Lessons From $50M+ Brands Navigating Ecommerce Chaos

23:50Brian Roisentul:In other words, you had multiple entrants coming into the space to support them and a very little and narrow total addressable market. So what it became is slightly a race to the bottom on price. New competitor comes in, offers a comparable solution just for less, just for less, just for less. It's like this race to the bottom. Now, what I'm saying here is nothing really new because that's the environment we lived in in 21, 22. And we dealt with it, right? We adapted to it. But that was just chapter one. Chapter two became about AI producing software. So this is still pre-ChatGBT. ChatGBT still wasn't a thing at the consumer level, but AI very much was in building software from the ground up.

24:39Brian Roisentul:And so then all of a sudden you see this new cohort of software emergence that basically built equivalent solutions to the legacy providers in a fraction of the time. in like four months. These legacy providers that we all know in e-com, when it took them years and years and years to build, equivalents or even better alternatives were being built in three and four months because they were being built AI first, AI native, built with AI from the ground up. They didn't need a smart engineer. They had AI to be the engineer for them. And these tools are built with like four or five human beings, right?

25:18Brian Roisentul:That's it. Software companies going up against software companies that had three, 400 employees. And these startups were this new cord where they were not funded by venture capital. They didn't need to sell their product at an extreme premium to support venture capital initiatives. They could sell their solution for a lot less because it costs a lot less to stand it up because of AI. And so AI created this new cohort where, boom, here comes another leg lower for the entire software ecosystem that wasn't AI first, right? In my view, many software companies in the ecosystem are in a lot of trouble, right?

25:58Brian Roisentul:And so because of that, because ultimately it's a limited total addressable market and a race to the bottom on price. And new and better competitors are popping up every single day because of AI, right? That was just chapter two. Chapter three is now where things got dark for me, where it was like, what does agentic mean? What does agentic commerce mean? Right now it's kind of a buzzword. We know Shopify is investing heavily in it without a question, but what does agentic really mean? We are coming up with ideas, but we haven't really seen it manifest yet. The only thing we know though is that two years from today, the world is going to look a lot different in the way we shop, the way consumers browse, the way the consumers discover, the way they transact, the way they express their behavior.

26:50Brian Roisentul:Just like COVID completely permanently changed the behavior of consumers when it came to shopping online, we are going through another consumer shift that's taking place right now that we're only at the beginning of, and we have no idea what it's going to look like. In other words, when you think of a Gentic, every customer at some point, this could be six months from now, this could be 12 months from now, this could be 18 months from now, but every customer is going to be able to have their own customer journeys, depending on how they prefer to shop, what their preferences are, and a customized experience based on their behavior.

27:21Brian Roisentul:Whereas right now we say, hey, Brian, you like that product? You must go to www.com, go browse that product, add it to product page, go through checkout. We force every single customer down that same funnel right now, right? Whereas Agentec is opening it all up, right? And so when you think about it, the brand's website is no longer as valuable to the brand going forward. Like the website ultimately becomes like a showroom. And so if you think of it a showroom, like what if people are buying in chat? What if people are buying in other platforms? What if people are, what if we have this open AI necklace that's just living around our neck and listening to our every want and dream and buying things on our behalf?

28:05Brian Roisentul:Well, then what is the purpose of the website? the value of the website goes down, which means all the software that supports that website also gets devalued in that process, right? So this is the crazy part where I'm like, okay, where do I go, right? So ultimately, it goes into one to two buckets. If you're looking to shift your career and really trying to bet on the future of where a company is gonna be, some people are looking for a job and they say, hey, I'm looking for a job for the next year. Great, more power to them. That's not me though. I need to join a company that I believe in, that I'm going to believe in three to five years from now, right?

28:42Brian Roisentul:And that they have a genuine path to success and an exit. And so then I'm thinking, like, who are those players? It's a race to the bottom in software. Who are those players? So it only comes into two buckets. Either A, you are going to be AI, driving the AI in a gentic conversation, and you are at the forefront of it. Well, guess what? We don't have any of those players determined yet. Shopify is the only one investing heavily, but we don't have any of these startups that are actually like gaining the traction to say that's going to be the winner or that one's got early promise. We're too early there yet.

29:15Brian Roisentul:Right. But if you do pick the right one, wow, you like amazing. But that's like a that's a moonshot. The other bucket is. Let AI do its thing and it will, as a category, be fine and continue to be successful. An example of that is logistics. You still have to ship the box to your consumer's doorstep. Let AI do its thing. Logistics will only get better with AI, right? Or think of post-pilot. We are still always going to have a mailbox. The Internal Revenue Service still has to have a place to send their notices, right? It's like we will always have a mailbox. And so when you think about it, there is a defensibility and AI can come and make direct mail better.

30:01Brian Roisentul:But the means, the mechanism, there is no algorithm. There is nobody else's rules at play. There's no platform. There's no regulation around it. There's no AI that's going to come and take it away. And so that's where it's like, okay, that's a defensible category. And so then when you look at the defensible categories, geez, logistics, no way. I'm a MarTech guy. Like, sorry, I can't sell logistics all day long. However, when it came to PostPilot, it's like, what? Let's look at Meta. what's a brand willing to spend on meta if it's performing there's no limit there's literally no limit if it's performing what's a brand willing to spend on app love and if it's performing that can go up forever right and that's why i was so excited about postpilot is postpilot has not just a retention engine but an acquisition engine to do what meta for mail we are doing Meta for Mail.

31:01Brian Roisentul:Truly programmatic, AI-driven, cold prospecting, building lookalike audiences, and targeting net new customers through acquisition, right? So as we continue to develop Meta for Mail, what's the potential if it performs well over time? There's literally no limit. And that's what got me so excited about the app.

31:23Neal Goyal:Brands that haven't explored direct mail yet, what do they need to know that they don't currently know?

31:31Brian Roisentul:Yeah. You know, when you think about direct mail as a category, it's like direct mail has been around since the beginning of time, right? Like digital's new relative to direct mail, right? But direct mail has been around for hundreds of years, right? So when you think of direct mail, you think of it in its most traditional sense. Like you have a print shop, you send out some mailers, send out some catalogs and whatnot. So you're predisposed to think about it in a traditional sense. And it does okay, right? It does okay. It's always done okay. That's why the industry has continued to survive. There was this time, though, when then, you know, 2016, 17, 18, 19 hit and the rise of everything digital where it fell out of favor, right?

32:15It fell out of favor largely because of meta, where meta taught us click conversion, click conversion.

32:22Brian Roisentul:then all of a sudden around attribution. Like attribution is a much bigger thing in the last five years than it ever was before, right? And so now performance marketers of today, they have a different rigor and a different standard for measurement, for testing, for incrementality, for looking for opportunity, for evaluating a channel. It's a completely different rigor. And they couldn't do that with direct mail before because the ability to measure the incrementality, the actual conversion impact just wasn't there. So it fell out of favor because digital was like we're programmed for direct response and therefore direct mail couldn't do that.

33:00Brian Roisentul:And that's where Postpilot was born. We have merged the channel that is considered been around since the beginning, but then also merged it with the performance market marketers rigor where we can track conversions to the card, where we can actually measure in dashboard incrementality. where you can see attribution data live as your customers are receiving cards and doing so in an autopilot fashion where there are no minimums, there are no budgets, there's no lift. It's just an always on programmatic system that can just be firing on all cylinders every single day on your behalf in the background.

33:40Brian Roisentul:Right. And so that's where direct mail, I think, typically fits is like is direct mail has kind of gone through this cycle where it's really exciting right now, particularly post pilot. But then if a brand's never done it before, like where do they start? Where does it potentially fit? Think about email. Email has gotten harder as a channel as every single year has taken hold, right? Email performance has gone down over time, largely driven by Google. Now, so-called own channel is email is no longer now that Google has continued to implement more and more rules over time. Brands are going to spam, put in a promo folder.

34:20Brian Roisentul:But then more importantly, it's so saturated. Like you would go look and you look at someone's inbox, promotion, promotion, promotion. The email inbox has actually fundamentally been destroyed for the consumer, right? It's viewed as a to-do list for the consumer. It's not viewed as a place of enjoyment or entertainment, right? So the channel has become so incredibly noisy and it's not just email. It could be SMS. It could be other digital channels too. And so this is where every single brand has a huge batch of their audience that is no longer responding to any of those emails. That's now unsubscribed.

35:00Brian Roisentul:Maybe they've bought from the brand before and then unsubscribed and completely disengaged. Well, guess what? What have they done in their CRM? They've suppressed that audience in Klaviyo to save on cost because that audience hasn't engaged. They've gone dead on email. Well, guess what? That's the exact audience we want to go play with. And we want to go fire off a card to them, right? And so we have the ability to reinvigorate that customer. We do, where we start with our customers is typically on winbacks or lapsed buyers. Someone that has bought once in the past or twice in the past just hasn't bought in over a year.

35:37Brian Roisentul:And we print, no pun intended, print money in that strategy. Stage of growth should a brand consider starting incorporating it into their marketing mix? There is no bottom line threshold. Quite frankly, it's a channel diversification play. Typically, if there was a situation with when you think of traditional direct mail, you need to send out 10 ,000 cards minimum. Okay, yes. Now you need to view a certain brand size in order to make that investment worthwhile, right? Whereas because you can send literally five postcards if you want to. Right. But do so in a very personalized fashion, too. So, Brian, you go to a website, you click on a specific product page, you spend over 30 to 60 seconds on that product page, and then you leave.

36:26Brian Roisentul:You don't even add it to cart. You don't give them their email. You don't subscribe to a flow. We have the ability to de-anonymize your physical mailing address, fire off one single card to you based on your on-site behavior that you were looking at this product. here's an offer to convert. And we can send you a card, one single card that day. So there is no minimum. Everything can be truly programmatic on a per customer level, on a one-to-one level, based on your purchase history, based on your onsite behavior. And that's where it's like, there is no limit on brand size on where they should start, right?

37:06Neal Goyal:That's great. And what type of, let's say results, conversion rate, or however you measure it, should brands realistically expect from this channel? Because let's say, again, brands, some brands have never done this. They don't know what to expect. They don't, you know, people are, let's say, they're not, they need to see to trust a channel, right? And if they have never trusted that channel, what can they expect realistically?

37:32Brian Roisentul:Yeah, that's a great call. First, performance is often going to vary based on category, AOV, and other factors, just like it would impact conversion on site. But when you think about where a brand will typically start is in those win-back campaigns or trying to re-engage lapsed buyers. We're going to see a 3, 4, 5x, 7x ROAS on those campaigns. I mean, these are customers that have heard of you before because they bought from you before. They just haven't bought in a really long time, right? And so those perform extremely well. Now, ROAS on its face is not also the measure that matters most. What matters most is incrementality.

38:15Brian Roisentul:Can you demonstrate that this was truly incremental? And so everything on every campaign that you run, we actually measure incrementality by doing a holdout test. In other words, if in a particular segment, you have an audience of 20 ,000 people, We'll actually only send 5 ,000 people, hold out 15 ,000, and then be able to measure the direct mail impact of those that received a card versus those that didn't, right? And then we show that incrementality with holdout in dashboard on a per campaign and per creative basis on how that performs. So ultimately, no matter how big or small you are as a brand, incrementality is really the only thing that matters.

39:00Brian Roisentul:And if brands are truly trying to reduce or offset customer acquisition costs, the last thing they want to do is throw more money at a channel when they don't know if it's actually driving a greater lift had they not. And so that's the other standard. Performance marketer standard is measuring incrementality, proving impact, proving lift. And that ultimately will determine where things should go from a test. So from a conversion perspective and performance perspective, very strong on the win back side. And then we learn what the brand's targets are, target CPAs are, and create a customized acquisition campaign that allows them to achieve their targets or even beat them.

39:38Neal Goyal:So I'm going back for a minute to AI and the future and the present, but the future mostly. Yeah, of course, we don't know what's going to happen. It's crazy the times we're living. But there's one thing we do know, and I would love your take on this. it's a classic you know that tv didn't replace didn't replace radio online didn't replace offline what do you think will happen with websites mentioned briefly that you know of course we don't know right but if you had to guess please will you think that it will be only a percentage of people buying through agents and then most of us will keep buying from websites and websites are not going anywhere?

40:29Or do you feel that maybe not in 2026, but in the short term, let's say that will happen?

40:38Brian Roisentul:Yeah, that's a great question. Websites are not going away, right? Long term. They're not going away, but their use and purpose is going to dramatically evolve. right, when it comes to the end consumer. And so if we are evolving from a place where right now we're saying every customer has to go down the same journey, go to this website and check out, and that's changing into an agentic world, the value of the website to the brand completely, A, evolves and goes down to a degree. It becomes a showroom of sorts. It becomes a place for brand story becomes a place for creative and drawing connection with a very engaged and high intent customer consumer shopping behavior is going to be dramatically different so in my view i don't the website is not going away like the yellow pages once did however it is dramatically going to be a smaller piece of the overall brand strategy direct to consumer i think as a term means website, right?

41:47Brian Roisentul:Direct to consumer two years from today will mean a whole lot, something different. And website is one piece of that pie, right? Over the long run though, it truly is going to become a showroom. I think a few proof points that are going to show you that, that that's true is obviously the way customers are discovering products in ChatGPT. We haven't even gone down the path of customers actually adopting purchasing there yet, but it's only going to get better. It's only going to get better from here, right? And we have made chat and other LLMs a permanent part of the way we discover everything in our life now, right?

42:25Brian Roisentul:So you have to assume that there's going to be more transactions there. Proof point number two, ChatGPT just launched ads, right? You launch ads in the same place where you can also convert, right? that's a pretty incredible place right now we're seeing ads in facebook and yes we can convert in facebook we see ads on google and we can convert in google but that is completely different in the way we're using chat gpt so what is that if you notice the word website is not coming anywhere in this mix right now in this conversation right now and it's for those brands that ultimately for the customers that say hey okay i'm really interested in this product i discovered it in this unique way I want to look at it more.

43:10Brian Roisentul:I want to learn more. Or I am an existing customer. I've already purchased a few times. I want to find something that draws me closer to this brand, right? That's the purpose of the website as opposed to a place to convert, right? And that's going to forever change. That's going to impact the software ecosystem. That's also going to impact in the way brands have to adapt in what they're doing for their specific website as well. But gone are the days where a brand says, I need to go invest a million dollars on a website. Like, I think we've officially and they're still doing it on Shopify. They hiring, you know, top tier agencies to build these websites.

43:46Brian Roisentul:But the reality is the value of the website is going down over time. Right.

43:50Neal Goyal:One of the last things I would say is that I recently heard something about agentic commerce that changed my perspective. And I'm sure it's not only it's even the beginning, you know, just the beginning. And the commerce is not someone was saying, I don't remember the name. Hey, I ate these for me. I'm Neil. I have three boys. Family of five, right? And we, I wanted to do the groceries for me for the entire year. I don't want to ever run out of shampoo, whatever. People already know, you know, your favorite cereals, your boys, and everything. and will start shopping for you in our current basis.

44:40Neal Goyal:It will know that you like to buy your groceries on making it at Walmart or Whole Foods or whatever, and it will purchase for you. It will find the best deal and it will completely forget about that. And I'm sure that's only the beginning of what it will do for us. But it's not just buy this one thing for me. And it was, wow. It's really going to change things.

45:05Brian Roisentul:We as brands were like, we've been stressing for the last several years of like, how do we hold our consumers attention? Right? How do we hold it? And it's become harder over time, right? In what you're describing here, you actually never got their attention. In the first place, someone is acting on your behalf. the emotional part of the equation gets eradicated, right? And that's pretty wild to think about, right? Truly wild to think about that consumer behavior is largely driven on emotion. And if you have someone acting on your behalf that acts unemotionally, what does that mean? Wow, can't even wrap my head around that entire thought process, right?

45:55Brian Roisentul:A wild one to think about.

45:57Neal Goyal:Neil, it was always a pleasure to have you here. I'll keep going for hours, but I know we are on time. Any final prediction, if you have to make one prediction, it could be something that you've already said, and you can say that too. For when we meet, hopefully for the first episode next January, we'll check on this and say, was it a reality or not? What would you say?

46:23Brian Roisentul:Now, more than ever, with the rise of digital and how noisy everything digital has become, and with this new world of agentic and whatever that's going to mean someday, the concept of having a human personal touch is going to stand out more than it ever has. that in real life moment is going to stand out more than ever. Brands who invest in developing a personal touch with their customer outside of a digital channel through in real life experiences are ultimately going to be able to cut through all of this noise. And it's harder than ever to keep their engagement, to keep a consumer's attention, to keep a consumer's love.

47:06Brian Roisentul:But one way that's going to prove to be an extremely effective strategy in doing so is just developing a close human connection with them in some form or fashion.

47:16Neal Goyal:Thank you again for being on the show. I again, admire you, follow you. So it was a true pleasure to have you here once again.

47:24Brian Roisentul:Hey, Brian, it's been an absolute pleasure as well. Obviously, I've followed your journey for a really long time and I love catching up with you. You're a legend and thanks so much for having me.

47:38Brian Roisentul:This episode was brought to you by BSR Digital. We help DTC brands grow through paid ads and email marketing campaigns. If you'd like us to help your business grow, head on over to bsrdigital.com and schedule a call with us.