromeda works when you're spending a million dollars a month in ad spend. But if you're a starting brand who's spending a couple of thousand dollars, $5,000 a month in ad spend, it's not going to be the same strategy of uploading a hundred ads. It's going to negatively affect the account and really waste your ad budget. Yeah, plus it's being that creative heavy is challenging for some folks, even with AI and tool sets today to build out something that's unique or something that they feel is gonna add value. And I think what people do is they get caught up in it. So if we're going after this a little tactically, so people can kind of follow along with a CBO campaign, different ad sets to variation, how many creatives should they go at like budget level? Say they're gonna spend 25, 50, 75, 100 a day. What do you recommend they use in terms of creative so they can kind of wrap their minds around what they should put into each of those ad sets? Yeah, I would say, I mean, if we're spending $100, I recommend a minimum is always gonna be $150 a day. If you are targeting everyone in the United States, you wanna have at least enough to be competitive there. But I would say at least six ads per ad set. Completely unique, copy, image, video, whatever is completely unique. Right, exactly. Sorry, I didn't mean to cut you off. I was just clarifying. No, you're good. That was pretty much the answer. Oh, that was the answer. Okay, cool. Well, that was simple enough. So for those of you paying attention at home, just to clarify, we should have about 150 a day in budget, which means you're gonna spend at least, I would say conservatively for the first seven days, would you agree with that? Yeah, I would say. And yeah, that's always just our recommended minimum. Of course, I mean, the scale possibility, also why I love meta. If you are seeing strong results because there is such a large audience on Facebook and Instagram, there really is a huge potential to scale there. So that's kind of a baseline to start. But if you're hitting your KPIs and you're seeing those results, you definitely should scale up because it's very possible on meta. On that 150 a day at the campaign budget level, how many ad sets would that be? Typically, if we're doing $150 a day, again, we are going to try and keep the account consolidated. We never wanna be spreading our budget too thin across too many campaigns, not allowing for effective learnings. So if we're doing that minimum budget, I would say at most, we have probably three campaigns running. 150 bucks per campaign. Yeah, and within those, we may have like two to three ad sets within those specific campaigns. Again, keeping it super consolidated, not testing too much at once, and then really focusing in on the creative strategy. So with this strategy, what have you seen? What has the results been? Give me some case studies, some examples you've got. Well, I can say, I mean, year-to-date across all of our clients on our average return on ad spend is about a 3.92. Looking at a specific case study, one of our favorite clients is Heretic. They are a perfume brand. They're a great example of a client who was able to effectively scale on meta. Now, they had a lot of good stuff going on their end. They did a collaboration with the movie Nosferatu. That definitely helped. But yeah, over a 12-month period, we were able to bring their meta ROAS from a 3.46 to a 4.82. Their total purchases went from about 16K to 34K. Their meta CPA, we were able to drop from about $23 to about 18. So yeah, really, really strong results for them. Excellent. It sounds like they had some brand power behind them, which is great. What about the business who doesn't have as much brand power going yet? Give us some examples of those you worked with. Yeah, it definitely helps to have a little bit more of that brand power. It's always going to be a different case for every brand, but we always give kind of the general starting point of you want to give it that six-month runway of really going heavy into kind of the testing and optimization in those early months. And we're always very clear that during that time, you're probably not going to be hitting the results that you want to see because we are spending our budget to get these learnings. We're trying to figure out what creative is resonating, what messaging is resonating, what campaign structure. While we work with a ton of clients and we generally know what works from brand to brand, it is different. It is different for every brand. So really what you have to do is you got to just put the money up there and you have to get in there and you have to test. Usually by about that four to six-month mark, that is when you start to get those learnings and you can double down on what is working and spend less on testing. And that's when you start to get into a group, into the account. Yeah, that's good experience. How did you get to this point? I mean, this is your business. Were you a ground media buyer, number one? Yeah, okay. Tell us about that. I was, I was. So yeah, when we first started out, so it was me and my co-founder, Nicolette, we were kind of doing a little bit of everything for everyone. We did organic social media, we did branding. I think we even dabbled in like influencer and eventually we realized this is why we weren't getting clients is because we were doing too much for too many people. So we really niched down to e-commerce and then we focused in on meta ads and email marketing as our main services, mainly meta ads because yes, I really did fall in love with it. I was the creative strategist and media buyer from day one. And yeah, I still, still love guiding a team on that. I still love digging into the accounts. So yeah, that happens. So like geeking out at the operator level. Yeah, yeah. That's cool. How long have you been at this? So let's see, it's almost a decade now. So yeah, we have been doing this since about 2017 as Milton said. Yeah, that's enough time to kind of know what you're talking about, I think. Yeah. Yeah, enough brands to have worked under your belt. How many are you now in your operation? Our team or how many brands we've worked? Yeah, our team. Oh, in your operation. Yeah, we're a team of 10 currently. Excellent, all media buying? No, we do, we have strategists, creative strategists, designers, media buyers, and some kind of like project managers. But yeah, we're a lean team of 10. Cool. Is AI involved in any of this? I mean, it's a big buzzword, but are you actually putting it to practical use? We are. Yeah, we use Claude. Claude started on ChatGPT, moved to Claude. Interesting, I've heard that a couple of times lately. Same story for a lot of people. Yeah, AI is crazy. So we had an employee go on maternity leave on end of April. It's now. If you want the CEO operator blueprint, join Neil Twa, CEO of Voltage on the next live workshop. He goes over the five steps to building generational wealth through almost automated income with FBA. Save your seat at voltagedm.com slash AI workshop. And now back to the podcast. What? End of June, so two months. By the time she went on maternity leave at the end of April, we were still on ChatGPT. We were using it to maybe help with right ad copy. And now, literally two months later, we're fully entrenched in Claude. We're using it to do a lot of the analysis within the accounts. We're connecting it to our creative analysis software, to the meta accounts, to Shopify, to pull in all the data. And it's just crazy. She's gonna come back and we're like, hold on. This is completely changed. A whole different job by the time she comes back. Pretty much. But yeah, we're thinking about how we, everything we're doing, we're thinking about how we can use AI to do that. So it's kind of a slow process of integrating it into everything we do, but I've challenged our team. When you're thinking of doing something, think, hold on, can AI help me with this first? So we're slowly building that out with more structured kind of systems and processes within the team of how to really use AI. And it's pretty awesome. It's a pretty awesome tool. I expect you're going to have to do that because your competition is. I would doubt they're not, right? 100%. So yeah, that's another thing is if you're not, if you're not using AI, if you're not embracing it, you are going to get left behind. It's just the way that it is. I mean, like I said, in the course of two months to see how much it has advanced in that time, it's only going to continue to get better. And I think, you know, a lot of people are worried about replacing jobs and yes, it will probably replace some of the things we do, but really it's just helping you to work at a higher level. We don't have to waste our time, you know, with data input or like implementing. We can focus on higher level tasks and AI allows you to do that. At some point, we'll just have this podcast. It'll be your AI person talking to my AI person. Exactly. I already feel that's how email is. I feel like it's one AI communicating with the other. Right? Instead of a, it used to be the joke, instead of a meeting, this could be an email. Now we're going to be saying, instead of a meeting, this could have been a cloud prompt. Pretty much. So we're getting there. Could this be a cloud skill? Do we actually need a meeting for that? All right, exactly. It's coming along fast, especially in the e-com world. We're six months into 2026 at the time we're recording this. Hopefully it's been a good year for you or a great year. How do you feel about the next six months? What are you foreseeing? Do you do any forecast and planning or is that on your radar at this point? And you know, what are your thoughts about the end of the year at this point as it comes in the next six months or so? Yeah. So I know, crazy. Time flies. I just realized as I said that out loud where we are. At this point, yeah, what we're already starting to think about is Black Friday, Cyber Monday, Q4. This is when we start thinking about it for our clients, thinking about the prep for that. It's definitely our busiest time of the year. So we always want to make sure that, you know, brands that we work with are starting to think about that now. It's crazy how this happens every year, but we get brands coming to us, you know, in October. Right. Asking about their Black Friday, Cyber Monday strategy. I was like, you should have came to us. You need to come to us in like August at the latest. It's too late now. We needed to have a lot of stuff in place. So we always challenge, yeah, the brands we work with to start thinking about that now. Well, certainly if you're gonna have success in their physical product businesses, which I think you work a lot with those, then you're inventory up because if things go well, you can't run out of stock halfway through Black Friday. So that's one of the things on our, you know, June planning is, do we have enough product pipeline and, you know, get cashflow set up and everything to ensure that anything is being ordered or was ordered a month or so ago will be here before we have to reorder again for the fourth quarter. So that's all the conversation we're having right now and all the brands is, you know, what are we doing with inventory levels for sure right now to scale up? You are on the other side of that, which is you would hate to have a brand that's kicking butt and find out they run out of product. That has definitely happened. And yes, it is. I guarantee it does. Yeah. Well, that's awesome. Hey, did I not ask any questions that you wanted to cover that we haven't covered yet? Let's think here. I mean, no, I feel like we delved into the meta ad side of things and yeah, touched on AI. Speaking of that, I got one. The creator side. So are you pure creator for creators or have you already started dipping your toes in the AI side for your creator side? Thought about it. Yeah, dipping a little bit. We are, we do use all human creators currently. So, and that's another thing that I forgot to mention. That's a big thing on meta right now that they are pushing hard our partnership ads. So they, you know, this whole volume within Andromeda thing, a lot of people think, yeah, the way to achieve that is with AI, but recently attended the meta summit and AI generated creatives maybe can work for more of your bottom of funnel, but they stress that more human led content. So, you know, still by being done by us really works better for the top of the funnel. So more of those creative concepts and then of course the UGC, the creator content, the partnership ads, that is just honestly, meta is pushing it very hard and it does work in the account. With creators, you, I mean, the whole point of UGC is for that content to feel authentic, right? So if we are using AI creators and I've begun to look into that and it is crazy, these AI creators. Some of the stuff I've built with it, I'm like blown away. Cause if you're, you know, I'm 50, which sounds like when I say it out loud, I'm like, that's really old, but I'm not actually old. And then I look into the, you know, the sixties or something, or even see like my in-laws are looking, I'm like some of those videos, they're not even going to know. They wouldn't know, it's crazy. They look so real. Really real. However, I know there's a lot of like laws and stuff that you have to disclose within meta that this is AI generated content. So that could, you know, different for the whole purpose. This is authentic content. And, and. AI sloth, right? Yeah. They don't want their system taken over by complete AI sloth, which I get. So they're pushing that a lot and that there's a lot in New York that's being drafted right now. That's supposedly going to fine you about five grand per pop if you don't disclose that you have AI. So I've did quite a bit. That is our hesitation. Yeah, right. Well, and there's other things, of course, watching out who you SMS like Texas, you know, when you send out SMS messages, be careful. That's a nasty little fine you can get if you start blasting text messages into Texas and not are paying attention to the law. Yeah. That's a very costly mistake in doing that now. So I see how some of these things are, well, they're sort of catch-22s. You want the creativity and the fun of being able to build those things. And then, you know, there's going to be nefarious people out there trying to jack grandma with a fake video, right? Or deceptively promote ads. I mean, I'll tell you, the worst in the world right now are the drop shippers. Like the drop shipper people are absolutely blowing it up at the UGC stuff and it's all fake content. And I'm just watching these videos going, you guys are just going to wreck the rest of the industry for the rest of us or get sued into oblivion for all of it. Yeah, I mean, you know, I think using AI to help with your workflow is one thing, but I mean, we're in the business of ads over here. So at the end of the day, a good ad is a good ad. It still requires that creativity. So yes, Andromeda wants volume and it wants all of this, but if you are creating an actually compelling ad that's still led by a human, maybe assisted by AI, that's what's going to perform. One of the things we've been testing real quick is the ability to take AI with CDANCE 2 and our creators who are real and making real videos can actually duplicate that content and switch it to unique. So it actually takes the sound, the video, et cetera, and just duplicates that person into a cloned video of what they originally created with a different background, different table, different angle, but exactly the same product, same look, same script, same whatever. It just flips a clone of it, which is really fun. That's cool. Cause yeah, that's like, you know, it's, we're starting, we're starting human, but then we're using AI to assist. And it just changes the scene and environment. Like we had a video that's done really well where she's using our chalk markers to go on the board and it's got this little squeaky sound cause that's kind of what the chalk markers make and it's the board on the table and she's drawing the whole thing. It did really well. The video popped off and went really great. I was like, well, I wonder if I can clone that. So I jumped in and I started building a skill to clone it using CDANCE 2 through an API and out came the end result. And I was just blown away. I was like, what? That is too real. Same thing, same very minimal errors. You wouldn't be able to tell the difference. And then I just asked it to start changing scenes and all of a sudden it's changing scenes in all these different locations. And it looks so real to the original one. I was blown away. I was like, wow, that is, yeah. Yeah. That's a cool use case. Some of the stuff that comes back with, you're like, wow. Yeah. Full personage, you know, full on a video, yapper videos and stuff like that. Some of those can get really well, but just the product demonstration videos in simplistic form, you know, and some of those don't even have audio and they're not showing the creator. They're just showing the outcome. They're showing the benefit. And those have actually been pretty powerful use cases. And they are legitimately the product and the video. And, you know, it doesn't look like we're trying to falsify anything by showing the actual product and usage. We can just change the background and identification and even change the hand that's doing the exact same thing. You know, change the person who's doing the exact same thing. We stopped to obviously put, you know, created by AI on it. But if I were to show you the video, you wouldn't know the difference. I know. You just wouldn't know the difference. I just read like some study about people who, you know, are affected by that or who don't care even seeing that AI thing. And I guess in like beauty is one where they do care, obviously. Lots of UGC videos come out on beauty. Just all over the place. Yeah, you can't fake it there. You know, once a fake actor is doing a before and after. So that one has like a negative impact. And then I guess like high fashion or like luxury items. For everything else, it appears people don't really care that much. But those are like the specific industries happens to be ones we work in where it appears that it does have more of a negative sentiment. Yeah, yeah. And it's just, it's the demographical type that's going after, right? And who's ultimately trying to look for it. Some of them have more discerning eyes. But as I have constantly reminded people and you'll understand what I'm about to say, this is the worst it's ever gonna get. So the next step becomes hyper realism and nobody knows what the heck is real. And it gets a little crazy out there. This is why I want everybody to be reminded what you're saying and what I will echo is there has to be a human interaction somewhere in this. There has to be authenticity. There needs to be experience. You can't just use AI for AI's sake. Using it as a multiplier of good things is promising, can add profit, add value and create better meta ads. But if it's just AI slop, it's garbage. It's garbage in, garbage out. You cannot 100% rely on it. Someone has to be in the operator's seat to oversee what's going on. Just don't let these things go automated. They are, they will not do as well as you think they do. Not right now, maybe. Exactly, maybe one day it will and I know it's only getting better, but you're right. I mean, I've been doing some analysis with Claude and having it dig into a client's account and their motion account, which is what we use for our creative analysis. And sometimes it spits back things and I'm like, hold on, those ads aren't even active. This isn't on. So you need to fact check it sometimes. And so it does. It'll even falsify revenue, believe it or not. If you don't, yeah, we had this happen last week and we went in and audited a skill that we were like, no, I don't think those numbers are right. And we looked at it again. It's like, well, it just added on 13,000 and that's extra revenue. Like, where did that come from? Nobody knows. Claude just made it up. Like, okay. Claude can hallucinate. Yeah, yeah. It's hallucinating. And it said numbers, right? Which you think, well, how would it not know one plus one is two and such and such? And how did it even come up with this? No one could know. All we knew was the calculations were wrong. Yeah. But we told it to train itself not to do that, right? So, okay. Like, we're still the intelligence here in some level. And I know that this is not yet what we all see is what we're gonna get because I know the next models are coming. Some of them are very crazy. For a moment, Fable 5 and others in Mythos, this thing, these things are super intelligent machines. They're dangerous in the hands of the wrong people in some cases, and they can be absolutely incredible when used for the right things. Yeah. But the only thing I'm trying not to do, and I know you'll appreciate this, I'm trying not to make myself obsolete. I think I have a lot of people who are literally operating under the idea of, I think in some ways we'll call it planned obsolescence. Only they're planning their own obsolescence and you're doing it in some capacity of freedom and choice and ideas that I don't have to work much or do a whole lot and I'll get all this output. Where in history has that ever worked? Yeah. We had someone, yeah, compare this like to the industrial revolution. It's like the AI revolution. It's just a, it's a different way. It is going to replace a lot of jobs. It is going to- Incredible things. It's going to restructure the way that we do things, but yeah, sometimes I'll be creatively writing or anything. And like, I do challenge myself not to use AI sometimes. I do sometimes too. Because you know, we still got to use our brains and- I got to get that muscle working, right? I got to try to keep up because it's changing so fast. I do agree. Yeah, you got to use that muscle and things are doing good. And sometimes it becomes a crutch that I personally step away from and come back to later when I've got my own thoughts in order. For sure. Yeah, that's an important lesson. Well, great stuff, Krista. Appreciate you. And thank you for coming on and just sharing some good word with us today and being a great guest. Appreciate it. Yeah, my pleasure. It's been great speaking with you as well. Where can everybody go find you? I'll put it in the show notes and everything else, but where should they go check out what you're doing online? Yeah, you can check out our website. It's www.miltmedia.com. That's M-I-L-K-E-D media.com. Awesome. Guys, check that out. And of course, go back and listen to this again. Take a campaign structure notes, creative. She handed you a pretty good blueprint today of the things you should do and should not do with your campaigns, especially if you're going to test out meta ads. She gave you a blueprint for the amount of budget you should assume and the amount of time you should assume. So don't go out there trying it for three days, spend a couple of hundred bucks and go, this doesn't work. I hate the whole thing. Give it time, give it patience, give it creatives and make sure that you're working on process. And don't just use AI as a crux for everything. Put some intelligence and experience in there and use that to your benefit while you're growing and building your brands. And of course, if you need additional help, hit up Krista and her team. She can help you guys out if you're going to grow and scale. Thank you for coming on the podcast. Thank you. All right, guys, like, comment, share, help us beat big tech as always. I don't care if you leave snide remarks in the comments, it just helps the algorithm. Share this with somebody who needs to hear a good word today who wants a blueprint. And as always, remember, you do something important. Well, what's my phrase? I almost lost it. Oh, crap. That's like my ending thing. I should know this from heart. Like, I totally blew this. Do something imperfect and something perfect will happen along the way. Stay high voltage and I'll see you on the next episode. Krista Carpen and I broke down why most brands are losing money on meta ads in 2026. The problem is not the platform. The problem is constant tinkering. You change one variable, then another, and you never let the data settle. We talked about how to stop the chaos. You need a system that lets you test one thing at a time. That is the only way to find what actually works. You do not need more e-commerce noise. You need clarity. Get the five e-commerce updates that matter. You get to see what changed and exactly what to do next. The Voltage e-commerce brief is free every Tuesday and Friday. It cuts through the fluff. It gives you the operator view. You can sign up right now. Go to voltagedm.com slash newsletter. This is High Voltage Business Builders. We help you build a business that runs without you. We will see you back here tomorrow. Until then, stay high voltage.