AI-Powered Media Buying: How Mark Cuban and Brecker Brees, Founder of Preflect Are Disrupting DTC

Tim Rowe:
If you search for AI in marketing on Google, you'll get back 11.4 million results. It's safe to say that there are no shortage of opinions on how exactly AI is influencing the evolution of marketing and advertising, but how many of those authors are actually in the pit doing the thing? In this case, the thing in question is AI in marketing, specifically for this conversation, AI in media buying. Why is it a problem worth solving? How are some of the fastest growing brands using AI and their media buying to arbitrage insights into impact and unlock new efficiencies as they scale? And what does the success look like with this new playbook? To unpack that conversation for us is Brecker Brees, founder of Preflex, a platform set on replacing media buying agencies with the help of Mark Cuban and AI. Brecker, thanks for being here.

Brecker Brees: Awesome. Thank you Tim for having me.

Tim Rowe: Absolutely. I'm excited to talk about AI in the media buying application. Maybe a place for us to start is where the idea for Preflect come from and ultimately the Mark Cuban story. I think you've got to tell us the Mark Cuban story at some point. It's a great part of the origin story, but where'd the idea come from?

Brecker Brees: Yeah, that's a great question. So, um, it all started back in early 2019. So I was actually approached by a buddy of mine that I went to high school with. So we were both in college at the time. He came to me, he said, Hey, I, I started this company. We're very early stage. It's just me. I'm pre-product, right? Everything's kind of in the ideation stage. I'm pre-product. I need good people to join me. But not necessarily in a position to go out and hire top talent. And so his options were just picking from a pool of people that he already knew. And we were both really interested in economics. We were studying economics in college. and interested in a lot of the research that was going on and a lot of the software products that were coming out based on modern 21st century experimental economics and things like that. And so Ian's concept was creating pricing software for Shopify businesses. And at that time, we weren't really focused on the S&B space. We were just more focused on mid-market to up-market Shopify brands and having pricing optimization software where we ingested a bunch of data and we were able to make optimal pricing recommendations for these large Shopify brands that had really, really high impact on profitability and sales and demand and things like that. And so, uh, Ian, um, Ian McHugh, he actually had coauthored a book with Mark Cuban when he was a teenager. Uh, and that's, that's a whole nother story, a long story. Yeah. Yeah. So, um, he, he had worked with Mark and the book did really, really well. Um, and so. At that point, he was not super close with Mark, but friendly. A couple years ago, they had made some money, wrote this book, got some really good PR out of it. And so he emails Mark and says, hey, I want to get into an incubator, some sort of program like that in LA, Chicago, New York. Would you mind writing me a letter of rec? And so Mark basically said, forget the incubator. I'd love to make a deal. And so we made a deal with Mark. The rest is history. We actually did, we did okay with the pricing software. The problem was the monetization, right? It's not really something you need on a recurring basis. You need it at the most once a quarter, but at the time we were working with mid-market brands. And so mid-market brands needed it, not even once a quarter, they needed it twice a year, realistically. And so getting recurring revenue and getting cashflow in, was an issue. That was always a hurdle. And so we started to create other features and other products that were in that platform. And so we started playing with using all that data we were ingesting for our pricing model. And one of those things we were doing, we were predicting customer lifetime value. And we realized that predicting customer lifetime value based on that historical Shopify data was great for building Facebook audiences. And so this was about a year into building that first product, and that feature really took off. So we started getting some referral business and doing some deals where people just wanted that. They were like, the pricing stuff is cool, but I'm more interested in these audiences, that these custom audiences that your model is creating. And we paused for a second, we were like, This could be, we might be onto something here, right? There's way more demand for this one feature than the entire platform. And so we basically packaged that as its own product. That did really well. And so we kind of sat down and said, okay, we're going to rebrand. We're going to relaunch and we are going to turn this feature into its own product. And so we were very fortunate that Mark and a couple other investors of ours had a lot of confidence in us personally. And really, we're able to sort of see the light in the traction that we were gaining from that feature, you know, in the old platform, and we were able to raise a little more money. And so that last round of financing is really what paid for the Preflect platform. And so it started as just a feature and our old product and it evolved into its own platform. And now we're here today, several versions of Preflect later and haven't looked back. So I still am so interested in pricing and I loved that product. But just as a business, this has turned out to be a lot better.

Tim Rowe: That's it. When, when, when you find a problem that people say, Hey, we'll pay you to solve this one, optimize in that direction. That's a pretty good lesson there. You started this in 2019 with the, with the pricing, then it's evolved into an AI enabled media buying platform. How did iOS 14.5 impact your visibility? Was that, was that an accelerant for the platform or did that set you back in any way? So.

Brecker Brees: To be honest, I don't even know if we would be here if that didn't happen. No kidding. Because we were able to, I think there was a lot of media buyers that were really relying on the targeting that was inside of Facebook. Okay. I think there were a ton of agencies that were, you know, they got really good at picking interests. And they got really good at using everything that was inside of Facebook to target customers. And while we do things a little differently than we did back then in terms of the ad buying side, that was really right. Those, they were lifetime value-based audiences. Facebook, hold it, it'll come to me. I can't think of the technical Facebook term off the top of my head, but value-based lookalikes. That's what Facebook calls them, is value-based lookalikes. So at that time, with that being kind of our crutch, that was working really, really well right at that time. And so even though there was a lot of data asymmetry, and obviously we weren't a data platform at that time, we weren't a attribution solution at that time, just in terms of pure results from a result standpoint, we were performing so much better than our competition, which at that time really wasn't other software competitors. It was boutique, small ad buying agencies in the Shopify space. That enabled us to grow quickly even as the new kid on the block and doing things way different than the traditional agency model or the traditional software model. So to be honest, I don't know if we'd be here. I think the fact that everyone else was doing so poorly kind of gave us that opportunity. And we had a lot of people that just said, you know what, I'm willing to try anything right now. And so it was a really good time to acquire customers because you had a lot of people that were just so disgruntled with what was going on and they sort of saw their Facebook funnels just fall apart, that they were so willing to try new things. And I think that was a really good thing for us.

Tim Rowe: Oh, that's an interesting time period to look back on as we get into this cookie list future. We'll we'll see if we can apply some of those same lessons you identified, or I guess you were identified as a natural competitor to those boutique agencies at that time. But you've identified this pocket of brands spending between three and $90,000 that have a specific problem set that you're able to solve. What is that challenge that is similar for brands spending in that range?

Brecker Brees: Yeah, the challenge for those brands, and again, I'd like to preface this as I'm definitely not knocking the agency model and the people that are in that line of work. There's very good people, very hardworking people in that line of work. But the problem is for some of their customers, if you just look at it from a sheer financial perspective, it doesn't make a lot of sense. So if you're a small Shopify brand and you're doing six, seven, 800,000, a million dollars a year in sales. And you're trying to grow, right? You don't want to be a million dollar year brand forever. You want to be a 5 million, 8 million, 10 million dollar year brand. You have two options, right? You either need to have solid unit economics in your marketing that's generating not just cashflow, but also profit that you can reinvest into growth. So if you're getting a really good return on ad spend, there's money left over at the end of the month, month over month, quarter over quarter, year over year, you're going to continually invest more. You have to spend money to make money. And so you have to spend money on ads if you want to make more money on ads. Gets to a certain point where you can only optimize so much, you have to start scaling. And so when you scale your business, you're gonna have to scale your marketing. And so the issue is to get a decent boutique agency, they're paying, you know, back then it was $1,500, $2,000, $2,500. Now, a couple years later, you're talking $3,000, $3,500, $4,000, $5,000, $6,000 a month. So think about that for a second. If you're a brand that your total sales, maybe right now it's a startup brand that's doing $50,000, $60,000 a month. Say that a huge majority of that's coming from paid social, maybe in the range of $30,000, $40,000 a month. If you're looking just on what's a realistic amount of profit that you're making, once you pay for the ads, once you pay for your inventory and all your other expenses and all the other softwares that you have, maybe you have one full-time employee that's also working for you on this brand, that $4,000, $5,000 a month agency retainer just eats up all the profit. Which is sometimes 100% of it. And that's the growth engine. That's the flywheel. That's what you take and continually reinvest it and just grow, grow, grow, grow and scale. Now that worked. Um, that works really well when times are good and performance is really strong. Okay. And if these agencies can actually get you great results, iOS 14 hits everyone's performance tanks, and now the flywheel is broken. So a couple of years down the line, now, like I said, there's two options. Option two is borrowing the money or raising the money, which for most merchants, especially small merchants, they're doing some sort of financing against their future cashflow. And so we're in an environment now where borrowing money, cost of capital is really, really high. And the e-commerce space, say that you've just been in the Shopify ecosystem for 10 years. You've never seen cost of capital as high as it is now, ever. on the 10-year scale. It's so expensive to borrow money. It's so expensive to do merchant cash advances. I mean, even if you're using Shopify capital, 8fig, there's a lot of different options. But right now, cost of capital is a lot. And so a lot of merchants are looking at that saying, well, that's not a great option. The cost of borrowing is super high right now. So that's not going to keep the flywheel going. My agency retainer is eating up all my profit at the end of the month. That's not going to keep the flywheel going. And right now it only makes sense to have one employee or zero, or I'm a solopreneur, or it's someone in their partner running the business, or someone in their brother or their sister or something like that. There's not funds left over at the end of the month to pay that agency or to take care of the interest on that financing. So they're looking at themselves thinking, I can't do this myself. I can't hire someone to do it. I can't contract an agency. I can't borrow the money right now. So I need a solution that works well and is also cost effective and is not going to drain a lot of my time. And that's where we come in.

Tim Rowe: And that's right. I mean, you just painted a scenario that I'm sure a lot of folks can relate to. And obviously, you're working with a lot of them. So we kind of understand the scenario of, hey, this is what happens if we don't solve it, is the profit gets sucked out. We never really grow the business. What happens when we do figure these things out? It seems like maybe that's kind of the, is that like the threshold to break through? Like to the next level, now you're scaling the business and now maybe it does start to make sense to work with agencies. And we're really, now we've gotten to that next tier. Is that kind of the, is, is that, is that the tier that I'm trying to get through into?

Brecker Brees: Yeah. And so I guess what you're saying is, are there people who graduate from us and get to a level where they need that?

Tim Rowe: Perfect.

Brecker Brees: Yes, exactly. And you know what? There definitely is. And that's why we say, and it's even on our website, we work with merchants that spend between three and 90k a month. And we've sort of found that when you start to spend north of a million dollars a year on your paid social ads, meta, you know, specifically, once you're spending more than a million a year, you're at the point where you need to start doing some more, some more advanced things. I don't really like that word. The diversification of your media buying becomes a lot more important. And so you're going to start, you're going to start doing a lot more testing and the attention to detail With that testing is going to require a team of people.

Tim Rowe: It's a great way to, yeah.

Brecker Brees: And so maybe one day we'll get into that market, but right now we're trying to keep our offer really affordable. And so, you know, could we come out with an offering like that? Yeah. Would it be at the price point that, that it is now? No. Right. So, so that's the thing. And, and, and the thing is it doesn't, so the cost of the agency. But in most cases, I know in some cases it does scale linearly and it shouldn't. It should not be linear. I've talked to some really large brands that have gotten themselves stuck in some agency contracts where it did scale linearly and they tell me how much they're paying their agency. And I'm just kind of thinking, wow, that's a lot.

Tim Rowe: But you're describing it as like a percentage of spend or a percentage of revenues. There's some element where as the company's growing that profit, there's a percentage that's being sucked away.

Brecker Brees: It's, it's, it's not even, it's not even that it's just like what, what they're getting for the money.

Tim Rowe: Maybe describe that. Describe that. Like what, yeah. What, what is that? Like what is an undervalue?

Brecker Brees: So I'll give you an example. I'll give you an example. And this example is important too, because it also makes it clear the space that we're in and why we're in the space that we're in. So an agency with a retainer, that's four or five grand a month that's working with our customers. Okay, that's a massive, there's sort of a floor of like the minimum that you're gonna have to pay actual humans to be inside your ad account every single day. And so for a highly qualified agency, that's around the floor. Maybe the floor is a little lower than that. Now, if you're spending 100, 200, 300 grand a month on ads, it should not be, I guess to make the example easier, if you have someone with us spending 10K a month, and they used to pay their agency five grand a month. And a similar merchant is spending 10 times more, right? They're spending, instead of 10K, they're spending 100K per month. They should not be paying their agency 10 times more. They can find really good people in a really good, big name, professional agency, with great references, to run their marketing for a lot less than $50,000 a month. You know what I'm saying? So it doesn't scale linearly, and so you get to the point where it's in the budget, very reasonably, to pay for an agency. But when you're in that 3K to 90K per month spend range, that's sort of the pocket where that extra, those extra funds, they need to go into growth.

Tim Rowe: You cannot afford the actual advertising.

Brecker Brees: And once you break through that, right, because the cost of contracting agency doesn't necessarily scale linearly, then it starts to make sense to contract out an agency, like just from just from a finance standpoint. And I think, you know, even on the larger end, I think I see a lot of agencies right now doing the same amount of work Uh, for someone who's spending a quarter million a month versus a brand that's spending, uh, 2 million, $3 million a month. And like I said, those are the instances where I see it scaling linearly. Yeah. Um, and. I think it's a pain point that a lot of people feel. But we will get into that space one day. We will tackle that space one day. But I think the immediate need in the Shopify ecosystem right now is figuring out how can you get these small merchants to not be small anymore. And I think that should be the number one goal of Shopify right now. I've said this a lot. I'll say it again. I think Shopify's number one priority right now is they make a majority of their money off of a handful of gigantic companies. If they want to be top dog in this space forever, not just sort of in this cycle that we're in right now, but If they want to be the gold standard for the next hundred years in this space, they need to figure out how to take these million dollar a year stores and turn them into a hundred million dollar a year stores. And a big part of that is freeing up capital and making it easier to reinvest in that flywheel and grow. And that's, that's our goal. And I, I, you know, I think that should be Shopify's number one goal too, because then they're, they're going to grow their customer base. exponentially, if they can solve that. And as much as people want to say it's not, the number one driving force behind that is paid social. You know, I talked to SMS guys, and email guys. And I mean, you know, even though you're big and out of home, it's still paid marketing, you know, but it's I go back and forth with the email guys and the SNS guys about this all day long. And my question always is, okay, you're driving traffic to the site and you have killer pop-ups and you get a bunch of SMS signups. Where did the traffic come from? It didn't come out of space. It came from somewhere. The traffic came from somewhere. Someone saw a billboard, they saw a TV ad, they saw an ad on Instagram, they saw an ad on Facebook, they saw an ad somewhere. Right? Or someone told them about the brand and they searched it on Google, but their friend that told them about the brand, where did they probably hear about it from? Yes. Instagram, Facebook, they saw a billboard, they saw a TV ad, they saw some sort of paid media. And so that is, that's sort of, you know, the lifeblood of that, of that flywheel.

Tim Rowe: How does maybe talk about the AI elements, the automation, the optimizations, how is the, the AI. In preflect functioning, I guess, without giving away the secret sauce, but how does the AI enable all of this to come together?

Brecker Brees: Yeah, that's a great question. So, um, I mean, that, that really is our secret sauce and that's also how we, you know, from a business end, keep our, our labor costs. Um, you know, very low compared to how many accounts we service because we truly don't, we have a lot of competitors where, you know, Oh, we'll run your ads for X amount of dollars. And it's really cheap and it's really not software. It's someone in some faraway country that is honestly, in my opinion, getting exploited and, and paid a very low wage, um, to do. work physically inside of your ad account. And so the reason why we have such a lean team for the customer base that we have is because 95% plus of the actual ad buying and ad creation and ad deployment and setup is done by software. Over 95% of it. So we don't need a huge team with a bunch of people pulling a bunch of levers all day long. So In terms of the actual automation side of it, everything is automated. So our users are uploading their ad creatives to our platform. We're scanning your site and looking at the ad you uploaded to write ad copy. When you deploy those ads, our system is structuring those ads into different ad sets, into different campaigns, deciding how much retargeting it's gonna do, how much remarketing, how much prospecting, where budget is going to go. So that's another thing. So we're Facebook, Instagram, and Google. You set one budget. You don't set a budget for Facebook and a budget for Instagram or just a budget for meta and a separate budget for Google. It's just one budget for all of your marketing. So our system is also taking control of your budget allocation strategy and sort of playing around with that over time. So another thing you don't have to worry about, right? That's a major task when, especially when you're trying to figure out how can I grow? Where are the profit centers? How am I going to generate cash flow that I can reinvest into growth? Um, you're constantly having to think of where are you putting your money? You just need to think about what you're going to budget on a daily basis. And then our system automates the rest for you.

Tim Rowe: What is, I guess, and how do you, how do you instill that confidence into the brands that, Hey, the, the decisioning behind the AI is aligned with your goals and objectives?

Brecker Brees: That's a great question. So, and that's one thing where I'd say we take a lot of control over that. And that's one thing some people don't like. Some people want to have a lot of granular control. And if you want to have a lot of granular control, I always suggest teach yourself how to run your own ads, right? You're going to have to make the time that you're going to do it yourself. We're very, very prospecting heavy. And so our goal, And again, this is one thing that's different from an agency. An agency wants to find a brand that has a really good list in their Shopify account that's not getting tapped, and then blast them with SMS and email and a ton of retargeting. And it's going to look like you're getting a super good return, maybe for a couple of months, and then you've burned the list and then everything just tanks. That's not our strategy. Our strategy is to have a really complete funnel and a really good foundation that's built on prospecting. We want to help you grow and to grow, you have to find new customers. That's the only way you grow, right? You can only sort of milk the cow so much with your existing customer base to a certain extent. And there comes a point where to grow, you have to prospect, you have to acquire a ton of new customers. You've never seen your brand before and are trying your product for the first time. You have to grow that base. And so we're very, very prospecting focused. And like I said, some people want more granular control. They don't like that. They want to be more heavy in retargeting or this, that, or the other. That's another thing. Our system takes control of that funnel and that allocation strategy. And it does what it thinks is best for you. through the lens of growth and to us, what growth means is finding more new customers. Not just doing a bunch of retargeting and having short term gains, but building a really good foundation.

Tim Rowe: I think that that would align to most people's, uh, idea of what growth is, is continuing to find new customers. How do you ultimately solve for like big platform changes? Like performance max. Or some privacy changes, how do you, how do you accommodate or solve for those things in real time?

Brecker Brees: Yeah, it's a great question. And, um, I can also touch on attribution a little bit here. So, um, about five months ago, we actually completely overhauled how our own internal attribution system works. So, and I think you said this earlier, you know, we're about to live in a post cookies world. There's a lot of things that are going away. that I think a lot of people, a lot of marketers, a lot of softwares relied on for a very long time. And so everything that we do, we try and focus on stability. So we want to invest in new things and new technologies and new ways of doing things that sometimes are not the best way right now, but they're going to have a lot of consistency and be less error prone over the long term. We don't want to do something and come out with some product that relies on extracting some piece of data or some piece of information that there's a good chance next quarter it's going to be a crime to collaborate. Because we don't want to get people used to doing something a certain way that we know with a high degree of certainty is going to very, very shortly go away. We want to find technologies and ways of doing things that people can use for years and that they can rely on and they can depend on. Because at the end of the day, yes, small businesses want to do innovative things, but they also want stability. Our customers, they're really hardworking people that, you know, have small teams or they're solopreneurs and they don't have time to worry about all these changes and all these things, you know, this is going away and it's getting replaced by this. And they don't wanna be bothered with any of that. They wanna use something that they know is gonna consistently work over the longterm. And so, one thing that changed with our attribution system I don't want to say it got less sophisticated, but we're not using certain data points that we're unable to collect anymore and that we know very soon are illegal to collect. We're trying to future-proof our attribution system. I'm sorry, what was the original question? Because there's a couple other things I wanted to get into.

Tim Rowe: Specifically, you kind of nailed some of it, how to solve for the platform changes. And then I think we're going to segue right into the attribution, how that's evolved.

Brecker Brees: Yeah. And so there's a couple other big platform changes on Facebook that we've had to deal with internally. We just had a major issue with the way that Meta processes ad text with emojis.

Tim Rowe: Say that one. Sorry, just broke up.

Brecker Brees: The way that Meta processes emojis in ad text. Oh, wow. Certain emojis wouldn't deliver in certain ad placements. there's always all these things that come up that our users don't even have to worry about, right? We take care of it for them. If there's some big change with how product catalogs work or some big change with they're taking away interests or adding interests or doing different things with targeting by age, geographic, there's all these different things that, you know, through the years Meta's changing. Sometimes it's giving new things, sometimes it's taking away. And it's not always their fault, right? Sometimes there's technical limitations where they gave something that they probably shouldn't have and they have to take it away. Sometimes there's regulatory changes that they have to adhere to. And so it's not always their fault, but the past several years, especially, it's been very dynamic. There's all these new things changing, this works differently. And our users don't have to worry about any of that. Didn't even feel the change. No, no, it's completely seamless. They don't have to be bothered by it. They don't, I bet you half of them aren't even privy to any of that kind of stuff because they don't need to be. We worry about all of that for them.

Tim Rowe: That's peace of mind. That's, that's that time back. The reason why we're doing the thing that we started out to do, right? It's the reason why they started the brand was to work on the company, not to be a slave to the ads manager. And we say that all the time.

Brecker Brees: If you sell golf polos or you sell snowboards, um, you started your brand to sell snowboards, right? You didn't start your brand because. you think it's gonna be so much fun to sit inside a Facebook ads manager every single day and that's gonna be great. Nobody thinks like that. Nobody starts a brand to do that. And so that ultimately is our main goal. That's what we're trying to do, is find people that have a great brand, they have a great product. It's inevitable that they're gonna be big one day. But it's what happens in between now and when that day comes. And so instead of them doing all these things that they have to do, we want them to do the things that they love doing and that they want to do, and we take care of the ads.

Tim Rowe: I can't think of a better exclamation point. Are you a reader? Are you a podcast listener? Where do you go for educational, motivational? How do you stay in the know? Give us a little personal on Brecker Breeze. What are you tuned into these days?

Brecker Brees: Yeah, so it might be kind of embarrassing to say, but the two main sources of information for me are Reddit and LinkedIn. Oh, I like that.

Tim Rowe: You're the first to use Reddit as a most common resource. What do you love about Reddit? Give us the 411. Why do you love Reddit?

Brecker Brees: I like it because I think that you can ingest a lot more information very quickly. I don't have a lot of free time. So to be honest, like for me, sitting down and watching the news is That's not even, that's like, yeah, that's just like, not an option for me. I'm just, I'm just not going to do that. And so I, I don't know, I try and limit my, I think some people call it like doom scrolling. Oh, yeah. On like Instagram and stuff like that. And I think that Reddit's format kind of translates to the way people interact with mobile social media and how there's a feed that you scroll through. I think that's a really easy and quick way to consume information. And so for me, when I'm reading anything about business and technology, I feel like I'm able to consume an insane amount of information very, very quickly with just the format that Reddit offers and also the way it's organized, right? There's a subreddit for everything, especially in the marketing and technology world. People are having conversations about marketing and technology. All day long, all day long. And it's just so quick just to see what people are talking about. And so I also like to see a combination of what's organically getting generated. If there's some new Facebook bug, of course it hits Twitter and hits LinkedIn, but where you find it first, on Reddit. That's where you're going to hear about it. You're not going to find it in some news article. You're going to have some thread and some subreddit that relates to Shopify or Facebook ads or digital marketing, something like that, where there's going to be a bunch of people that two days before it made news headlines, they were complaining about it saying, Hey, are you having this problem? Yeah, I'm having this problem. Oh, I'm also having this problem. And then people talk about it. Stuff like that, yeah, I think Reddit is honestly, it can't be beat.

Tim Rowe: Man, that is a pro tip. I think there's going to be a lot of folks jumping onto Reddit after this conversation. Give us the lat and long. Where do we get in touch? Where do we follow you? Where are you most active on socials? How do we find you on Preflex?

Brecker Brees: Yes. The best way to reach out to us would be to go on our site and actually book a demo. So Preflect.com. Preflect.com and preflectads.com. We own both those domains. It goes to the same place. But yeah, our website's very simple, like our product. We don't like to bore people with a bunch of information, but it's all over the place on the site. Bunch of forms where you can fill out very quick form. We don't ask for a ton of information. And then from there you can book a demo. Beautiful.

Tim Rowe: We'll make sure to link to all of that below. Brecker, thanks again for being here. Thank you, Tim. Absolutely. If you found this episode to be helpful, please share it with someone who could benefit. As always, make sure to smash that subscribe button and wherever you're listening, leave the podcast review. That's how you help us grow and we'll see y'all next time.