Episode 173 ยท Season 7 ยท ยท
The Agentic Surface
John Searby, Chief Strategy Officer at HUMAN, on why over half of all web traffic is now automated, and why the brands that win will see every agent and still reach the human guidingย it.
John Searby, Chief Strategy Officer, HUMAN. In conversation with Justin Cooke.
The Agentic Surface
0:0040:00From the cyber intelligence world into the centre of commerce, John Searby now sits where advertising, fraud and agentic AI collide. He argues that over half of all web traffic is already automated and no longer human, and that the brands that win will be the ones that can see every agent and still reach the human guiding it.
Searby is Chief Strategy Officer at HUMAN, the security business that sits in traffic on advertisers' and merchants' websites and across the advertising ecosystem, protecting the consumer journey from bots and invalid traffic. He came from cybersecurity, specifically the cyber intelligence world, where the job was to gather the most data possible on what bad actors were doing and to layer defences and controls so no incident or loss slipped through. By his account HUMAN has spent 14 years detecting automation, and the company has now carried those same detection tools into a world of consumer-side agents, crawls and real-time retrieval. More recently it moved into broader, data-driven ad verification, with Searby pointing to a partnership with OpenX announced that week.
In this conversation with host John Horsley, Searby argues that the internet was built for humans but is now being walked by machines, with over half of customer traffic automated and agent traffic up over 7,000% since tracking began. He says the first job is visibility, then controls, shaping which agents can browse, shop or check out on a case-by-case basis. The deeper problem is identity: verifying who the human behind the agent is, and building trust on both sides of the transaction. He is an unapologetic AI optimist, framing the LLM as the best surface we have ever had to reach a buyer, while warning that fraudsters were handed the same powerful tool. See it first, then shape it.
Size up that 50% forย me
“We saw it over 7,000% from when we started tracking last year.”
The internet was built for humans. For years we had automation and an implicit deal with Google: scrape my site and I get human traffic. AI skyrocketed that, first crawling to feed the LLM, then real-time retrieval, and now agents acting like humans. The launch of Muse was a critical inflection point. Agentic traffic grew over 7,000% since we started tracking last year, and it has skyrocketed in the last few weeks. Right now Muse is about 70% of all agentic traffic.
Where are we on theย journey?
“All of a sudden he's a new audience that wasn't traditionally”
Early, and we all are. This is the fastest moving news cycle and innovation cycle. It is unreasonable to expect anyone has this nailed, and anyone who tells you otherwise is not paying enough attention. Uniquely it is changing at the consumer level first. My father is 80 and never figured out a computer. Now he talks to his LLM every day and can give it agency to shop and do things he never managed. All of a sudden he is a new audience that would not traditionally have been on most websites.
Where is fraud in this conversation?
“It became a very low entry point for them to spin up a fake merchant”
Everyone is trying to solve a similar problem. Fraud has always been a cat and mouse game. We just handed consumers an incredibly powerful tool, but we also handed the fraudsters. It became a very low entry point to spin up a fake merchant and a legitimate-looking website that says it will ship you shoes that never arrive and disappears overnight. We also lost visibility and signal as the old customer journey was compressed and replaced. Fraud is catching up, and we work on linking traffic, device, identity and payment so we can identify it. It comes back to collaboration and integration.
Put yourself forward or recommend a guest for us to consider.
Request an interview โMost companies try to reach senior decision makers. We give them a reason to engage.
Partner with Us โHave your say.Add your voice to the conversation.
Voices is where senior practitioners share what they're seeing, what they believe and what they'd change.
Become a Voice โHow does this change the supply of places toย advertise?
“It's not only a new surface, but it's the best surface we've ever had, right?”
Attention is moving, the same way Google Search changed it and the walled gardens came. This is the new search, and it will have the best data on anyone's interest. It is not only a new surface, it is the best surface we have ever had to gain attention, so dollars will flow that way. But you still need a good product and something people want in the end, because the human comes back into the loop to buy. You will still see retail and commerce media networks and open web advertising, just reached by a slightly different path.
Your buyer was the CISO, now it's closer to the whole C-suite. How do you square risk andย growth?
“I'm not the solution. I am a piece of this puzzle.”
We get much closer to our customers' workflows and integrate where people work, in Adobe and Google Analytics and with the DSPs and SSPs where the full signal is. We have worked with the CISO and fraud team, increasingly with the CMO, analytics, product and commerce teams. I am not the solution, I am a piece of this puzzle. I give the data and controls where they operate, redirect the traffic, work with providers on the policies, and act as an integrated control plane. In fraud or marketing it is the same conversation: let us actually integrate, because we ignored each other for too long in the tech stack.
Never miss an episode. Follow Business of Marketing.
Read the full transcript
Lightly edited for readability
Speaker 1: I'm delighted to welcome to this episode John Searby, who's the chief strategy officer of HUMAN. Welcome.
Speaker 2: Thanks for having me. Yeah, appreciate it.
Speaker 1: So, um, why don't we kinda kick off? You run strategy. Um, why don't you tell me a little bit about the kind of the role and about the business?
Speaker 2: Okay. Um, so HUMAN is, uh, as a business, we are in the business of protecting the consumer journey, the customer journey. We're doing that across your advertising ecosystem, ensuring that real human eyes are being served ads. Uh, increasingly, we've for many years, we've protected that from invalid traffic, making sure, again, we're cleaning that supply, working with a variety of platform partners. Uh, and obviously, as the world is continuing to change here, and we're all handed autonomous reasoning robots in our pocket, how is, uh, AI affecting that has become a critical part of our business. But we also are, uh, sitting across many of your advertisers, merchants, et cetera, and we sit on their, in line on their website. We're, we're sitting in traffic. We're protecting from bots traditionally. How are they interacting with their, their web assets? Are they performing fraud in some way, trying to do some form of automation that is undesirable? Uh, and so we, we've, you know, hundreds of customers across that, that we've been protecting for some time. So seeing that journey from the moment an ad is served to the traffic that comes to the site, to the login, to the shopping, to the reading, the, the subscription, et cetera. Uh, and so we're sitting very much in the center of commerce, and as I said, the world of, uh, agents has come, and we've seen a massive uptick in, in traffic. Uh, you know, over, over 50% of all the traffic our customer sees now is, is automated and no longer human, uh, in some nature.
Speaker 1: That's incredible. And, and just, just size that up for me just so that sort of someone listening to this, watching this kinda get, gets a sense of, like, 50%. What does that mean?
Speaker 2: Yeah. So, you know, traditionally, again, the internet was, was built for humans to interact. We advertise to humans. H- sites are built for humans to, to come and shop and click, et cetera. Uh, and, and for many years we've had automation, various forms. We've had search bots coming and crawling. We had an implicit agreement with the Googles of the world, "Hey, you can scrape my site, in return, I get human traffic." Uh, and what we've seen over the years is automation increasing in a variety of ways, and AI really has just skyrocketed that. Now it's about how do you crawl to get to feed the LLM, but also how is it doing real-time retrieval? When you're asking your LLM about a product, et cetera, it's live searching the site. And now the, increasingly we're seeing the agents come. They're actually performing and acting like humans. And, you know, the launch of Muse has been a critical kind of inflection point where as we saw this coming, we saw agentic traffic on sites. Uh, it was, it was real. It was, it was a real thing. We saw it increasing. We saw it over 7,000% from when we started tracking last year. We saw traffic increasing of, of agents on sites, and that's literally skyrocketed in just the last few weeks as Muse has become widely available. Uh-
Speaker 1: Wow. 7,000%?
Speaker 2: Yeah. Yeah. It's, uh, right now, Muse is about 70% or so of all agentic traffic, of the agents that are on sites. We a- again, saw it overnight that, uh, the traffic became Muse and coming and shopping and trying to do many things, right? Um-
Speaker 1: You must be very busy.
Speaker 2: It is, it is an exciting time. It's a very exciting time.
Speaker 1: So you're spending your days, and maybe your nights as well, telling good machines from bad ones.
Speaker 2: Yeah.
Speaker 1: Um, how do you make the call now that, you know, there are allegedly good, good, good actors kind of-
Speaker 2: Right
Speaker 1: ... you know, growing at sort of exponential speed, um, from the bad ones? Um, and, and how do you make that call now that so much kind of machine traffic is doing something good or legit?
Speaker 2: Yeah. You know, it's interesting. The first step, um, for, for many of our customers was, "Let's just make sure we're actually seeing it." Right? Visibility was, was-
Speaker 1: Yeah
Speaker 2: ... king. I think there's just a lot that we had no idea. Um, our customers had no idea. We, again, we'd see that traffic. We, uh, we have in our practice, we're very focused on what are bad actors doing traditionally. We have intelligence teams that are tracking campaigns, building detections, et cetera, looking for that fraud, right? Um, across, again, the advertising ecosystem and the, and the, the website. Um, but as the agents came in, we said, our- we were working with our customers, it was, "Hey, some of these might be good. We, we might want them to shop." And it very much is a per business dependent, you know, uh, question. They might be investing in their own chatbots, and they wanna control that surface entirely. They might be more a publisher, and they would wanna make sure that none of their content is getting out of the LLM. So that, uh, those business decisions have become, uh, an interesting convergence we're seeing amongst a variety of teams at our customers. It's not just one decision point. But the first step was give them the visibility. So the same, uh, tools we've been using to detect automation for the last 14 years, we, we now are applying to how do we detect the Muses, the GPT dots, the agents, the, the Claude scrapes, et cetera, all of the different consumer-side agents that are coming to sites. Uh, and so by rolling that out to them, it was let's get it visible, and then we gave them controls. We give them the ability to tweak the policies. "Hey, I, I want this one being able to see products. I want this one being able to check out." Uh, really kind of shaping that traffic, and that's just, I'd say, getting adopted as, as this has become a real reality. But it's, it's very much a case-by-case decision that we're working intimately with our, our customers on.
Speaker 1: It's amazing.
Speaker 2: Yeah
Speaker 1: ... every click, whatever. Kind of where does that sort of assumption break first?
Speaker 2: So I think it, it all comes back to, you know, where does it break first? I think, you know, how are we first defining what is the audience, and how are we measuring it? Um, there's still, you know, you and I are using these things. I know I, I use it every single day, right? I'm- Uh, I got my agents rolling now, and they're shopping for me and doing all sorts of stuff, and, you know, ultimately it's still me guiding it.
Speaker 1: Mm-hmm.
Speaker 2: You still- It's still my interests. You still wanna know who I am. But how are we getting to that point of understanding the identity of the human as it relates to the agent? I think that's, uh, you know, I, I have conversations across not only the ad tech space, but a variety across broader martech into, uh, into your security space, cybersecurity, fraud, uh, general commerce. I'm, I'm having a lot with our kind of peer organizations, those we integrate with, and our customers, and everyone's getting that same problem. Like, everyone's trying to fundamentally say, "Who is the, who is the person? Is it John Searby? What agent is he using? Can we verify it's him? Can we verify that it is in fact an agent, and that he's guided said agent, uh, so that we can protect from fraud, but we can also maximize towards, towards, uh, you know, continuing to, to optimize for the good commerce we want from it?" Um, you know, I think that's where it all comes down to. Do I have the visibility if I don't, and, and do I know the identity of the individual and the agent behind it? So that's really what we're, we're seeing breaking first again, not, not I wouldn't say just for advertising, advertising-
Speaker 1: Yeah
Speaker 2: ... right?
Speaker 1: Across the piece. Across. And, and when you show a, a marketeer, a CMO what's really hitting their site, what's their kind of reaction? Are they t- typically sitting down?
Speaker 2: Uh, I think, uh, people are generally shocked, right? And it's, "Hey, can I get that data here or here? Can I use it?" Like, it's, it quickly becomes a, a large integration effort. Um, I think, you know, as a business w- interestingly, we, we work heavily with security and fraud teams as well, and, and not just the marketing teams. Um, and so often it, it requires some collaboration internally-
Speaker 1: Yeah
Speaker 2: ... in the organization. We're seeing that convergence, and they're saying, "Hey, let's make sure we're all in a room."
Speaker 1: Right.
Speaker 2: And I, I'd say more than ever in, in my career, we're having conversations or our partners are having conversations where they have multiple stakeholders sitting there in the same room and saying, "Let's share our tech. Let's share our data that we otherwise in the past weren't even aware of." Um, so I think generally they're, they're pretty shocked when they see it, and it's more about, "Okay, how can I use this operationally?"
Speaker 1: And who do they kind of typically kind of bring together when, when, you know, is it, is it pretty much the entire kind of C-suite sort of thing in, in that-
Speaker 2: Yeah
Speaker 1: ... in that regard?
Speaker 2: Yeah, it's a good question. I think, um, again, you think through those, you know, responsible, depending on the organization, might be product teams-
Speaker 1: Yeah
Speaker 2: ... it might be e-commerce teams. We're seeing a lot of analytic teams, data analytic teams who are trying to get all of this data in one place for them. Uh, and so, you know, that might be with a digital officer or a CTO. Uh, and so those organizations, it, it is case-by-case dependent. I think people are trying to figure out where does the responsibility sit, and, uh, you know, the reality is, is the responsibility's across the board. Like, uh, just like everything we're seeing, this is a fundamental shift in how humans are interacting with machines on, on a, on a most basic level between the human and keyboard. You know, we, we all grew up, uh, you know, mouse and keyboard. Our kids are on touchscreens and, you know, the Siris and personal assistants, and now we have this robot that speaks like a human, right? And so that shift is so fundamental that every m- component of a business needs to understand how it's affecting them, and we need to be sharing signal and data amongst each other.
Speaker 1: And where are we? I mean, you've been doing this 14 years, so that, you know, you've got a lot of, a lot of kind of experience, a lot of expertise, a lot of data. Uh, and obviously along that way, various kind of best practice, standards, protocols have, have kind of been set. But obviously now, as you said, sort of it's shifted again, right? And we're sort of, you know, massively trying to pl- kind of play-
Speaker 2: Yeah
Speaker 1: ... kind of catch up and, you know, to your policy point, I suppose in a sense, sort of like where are we on that side of the journey in terms of, you know, okay, right, you've shown, you, you've shown me and, and then you've given me the chance to kind of create policies-
Speaker 2: Yeah
Speaker 1: ... but I have no idea as to what-
Speaker 1: ... I should or shouldn't be doing in that regard.
Speaker 2: Uh, yeah, it's a good question. We're early in the journey.
Speaker 1: Yeah.
Speaker 2: I think we all are.
Speaker 1: Yeah.
Speaker 2: And, you know, as we, it all changes.
Speaker 1: Yeah.
Speaker 2: I mean, it changes every day.
Speaker 1: Yeah.
Speaker 2: Right? This is the fastest moving kind of news cycle and innovation cycle. It's, uh, you know, that's a real thing that you feel like you're missing out on-
Speaker 1: Yeah
Speaker 2: ... on the news 'cause you're like, "I haven't, I didn't pay attention for 12 hours," right? Um, and so I'd say, you know, it is unreasonable to expect anyone's got this nailed-
Speaker 1: Mm-hmm
Speaker 2: ... and anyone who tells you is, is not paying enough attention because the reality is it's changing fundamentally. And I think, um, you know, uniquely it's changing at that consumer level first. Uh, you know, I always joke, my father, 80 years old, could never use a computer. He just never figured it out, no matter how he hard he tried, and he had to work on it for years. Uh, and I'd always be helping with the most basics. And now that he is talking to his, his LLM every single day and engaging with it, and if he can give it agency to go do the things he never really accomplished, whether it's shopping, et cetera, like it just changed, right?
Speaker 1: Wow.
Speaker 2: And all of a sudden he's a new audience that wasn't traditionally, he wouldn't be on most websites, right? Um, and it's, it's pretty wild.
Speaker 1: It is, and that's a lovely positive sort of democratization kind of side t- you know, to counter the kind of the sort of the, the threats and, and sort of potential kind of negative side of, of, of, of the world that we're entering into. Um, who do you think needs to be leading, um, sort of those guardrails? You know, um, I, I'm thinking sort of, uh, you know, uh, on the basis that sort of this industry has done a reasonable job, I think most people would say, of kind of self-regulating.
Speaker 2: Hmm.
Speaker 1: Wh- wh- who, and who and what sort of, you know, do you see that kind of playing out?
Speaker 2: Uh, you mean kind of just broadly across or-
Speaker 1: Standards
Speaker 2: ... like in an organization or-
Speaker 1: Broader industry terms, I guess. Like, you know, like we, we created the IAB. You know, we need, we came up with standards. We came up with guidelines. We came up... Like, what, what, you know, you said we're early, right? Is that, is that, is that even too early to even be have asking that question?
Speaker 2: Yeah. I mean, I think, and everyone's trying to lead it in one flavor or fashion, right?
Speaker 1: Mm-hmm.
Speaker 2: We're all making a lot of noise about it.
Speaker 1: Yeah.
Speaker 2: Um, I don't think it's one group or, uh, you know, it's not a single organization. This is a collection. As I said, this, this is involving your, uh, security, your fraud, your payments teams, your commerce teams. You know, you're looking and seeing some of the interesting things Stripe's doing or Shopify is doing around this, uh, on, on customer sites, and then you're seeing that, you know, how does that translate back into, uh, into these new services that we're advertising on? I mean, ultimately the, uh, the Frontier lab's building these, right? I think ultimately have to be working with industry-
Speaker 1: Wow
Speaker 2: ... and we have to be working with them to say, "Hey, let's make sure we're recognizing what your traffic is, what your agent is, and who's the person behind it, and let's build those protocols together, uh, so we can better understand that." And then, you know, similarly, if we're gonna start, uh, opening the opportunity to advertise on those surfaces as the new walled garden, the new search, whatever it might be, uh- We need measurement. We need to see it, right? And it's not as simple as, you know, here's a ser- here's a link, and here's a click. It's not, that's not necessarily gonna be the, the same measurement. So I think it is a, is a really a consortium of folks and, and it goes back to those who are currently ho- holding the attention and, and us working with the frontier labs directly.
Speaker 1: And if, if your agent's doing this, you're shopping for you, um, what's the brand adverti- you know, who's the brand advertising to?
Speaker 2: I mean, it's ultimately still me, right? I mean, it's still the human behind the scenes. It's just another mechanism to get, get to them. It's a new surface. It's a, um, you know, you're seeing SEO become AEO and GEO, uh, and it's, it's really gonna just be optimizing for this new path. Uh, you know, I'll take, I, I got into, uh, to woodworking this year. It was new. I like acquiring random hobbies. So I got into woodworking, and traditionally when I would start, I'd, I'd be searching, uh, for, "Hey, how do I, what kind of tool do I need?" Et cetera. And I'd probably Google it, and then I'd go to Reddit, and I'd read some threads, and then I'd flip over to Amazon 'cause I had inherent trust in Amazon that they'll ship it, it's convenient, et cetera. Uh, and then I started using the LLM, and it got incredibly good at directing me and saying, "You need this tool and that." And it started presenting me with other merchants providing me the same tool, maybe cheaper, et cetera, and all of a sudden my behavior shifted materially where I started shopping at other big box stores. It actually told me, "Hey, you can get in the car and drive over there and get it now. It's on the shelf." And I was like, "Oh, that's more convenient. I'll do that." Um, that's a behavioral shift that happens, but it was still me. It was just how it got presented to me. And it's, and I see it as m- very much a, an incredible opportunity. I'm, I'm kind of a, a very much an AI optimist when it comes to this.
Speaker 1: Yeah. Glass half full. Absolutely.
Speaker 2: Yes. Um, and so I think that, yeah, the, the, uh, the opportunity is large. So you're still thinking about how do you gain my attention or, you know, a human.
Speaker 1: Yeah. I mean, ultimately if you're able to deliver, you know, the best price, the best quality, right, and deliver the best service, then, then you, in theory, you should win.
Speaker 2: That's right. That's right.
Speaker 1: Right? Uh, and, and, and then how do brands, I suppose, kind of working to be, you know, um, the kind of the agent's preference, like what, what does that look like in practice?
Speaker 2: How do they actually show up, you're saying? Yeah.
Speaker 1: Um-
Speaker 2: Uh, you know, the AEO world is, uh, we're still building that.
Speaker 1: If, if I had an answer to that, I'd-
Speaker 2: Yeah. Exactly. Uh, I- you're gonna see, I always go back to the incentive structures of, you know, again, who's owning the, the LLM and, um, you know, everyone's gonna approach it differently. Ultimately, whether it's a subscription, they're gonna sell advertising space on it, uh, or they're gonna be part of the commerce thread, and they're gonna take a taste of it. It's gonna be some combination of that where they recoup their costs. Um, and, uh, you know, ultimately they, they're gonna give those pathways, um, for individuals. Some of that is, uh, is going to be overt and, "Hey, here's how you advertise this," and it's gonna be, again, on them to make sure that that's clean supply for the consumer, that's, that, that it's g- trackable and measurable, et cetera. Uh, all the things that have been baked into the cake for years now and, uh, you know, the space we live in that this is real humans looking at it still. Um, but, uh, that, uh, you know, I think the, the incentive there for them, though, is the con- for the consumer is it still needs to get to the outcome. Like, it's still, I expect Muse to shop for me and get there, and so they have to be presenting us with those, those, those, those data points for brands that they can say, "Hey, here's how you show up." Right now, I think it's, it's, it's a little blurry. We do know that w- in a, in, with AEO, there's certain ways to structure your product pages. There's certain ways to get your data out. It's very much where we see an opportunity for us is it's changing depending on if it's the crawl from the LLM, it's real-time retrieval, or it's the agent. And so shaping your traffic, you know, I see a world where, hey, if it's the crawl coming in, present it this page. If it's a live search, present it this page.
Speaker 1: Nice.
Speaker 2: If it's an agent coming to my site, I'm gonna treat it differently.
Speaker 1: Mm-hmm.
Speaker 2: I'm gonna give it different rails. Maybe I'm in the business of controlling that. Maybe I'm in the business of optimizing for how the agent expects it, and this is where you get into certain protocols and things that could help. And so very much it's a, I'd say it's a hybrid-based approach. When I'm a brand, I'd be looking at it and saying, "It's gonna be..." You have to be very thoughtful of how am I presenting to these, these AI agents and the LLMs and scrapes and crawls, uh, in a kinda case-by-case basis-
Speaker 1: Yeah
Speaker 2: ... aligned to my business plan. It's not as simple as, "Hey, I got, I got the right keyword showing up."
Speaker 1: Yeah.
Speaker 2: Right? Traffic's gonna change. Behavior's gonna change in the agent.
Speaker 1: Yeah.
Speaker 2: Uh, and so it's, it's gonna be an interesting, uh, time, but I'd say the, this is where I bring the relationship between hard traffic and what we're seeing in the analytics side to how we're measuring that attention, and that convergence is like, it really, it does have to come together because it's not as simple anymore of a human clicking button.
Speaker 1: Yeah, and you're seeing the full picture.
Speaker 2: Yeah, exactly.
Speaker 1: Yeah, yeah.
Speaker 2: And so you need that, that full picture, so that's-
Speaker 1: Yeah, yeah. That's really fascinating.
Speaker 2: Well, I love-
Speaker 1: I love, I love that as an approach, and it's sort of linked to kind of my, like, like a couple of, couple of kind of questions, I suppose, are linked to what you were sort of saying there. One, one of which was what's the consumer's role in that, you know, that decision-making kind of process? Because, you know, am I giving kind of Muse my, my kind of, uh, full kind of authority? Am I signing different disclaimers with that brand-
Speaker 2: Mm-hmm
Speaker 1: ... to allow that agent to act on my behalf? And therefore, if I've done that, then does that change your kind of policy with regard to that agent acting kind of on my behalf, right?
Speaker 2: Yeah.
Speaker 1: 'Cause I've effectively opted in in a very brand com- you know, compliant way with you sort of thing. And I guess linked to that sort of, you know, you, you, okay, so you've, you've shown me the light. You've given me the full picture. Right now, what, you know, when you're, when you're saying, "Well, look," you know, um, are you, are you, how, how kind of, uh, involved are you, or do you have an ecosystem around you in helping then organizations, um, come up with those policies?
Speaker 2: Yeah. Uh, I mean, I think what you're getting at is this, this fundamental theme of trust, right?
Speaker 1: Mm-hmm.
Speaker 2: That we're all talking about and, and very much we're, we're focused. It's trust that it is actually OpenAI. First, there's, there's an opportunity to spoof it. There's a, you can, you know, bad guys can claim they're OpenAI. Then it's trust that, you know, John Searby's conducting is, is guiding that, that bot in some way. Uh, that is trust from me to the LLM to, like, signing up for all the, the payments and logins. Do I hand it all my information to run, and what am I opting into? So a transparency there of saying if I know- What I'm giving in, and I'm expecting the convenience of this thing's gonna shop on my behalf or interact on my behalf. I know that I'm, I'm putting my data in, and it's gonna be re-resolving my identity to that. My trust that they're gonna use that appropriately, and my trust that it gets me through the door, and more importantly, it gets me to the convenient outcome I want. And the flip side of that is, you know, the brand and the merchants that are sitting there and saying, "Hey, do I trust that I see the complete picture?" Right now, everyone's flying fairly blind across this. Uh, and we can now see, we see that traffic, and we can say, "This is, this is news. It's coming, it's shopping." Uh, but resolving that to the identity so that there's trust on both ends of the equation, that was built for, you know, decades and decades through lots of organic growth in the world of e-commerce, et cetera. You know, it's kinda, yeah, quite when I first, we first started in e-com, like we didn't, we weren't comfortable putting our credit card out on the internet to buy things. It took time, right? And this is moving faster than, than anything we've seen, but you gotta build that trust model, and it does start with the consumer trusting in it as well as the, the merchant side of the house.
Speaker 1: It's great you mentioned the credit card thing. I was gonna sort of zero in a little bit more around fraud, sort of-
Speaker 2: Mm-hmm.
Speaker 1: Give me kind of where, where, where, where, where, where is kind of fraud in, in this, in this conversation?
Speaker 2: Yeah. Uh, so like I said, uh, if you have conversations, everyone's trying to solve a similar problem. So, you know, traditionally, the tools and mechanisms we had to detect, uh, you know, fraud and, and, and financial fraud in some way were different schemes, different mechanisms, different types of scams that actors, they always find, you know, the, the way to slip through. And, uh, it's a constant cat and mouse game everyone in security and fraud is constantly in. Uh, we just handed not only the consumers this incredibly powerful tool, but we also handed the fraudsters, right? And, and ultimately, it became a very low entry point for them to spin up a fake merchant and stand up a s- a, you know, a legitimate-looking website, and, "Hey, I'll ship you these new shoes," and it's, you never get them, and it doesn't exist. It disappears overnight. Um, that applies, I'd say, across the board. That's, that's in the ad space. What fraud are they presenting? They're trying to get people to things like scam merchants. How are they getting that attention? That entry point is so easy now because of the, the power of these tools. Um, similarly, the loss of, of visibility and signal. We took the, we took the customer journey again, where we had certain mechanisms that we expected over time. Hey, a human's gonna come to site. They're gonna fill out this. There's gonna be this payment processing that happens, and here's how we all expect it to work. All the mechanisms have been there, and we just compressed all that, and we're now presenting whole new ones. So fraud is catching up. The fraud ecosystem is catching up to that, and, you know, we're working pretty directly on how do we show the traffic level, the device level, equate that with them to the identity and the payment level, and again, bringing that all together so we can effectively identify when, when this is, is fraudulent in some way. Um, but there's, there's work to be done, and it's really about, uh, again, I'll just keep going back to collaboration and integration. Like it's, it's critical.
Speaker 1: I, I was really looking forward to this conversation 'cause of your background. You know, like you were, you came out of kind of cybersecurity, so you've got kind of a very different kind of lens, and also it just feels like that lens right now is kind of the most appropriate lens we could be kind of using to kind of look at what's going on right now.
Speaker 2: Yeah.
Speaker 1: What, what is it that you kind of brought with you into kind of more the world of kind of ad tech and, and, and kind of commerce, um-
Speaker 2: Yeah
Speaker 1: ... you know?
Speaker 2: Um, you know, in the world of security, uh, and, and where I've come from, it's always, you know, specifically I come from cyber intelligence world.
Speaker 1: Okay.
Speaker 2: And it was always about how do I, how do I have the most data about what actors are doing and what they're gonna try and do, what are their techniques, et cetera, so that I can put in layers of defense and controls throughout so that, you know, we don't, uh, end up with an incident of some kind or a loss. And I think, um, accuracy in the data and understanding of the actor was, was very much the world I came from. And how do you use signal from all these disparate points to better inform your decisions, your technology stack, and what you're doing? And so, you know, similarly, I, I think as I look, like I said, across, uh, uh, ad tech space and us looking at it from, from all these different segments, it's how are we bringing together that data? How are we understanding not only how the fraudsters are doing it and the bad guys, but how the good guys are doing it? Applying that same methodology of, okay, and where are the, where are the layers of defense across the whole tech stack? And maybe not defense, but where am I adding friction and where am I adding flow? And do I have enough intelligence to inform those decisions for how do I present them the ad, how do I get the traffic, how do I then convert them, and how do ultimately do I enable them to transact? Um, and so very much of that, again, that, that world of layered defense and informed data with kind of a rigid understanding of identity, uh, I think is, is key. I think in, you know, security world, we've, for a long time, we've had security and fraud. We've had, you know, very rigid tools to make sure you actually are who you say you are 'cause that's a critical path. And then you have the right roles, responsibilities, et cetera, and accesses. I think that's now gonna have to extend more. We're gonna have to extend that further into an understanding of the human behind the agent, hard verification around that, uh, hard verification of the agent itself. So a lot of those mechanisms I think are, are really, again, moving left, uh, in the, in the customer journey back towards, yeah, attention, getting their attention in advertising too.
Speaker 1: And, you know, this industry has become notoriously kind of technologically kind of fragmented, right? It's, and, and, and then, you know, not even just technologically, but there are so many other kind of actors in the, in the ecosystem. If you can have an actor in an ecosystem, I don't know if you can, but-
Speaker 2: Right
Speaker 1: ... um, you know, whether that's an agency, whether that's the publisher, whether it's the, the, you know... How, how, you know, do you see that sort of that more kind of, I suppose, kind of complicated kind of landscape sort of, um, you know, um, acting as a, a challenge to kind of, you know, where we're trying to go? How are you working with the ecosystem?
Speaker 2: Yeah. You know, it's like any. It's, it's big. It's complex. There's a lot of people with different incentives around it. Um, you know, I think, um, it's, it sounds like I, I'm gonna keep on bringing this into every-
Speaker 1: That's okay
Speaker 2: ... the data, right?
Speaker 1: Yeah.
Speaker 2: The data is the most important.
Speaker 1: Yeah.
Speaker 2: The visibility into it, and we have to be able to share and move that between. And, you know, I'd say I think that we, when you look across, like, the, uh, the, uh, the, the real-time bidding process, supply path, it's for years these mechanisms have been built for a long time. Yes, there's a lot of players, but, like, there's the movement of data and measurement is critical to the business.
Speaker 1: Right.
Speaker 2: It's just shifted, and so I, I don't think, like, um, there's a fundamental change in that way. It's like we're all gonna be adapting to this new, new normal in, in variety arrays. And, you know, I think publishers are in an interesting spot as an example of, "Hey, I'm gonna put this thing on full block," right? I'm just gonna-
Speaker 1: Yeah
Speaker 2: ... full block all the time, and I'm gonna find a different way to monetize my content. But in the end, if you're showing up, if you can still get that traffic, you're showing up in the results, then, then you're one of the one-- You're still driving what you need. So it's gonna be, um-
Speaker 1: Mm-hmm
Speaker 2: I, I wish I had an, I wish I had a good answer, to be honest.
Speaker 1: No, it's-
Speaker 2: You know, the crystal ball right now-
Speaker 1: That's it. Yeah
Speaker 2: ... and it's like, uh-
Speaker 1: Yeah
Speaker 2: ... it, it's-
Speaker 1: Well, we could all go just, we could go to our, our island, couldn't we?
Speaker 2: That's right. That's right. So yeah, it's, um, it's an interesting time, but I think r- for me would be it's here, it's now-
Speaker 1: Yeah
Speaker 2: ... it's real for everyone, so if, if you aren't thinking about this actively today, if you aren't talking cross-functionally to your peers in the, the ecosystem, the organizations, as well as in within your organization, then, like, do it today, right? That's, uh, that'd be the main thing I'd say. Um.
Speaker 1: Yeah, I think one of, one of the things, uh, you know, I, I've been talking about this week, um, has been, you know, the, the publisher kind of side of things and y- as you said, that sort of, that, that debate, that sort of, you know, quandary, right-
Speaker 2: Yeah
Speaker 1: ... of, of do I, don't I, and, and when do I and when don't I? And, uh, but also I think it's making publishers think about, like, what type of content they actually kind of create, right?
Speaker 2: Totally.
Speaker 1: And I think that's really interesting kind of way of thinking about this in terms of, you know, a, a, another wave of, um, whether it's evolution or but, you know, we're human beings. We, we tend to be kind of, um, you know, inventive, you know, in moments of necessity, right? And I think, I think that this is gonna kind of move on the whole kind of, um, you know, the experience, right? A- as a, as a, as a, as a challenge and an opportunity, I think.
Speaker 2: Yeah. Absolutely.
Speaker 1: I was-
Speaker 2: Yeah.
Speaker 1: Yeah, go ahead. Sorry. Please.
Speaker 2: I, I, I mean, it's-- I see it as, again, I'm gonna go with my glass half full here. It's like, okay, so now there's a new way to get my content in front of people.
Speaker 1: Yeah.
Speaker 2: Yes, it might be distilled, it might be changed, and I can, I need to put controls in there. But if I can work to better optimize so I do appear in that, maybe it ends up being a, a more of a net positive for, for some of these folks than they, than they realize. But it's like if you're not intimately testing it and trying it and working and innovating towards it, and if you're just saying, "Let me throw up my walls and hope it never comes," I think that's where it's gonna be a challenge. I think it's viewing it as an opportunity is, is key. But if that makes sense.
Speaker 1: It does, 100%. Um, I was with, uh, a, a, a, a, a programmatic ad tech earlier, and the, the, the, the drum they just continued to kind of beat into me was, um, you know, if you're a publisher, just, you know, focus on authenticated kind of users and whether we're, we'll have authenticated agents. You know, I think, you know, uh, uh, and then another, ano- a publisher was saying, "I'm just, I'm just changing the way I think about what content I kind of create."
Speaker 2: Yeah.
Speaker 1: And so, you know-
Speaker 2: Yeah
Speaker 1: ... it's, this is already happening, right?
Speaker 2: Yeah. It's great.
Speaker 1: Sort of playing out in real time.
Speaker 2: Yeah.
Speaker 1: Really interesting. Um, so w- in terms of kind of the supply of places to advertise, how, how, how does, how does sort of, you know, this new world kind of change that?
Speaker 2: Uh, again, I, I wish I had the full crystal ball to say this is exactly where it's coming, but, uh, you know, ultimately, but we see attention shifting, right? I mean, it's the same way we saw Google Search change it and then the, you know, walled gardens come.
Speaker 1: Yeah.
Speaker 2: And, you know, this is going to be, not only is it n- now the new search, it is gonna have the best data on anyone's interest. I mean, this thing knows everything about me, right? It knows every interest, it knows all my kids and details, and so when you, when you look at it and say there's a new surface, it's not only a new surface, but it's the best surface we've ever had, right? To, to try and gain attention. Uh, and so, you know, obviously we're gonna see dollars flow that way as the mechanisms, uh, are created for, for folks to spend. Um, you know, I think you still are in this world where ultimately it needs to get out and take action on, to your point earlier, was like, hey, it's still, you still need a good product. You still need something people wanna get to in the end. Like, this is just a mechanism to get there. And so there still is opportunity that at some point the human will come back into the loop and be looking at your content and trying to buy your product, et cetera. Uh, and so I think there's still very much an opportunity of how is that getting presented on my site. You're gonna still see retail media networks and commerce media networks. You're gonna still see a lot of, uh, open web advertising, et cetera, but it's just gonna take a slightly different path in how we get, people get to it, I think. Um, but yeah, there's some, there's some interesting shifts happening that I think are, are coming there, but I don't know.
Speaker 1: Yeah, I was chatting to a former CEO of WPP. He's k- launched a new event called Prompt, so it's pretty cool actually.
Speaker 2: Mm.
Speaker 2: Hmm.
Speaker 1: And that was kind of interesting-
Speaker 2: Yeah
Speaker 1: ... in terms of-
Speaker 2: That's, again, it's gonna be such a changing-- Everyone's relationship's gonna be different with this, and that's been true forever. Like, you know, how you interact with Facebook or Google and, you know, everyone's, there's some normal ways we've kind of, people have normalized to it, but, like, there's gonna be a nuance here, and I'd say it's even more personalized, right? Because you're having this conversation that this thing seems to understand you. It's like, what, what am I comfortable with you doing? Again, that trust layer between us and, and the agent's gonna be critical there. Um, but I also see it, you know, as an opportunity as, uh, you know, if, if I am, if I'm, if I'm in fear of my supply not being presented there, do I build a plugin in these LLMs? Do I use it as a new surface? There's, you know, the ecosystem around it is becoming something really material that I think, again, I'm going with the optimist view of this. There is, there is opportunity. It just requires us paying attention and trying.
Speaker 1: And, and you know, if, if I'm a, if I'm a, a brand, I'm a marketer listening to this kind of conversation, you know, kind of what's, what's your kind of advice to me- About what I do now
Speaker 2: Um, yeah. So first, uh, is get visibility. Do you actually see what is coming to your site? Do you see what is actually viewing, you know, uh, y- your ads in the wild? Do- are your providers today giving you that? Can you measure it? Uh, and so I think that, that is first and foremost. So the expectation should be, hey, where is it showing up as, uh, and can you tell me how much of this is being affected by ad? Uh, and so that might require you working cross-functionally. So I'd say pick up the phone, call your peers in your organization, and say, "Hey, do you have something that can help me here?" As I said, a lot of our tool was traditionally sitting with a CISO, and the marketing team didn't even know the data was there, right? And now they're, you know, they're raising their hand saying, "Hey, get that into my Adobe Analytics or my Google Analytics and let me see it-"
Speaker 1: Yeah
Speaker 2: ... all live." So I'd say that's, that would be first. Um, and then, and then you're, you know, it's develop the policies then.
Speaker 1: Yeah.
Speaker 2: And then it's the controls and ongoing measurement.
Speaker 1: Yeah.
Speaker 2: Um, and, and as I said, it's gonna be, uh, you working with a collection of technologies and providers and partners that you've used for some time and, and really be demanding of us in that space, saying, "Can you provide that to me? What is the visibility you can give, and then how do I integrate it into my workflows, into, into my processes?" Uh, and that, and that should be the expectation. That should be the expectation of the platforms you're working with and the agencies who are supporting you and your, and your spend as well, uh, is can you, can you provide us this?
Speaker 1: That's extremely sage advice. Um, let's talk about HUMAN and your strategic priorities now, and sort of I guess, you know, what trade-offs you're having to make or where you're double-downing.
Speaker 2: Yeah. Uh, well, I think you're probably honing in on the thing I'm kind of most obsessed about here, right? Uh, given this conversation. So it, it was natural for us, uh, as we, uh, as we really entered this space. I mean, we forever, we are the guys sitting on websites, sitting in traffic. We are- expectation is we tell you about automation, and the w- world's automation just changed and became bigger and more interesting and a new opportunity, not just, you know, protecting from bad. So that obviously became a huge, a huge part of our strategy. So our Agentic Trust offering around that, really making that available. How do we deploy and integrate that as widely as we can and make it, uh, make it available for our customers? But then also working with our partners, how do we get that data into those flows? How do you get it into your fraud systems? How do you get it into your measurement systems? So that's a, a big part for us is, is that. Um, and then, you know, similarly, when you look across this, this entire space and what we've traditionally done in IVT, you know, we saw a gap in the market and an opportunity that was bringing a new data-driven, transparent approach to broader ad verification of those services there. And so we've, we've launched that. We announced, uh, you know, our, our partnership with OpenX, uh, around that, uh, this, this week, and it's been a, you know, very exciting time where we are really stepping into that space. But that's most interesting to me as it comes together in what we've been talking about here.
Speaker 1: Yeah.
Speaker 2: Right? So that full completeness of signal and visibility from, okay, what is out there, you know, how am I advertising? Is it, again, verified, safe, viewable, brand safe, all those things. Uh, is it being, being viewed by humans or bots? And then am I seeing the outcomes from that when it comes to my actual traffic, and can we bring that signal all together? And so it's very much for us of AI as a, as a thread for us across this, and we want to ultimately move towards, you know, being the control plane for folks across that customer journey and being able to, again, turn those knobs, add the friction or flow where you, where you need it. I'm not gonna walk in the door and say, "This is the right policy in how you do it." We wanna give the tools to our customers to do so. So, uh, you know, the, the trade-off there is it's moving really, really fast, and it's very hard to keep up on it, and we're innovating at rapid pace and investing like crazy into doing it, and I think, uh, you know, everyone who's, who's operating around this space who's like, "Oh, yeah, I'm getting in that game. I'm gonna say I- AI on what I'm doing," you know, we, uh, that's, that's a hard invest. If you're not really investing there, um, and you can't prove it, then that's a challenge. So that's, you know, for us is the trade-off of continuing to do our core opportunity around what we've done for forever and balancing that with this innovation cycle, you know, is, uh, is an exciting but, you know, challenging, uh, decisions that we're going through.
Speaker 1: Yeah. Well, I mean, you used the phrase earlier, but it, it is, it is in a sense a whole new frontier for you, right?
Speaker 2: It is.
Speaker 1: A completely different way of thinking about the proposition and, and what you're, what you're offering the market.
Speaker 2: Yes.
Speaker 1: It's enormous. And, and, you know, to that point, I guess, you know, your, your kind of typical sort of, you know, ICP may be the CSO. Now you're typ- you, you, you have an atypical sort of, uh, kind of ICP in some sense, you know, being like the entire C-suite, but maybe with more of a, a CMO lens on it.
Speaker 2: For sure.
Speaker 1: One's thinking about risk, and the other's hopefully thinking about growth.
Speaker 2: Yeah.
Speaker 1: Um, how do you sort of, you know, square that circle, bring that, bring that, that, that into a useful conversation?
Speaker 2: Yeah. I think, you know, for us it's getting much closer to our customers' workflows as they work today. So CMO and team are gonna operate a certain way. They're gonna wanna, and I mentioned Adobe and Google Analytics, and let's integrate where people work, right?
Speaker 1: Yeah.
Speaker 2: And that's, that's a critical piece for us. Let's make sure we're integrated with, um, your GSPs and SSPs where we're seeing the f- the, the full signal there, right? So it's integrations are a, a big piece of that. Um, but, you know, traditionally they each giving them the data and the workflow, you know, we've, we've worked, again, like you said, with the C-CSO, CISO, uh, the fraud team, increasingly with the CMO and the analytics team or product teams, uh, or commerce teams. Again, depends on-
Speaker 1: Right
Speaker 2: ... the organization, right?
Speaker 1: Yep.
Speaker 2: And what you might name it. Uh, but, uh, it's, it's very much, uh, for us that obviously creates complexity, but I don't expect that I am not the s- I'm not the solution. I am a piece of this puzzle. I just want to give the data where they need to operate.
Speaker 1: Yeah.
Speaker 2: I wanna give the controls. I wanna be able to redirect the traffic. We were talking AEO, et cetera. I wanna work with those providers. I wanna work with Adobe, et cetera, and say, "Hey, let's give you the policies, and we're just, we're just a control plane that helps you get there and an integrated data point." Um, you know, whether that be in the fraud space or, or marketing space, it's the same, same conversation as Let's, let's actually integrate this thing 'cause we, we, we kind of ignored each other for a little too long, right? In our tech stack.
Speaker 1: Yeah.
Speaker 2: You know? And this is forcing some-
Speaker 1: This is overdue.
Speaker 2: Yes, exactly. Yeah.
Speaker 1: I love it. That's a great answer. Um, I... Last, last sort of thing I wanna kind of just chat to you about is, I suppose in a sense the next generation and, and sort of like, you know, how you kind of learn this job now, right? You know, if you go back to the John when he was 23, um, you know, kind of what would you be advising him to sort of, um, you know, go and do now?
Speaker 2: Become a plumber.
Speaker 1: Electrician.
Speaker 2: Yeah, yeah. Electrician.
Speaker 1: Yeah. Yeah.
Speaker 2: Uh, use AI to build that business and do it fast, right? Uh, I think there's a huge opportunity there. No, I, you know, I think, um, uh, for me, uh, you know, if I were entering now in this timeframe, it's, it's very much like you gotta keep up with the speed of innovation. You actually gotta roll up the sleeves and try it. You can't just read the news article and move to the next thing or take someone's word for it. For me, it's like I, I try to play every single day with it. I try to use it every single day. Uh, and so how do you, uh, how are you staying a-ahead of the innovation curve? And that means working with your peers, talking to partners, actually trying, right? And so I think, um, get out there. Like I, I don't know what-
Speaker 1: Yeah
Speaker 2: ... what's the right way to say it, but-
Speaker 1: Roll your sleeves up.
Speaker 2: Yeah. Like-
Speaker 1: Just do it
Speaker 2: ... just roll your sleeves up. Actually use it.
Speaker 1: Yeah.
Speaker 2: I mean, I have a number of children growing and, you know, advise young folks going into college and things, and it's do what you love. Do what you're interested in. I don't care what that thing is. Electrician, probably. Whatever. Building something or whatever it is, but you get a minor in AI. Everyone gets a minor in AI. Everyone should be using the tool every single day because you... It, it's a reality of the world now, you know?
Speaker 1: That's a great answer. I'm gonna hit you up with three final quick-fire questions.
Speaker 2: Okay.
Speaker 1: Um, what's the one resource you rely on to keep up?
Speaker 2: Uh, ChatGPT.
Speaker 1: Might be the same answer to all three. What-- Where do you go to for inspiration outside of work?
Speaker 2: Ooh. Uh, I like building with my hands. I like-
Speaker 1: The woodwork
Speaker 2: ... whether that's, you know, the woodworking or it's, uh, you know, doing, uh... I got into 3D printing and designing. Like the outcome of something real and tangible. Barbecue. I like cooking and having something come out. So, uh, for me, it's like, you know, sometimes I gotta lift my head off that AI, uh, world and the email and, and actually go back into the world and remember what's good about it, right?
Speaker 1: Mm-hmm.
Speaker 2: And why we're doing it. So I... That's, that's, that's key. And then, you know, my family. Like I said, I have, I got five little kids at home and, uh, you know, they keep me on my toes every day. Uh, and they're, they're entering a changing world, so it's important that I'm, you know, understanding the changing world to be part of it with them, so.
Speaker 1: And a brand you admire and why?
Speaker 2: I wa- I'll go with one that just kind of off the cuff just 'cause it came to mind, but, uh, Sonos. I love Sonos. I'm a big Sonos guy. I got speakers all over my house. I love the simplicity of it. I love their branding. I love everything about it. Um, you know, they continue to innovate in really interesting ways as well, and integrate. You know? It's the themes of the day here, huh? And then they're, they're very big on that, and so I'd, you know, I'd say that's, that's a, that's a good one for me.
Speaker 1: It's a great choice. Hard and soft.
Speaker 2: Yeah. Say again?
Speaker 1: Hardware and software, right?
Speaker 2: Absolutely.
Speaker 1: That ain't easy.
Speaker 2: Absolutely. Yeah.
Speaker 1: That's a great choice. It's been a real pleasure. Thanks for joining Small Business Marketing.
Speaker 2: Yeah. Thank you. Thank you for having me. This has been great.