AI Beyond the Sandbox

Drew: Hello, Renegade Marketers. If this is your first time listening, welcome. If you're a regular listener, welcome back. I'm Drew Neisser, Penguin-in-Chief of CMO Huddles. And today, it's a special Drew-on-Drew episode where I interview myself. So the guy answering the questions will be the one with a hat. The one asking the questions will be the guy over here. Okay.

Drew: So, yeah, Drew, it's always an honor. Though I have to say, after the Knicks' comeback on the 19th — that's just a few weeks ago — I'm sure you don't have any voice left.

Drew: Well, that's fair enough. I mean, they were down 22 points in the fourth quarter. I was ready to declare total loss, and yes, once again, the Knicks were gonna disappoint. But with eight minutes left, somehow they came back from overtime to win. My son and I were texting like it was 1999, which for the Knicks is basically the last time hope had a forwarding address.

Drew: Yeah, the whole — you know, Spike and the gang must have aged backwards, like, several years. It was a crazy, crazy game.

Drew: You know, maybe Ben Stiller actually smiled in public. It's hard to confirm, but here's why that game stuck with me. It's not just the frantic energy. It was the momentum, the execution, and the belief meeting exactly at the right time. And that's where a lot of CMOs are with AI right now. The scoreboard is changing. The old plays are not enough, but there's still plenty of game left if you call the right plays. And so, yeah, I just managed to go from the Knicks and basketball to AI.

Drew: Okay, so Drew, today, obviously we're talking about AI and marketing orgs and how CMOs can move from sort of random experimentation — or as one CMO said in a huddle recently, to letting 1,000 flowers bloom, I love that phrase — to something closer to an operating model. That is a goal here, folks, and maybe we can talk about that on this show.

Drew: Okay, I don't know who's asking the questions here or who isn't, but I think we're gonna pull from two recent CMO Huddles leader huddles and from Noah Brier's big picture thinking, which he did at the Imaginarium Summit. Scott Sedman, thanks again for having us there. If you've never checked out Imaginarium, do yourself a favor. They're doing some amazing things with data and marketers. Anyway, it was a unique conference, and so between the huddles and Noah's comments, I'm gonna try to give you some practical, maybe slightly silly, and hopefully more fun than watching a Knicks fan during the first three quarters.

Drew: So okay, all right, question number one, Drew. Why start with a Knicks comeback? Is this just your excuse to talk basketball?

Drew: You know, yeah, I don't mind talking basketball. I'm not as expert as my son — I can't name all the players even — but it's a useful metaphor, and so that's what I wanna sort of get at. They didn't win because they ran faster randomly. They won because the right players made the right plays at the right time, and I think CMOs are facing a similar AI moment. They've got a lot of energy and a lot of pressure and need to turn this scattered activity — which it really feels like random acts of AI — into momentum. So the lesson is not to panic in the first quarter, and honestly, if we're pushing this analogy, it's not even the first quarter with AI. But we gotta know the score, you gotta know your plays, and you gotta know who should have the ball.

Drew: What did you hear at the leader huddles that made this AI conversation feel different?

Drew: Interesting, and it has been quite an evolution. I think there's no doubt that CMOs are not wondering, does AI matter? I mean, come on — that conversation, the lights are out on that one. This conversation has shifted, though, from "what tools are you using" to "how does this change how marketing runs?" Huddlers talked about organization design. They talked about governance because this stuff can get out of control — not just from a security standpoint, but also a cost standpoint and a brand standpoint. In these conversations for the first time, build versus buy came up, sales alignment and making sure that we're using these tools for them, and of course, measurement and team readiness. One CMO described all this AI work as happening off the side of people's desks, which I think captures the moment perfectly. And I quote — yes, I'm back to the Thousand Flowers bloom story, because I just can't get over it. That was what we were thinking about this six months ago, and really now there's this central tension. Experimentation created momentum, but operating models create scale. We need to be thinking about this on a strategic level.

Drew: All right, so Drew, big question here. What's the biggest risk for CMOs right now?

Drew: Oh, my God, there are so many, but I think the biggest one is confusing activity with progress. And, you know, God forbid you work for a company that has a scoreboard based just on token usage — which is stupid. You could spend your entire day working with AI but not actually get to any outcomes. So I think if we talk about the biggest risk, we can put it in multiple areas. There's a "God, is any of this really gonna work?" risk. And so I do think that we have to start to think more strategically about what problems we're trying to solve and how we're doing it as a team. So yeah, teams are launching more content, campaigns, agents, and workflows, but measurement — like, is any of this thing working? I think there's an inherent understanding that yes, they're able to create more content faster. We still don't know: is that content better? Is it performing better? And so that's problematic right now. One Huddler noted that teams can cite individual productivity wins — "I can't believe I can do this podcast, which used to take 10 hours, and now in 30 minutes" — but not necessarily relate the total impact on a function or business result, and I think that's so problematic. So what are the big strategy areas that you could solve? And again, I think the easiest way to frame this is: do we wanna become the most customer-centric organization in our category? And if so, how could AI help? The answers that would produce would be extraordinary. Another Huddler talked about how the output is happening in real time, but the measurement just can't catch up fast enough. So all right,

Drew: Drew, important question. What should CMOs measure first?

Drew: I know I'm supposed to be answering these questions, and honestly, I don't have all of the answers, but I do think I have some guideposts here for you. This came up in a huddle. One Huddler suggested: pick one CMO-friendly metric and don't try to measure every micro-experiment. So let's say, for example, the goal is to lower cost per acquisition. I know that's a big thing, but now we can start to think — let's break this up into pieces. We know how much a lead costs. We know how long it takes to nurture that lead, and we might even know what value per customer is. So where can you use AI to accelerate some of those things and lower those costs? Because if you can lower your cost per acquisition, you'll be a friend of that CFO forever. So another example might be scaling revenue without scaling the BDR team. This is a moment where I'll probably plug 1mind — I might do that later. I just love that tool — but 1mind is a way of scaling your SDRs. One SDR at HubSpot reported that it was the equivalent of, like, 98, and it works 24/7 and doesn't need a water break. So how could you scale revenue without scaling your BDR team? Well, one answer might be to have a conversational superhuman on your website. So another metric might be: more qualified meetings without adding headcount. That's an interesting one, and so that might get down to: how do we make sure that our ICP is stronger? How can we engage with the right target more effectively? And what role could AI play? Because in all of these things — and then maybe pipeline created with AI-assisted workflows, I think that one's gonna be harder to measure — but in the earlier ones, we created a business objective and then said, "How could AI enable?" And what's so interesting about all of that is when you get into that conversation, you realize it's not just a marketing problem, it's a go-to-market problem, so you have to involve sales. You have to involve customer success. And to me, that creates a unique opportunity for CMOs right now to lead and help transform not just marketing but go-to-market. So I think a way of thinking about this is: if the CEO can't repeat your AI metric, it's probably not your headline metric.

Drew: All right. This is all really, really interesting, Drew. You kinda answered the question — good job, sort of. Where are CMOs seeing real tactical wins right now?

Drew: Well, we moved from strategy to tactics pretty fast on this show. Yes, we did. So I'm gonna talk about AI SDRs. One CMO shared that they had already contracted with Qualified, and it's funny because they had seen the 1mind demo, but the deal was already struck with Qualified. They went, "Oh, I really wish I had 1mind." But anyway, what they were happy to report was that with Qualified, they were still able to get 20 meetings booked and 12 opportunities with — and I think it was like $29 million worth of pipe. So that was a real tactical win. The AI SDR was booking meetings at all hours, including while human SDRs were, you know, sound asleep. So obviously this matters because it solves a business problem. But I also wanna point out that what's interesting about getting an AI SDR up and running is that that's maybe one person that you didn't hire — an SDR that you didn't hire. But it's a standalone project that won't break because you bought it, not built it. And I do think there's a lot more risk in build than any of us thought about three months ago, because the truth is the pricing of these tools is way underpriced. And so if that pricing goes way up — as it could, and we can see it coming — then we're gonna have a problem with all of these agents that we built and we're running our business on. So I do think this is a moment to think about build versus buy. That doesn't change the strategic angle at all, because again, AI SDRs most likely are just gonna help you be a more customer-centric, more sales-centric organization. So this is an open — if we go for the basketball frame, you got a clear look on a three-point play, you gotta take the shot.

Drew: Okay — yeah, that basketball thing. Yeah, dribble. You're dribbling, Drew. So does this mean that CMOs should buy instead of build?

Drew: Let's put it this way. I think a week ago I probably would've said, you know, buy the things that are absolutely essential to the business. And I think I probably would double down on that again, but I also would've encouraged folks to build all sorts of workflows and agent systems and so forth. But where I'm getting a little more reluctant is this notion that they're gonna have hundreds of agents running with just maybe one person in the loop.

I don't think we know what the cost of those hundreds of agents are. And while you're busy destroying the marketing department that you have that's actually driving the pipeline that you have, you need to be careful about this. And so I do think that hedging your bet a little bit and buying a few tools that are fundamental is not a bad bet right now.

That doesn't mean you can't build a couple of individual workflows that will improve productivity dramatically. I mean, I'm just loving the agents that I've built with Codex that are giving me daily briefings for all of our meetings. I'm loving the brain that I've been able to build as a result of the 520 podcasts and being able to find information in that.

There's a lot of things that I think individuals — oh, I had a calendar analysis that looked at where I was spending my time relative to the priorities. Again, these are productivity enhancement things. Nothing wrong with those agents, you know, and this is the moment to build all of those things because it's never gonna be any cheaper.

So I think it's gonna be a combination of build and buy. One thing I haven't really talked about at length is there is a tremendous risk on the build side, from a compliance standpoint, from a cybersecurity standpoint, from a brand protection standpoint. So in all of these things, anything that includes customer-facing information, I do think that you need to really think about cybersecurity and involve your side.

So I do think you need to make sure that brand governance and there is a governance and a human is in the loop. And again, that's why folks buy — because that's part of what is built into some of these tools. So yeah, I'm doing a little bit of a flip-flop right now. I think while I want everybody building and learning how they can be dramatically more productive as individuals, I still think organizations, and particularly larger ones, are gonna buy some pretty important tools.

Drew: What's the deal with agents making tiny mistakes?

Drew: Yeah, this one came in from a huddle and it's really relatable. A CMO was describing an agent that had built a very sophisticated ABM strategy and was able to do a lot of the things that you wanna do in ABM, but one thing it couldn't do is get the calendar date right in any of the emails. And so it's just they can't use it, or they would always have to check it. And so if you're trying to do these things at scale and it can't even get the date right, that's a problem. Another huddler noted that as a recipient, one obvious error can invalidate everything good. And I personally have this story that I love.

The same day this happened this week, I got two different emails from the same organization wanting to appear on this very podcast. One of them was addressed to me and it said, "Oh, Drew, I just listened to the last episode. I love the point that you made about XYZ. I think my guest would be perfect because they can speak to this topic."

And then the email that was right there was written to a different CMO organization and it said, "Hey, podcaster," — whatever the name of that individual — "I loved your most recent episode." It was the exact same phraseology. So obviously there was a data problem there with that tool, and there's no way that individual who was pitched is gonna be on my show.

So again, I know this is about mass personalization, but you really have to be careful because you are going to turn some people away. So maybe it's a numbers game and you don't care, but I do think there's some brand risk involved. We also had a wonderful expert huddle with David Horsager, who wrote Trust Matters, who makes a convincing case about how individuals and companies build trust.

And I think that we really have to be careful with AI. Think about it as brilliant and brittle at the same time. So you need to start to think about your agents on a risk level. Low-risk agents, like if they can monitor, summarize, and report — great, that's internal. Medium-risk agents can recommend actions. High-risk agents touch prospects or customers. Those need human review. You've gotta figure that out.

So, again, as a basketball fan, it was a painful moment at the end of the Duke season this year when all Duke had to do was hold onto the ball, but a rookie who should have known better tried to throw it over somebody else, and the other team hit a three-pointer. It was so painful as a fan, I gotta tell you.

So anyway, you don't need to make rookie mistakes. You're all more sophisticated than that. So while we wanna use AI to solve the big pictures, the big problems, the things that you build and the agents that you build, you may wanna think about sort of the risk levels involved.

Drew: Okay, we're still going. Question number eight: Where does sales enablement fit into all of this?

Drew: I am so glad you asked that one because this is like the exciting and unappreciated AI win out there. One CMO sort of describes sales teams using all the marketing thought leadership inside a prompt. So they built this wonderful brain. I think they used an outside tool — I can't remember if it was Writer or Optimizely, Misay, or whatever.

But all of the brand guidelines, all of the style, all of the deep information about the product was there, along with the ability to create presentations and so forth integrated into this tool. So now the sales folks are saying, "Oh my God, this has all the answers. The format is correct. I'm creating better output, better emails, more useful case studies, and the messaging is just so much better as a result of this."

So I love that story, because usually sales enablement is just something that drags on both marketing and sales. Marketing creates a bunch of stuff and sales never uses it. Well, if everything is created on the fly but it's all on brand — that's a huge opportunity. So this is a huge shift because marketers spent years wondering whether sales would use this content.

Well, now this tool not only measures it, but it's giving it customized and personalized with a human in the loop. So I think this is gonna be a moment where marketing and sales really combine and sales feels like, "Oh my God, they really get it." So you're gonna need to build these approved content libraries and brains, and that's an agentic opportunity that I think exists for every marketing and sales organization.

So let's use these brains to generate more relevant outreach, meeting prep, and follow-up with a human in the loop. Okay. I guess the basketball metaphor is, you know, it's marketing gets an assist and everybody appreciates it.

Drew: All right. So talk to me about what Noah said and why you're even bothering to talk about it, and who the heck is Noah anyway?

Drew: So Noah Brier is the founder of Alephic. He's a longtime friend. He runs a terrific conference in New York City called Brand X. I strongly encourage — he just opened early bird tickets to that. And he was at the Imaginarium Summit, which was hosted by Scott Stedman. Love Scott. Anyway, his big idea, or one of the big ideas that he talked about — because he helps people think about AI on a big picture —

He talks about AI as a mirror, not a crystal ball. And this is such a concise and important insight. AI reflects your organization back to you. If you have messy processes, they will become more visible. If you have weak training across the board, that will become more obvious. If you have disconnected data or bad data, that becomes a bigger blocker.

And if you have bad corporate writing from the beginning that you're ingesting into your tool, guess what? You're gonna have bad output. So Noah really talked about and warned against expecting use cases to just come down from the mountain. He talks about teams — yeah, they need to explore and test and apply imagination to their work.

And so in his case, he talks about don't ask executives what are the AI use cases. Ask what important work are we trying to improve, accelerate, or rethink? And this goes back to the very point, and obviously Noah and I had a Vulcan mind meld. Think about the strategy that you're trying to solve and the fact that AI will enable you to do things you could never do before.

So in Noah's words, "What important work are we trying to improve or accelerate or rethink?" That answer, to me, is pretty profound.

Drew: Yeah. Okay. We're trying not to get too profound, Drew. So what does team readiness look like?

Drew: So I think this is a really interesting and important thing, and why I'm gonna raise this up a little bit. We are in a moment where marketing departments are — people are thinking about revising what their marketing organizations look like, and it's pretty vast in terms of how they might approach this thing.

And so however this is gonna look, whether you're gonna have, God forbid, humans reporting to agents — I hate that idea — but humans and agents working together, everybody in your organization moving forward is going to need people who are ready, right? So team readiness. And it's not just about having access to the tools.

I mean, it's interesting — several huddlers have power users and task forces and curious teams, but not everyone has time to learn technically, well, how do I use this stuff? So as a CMO, an empathetic leader who is helping people grow, the greatest gift you can give is to help your employees see where this is going, why it's important, how it will empower them, how it will enable them to be superhuman, if you will.

So I think — and not everybody's gonna be as technical. Team readiness, to me, looks like you've got people, as one CMO talked about it, who are empowered and enabled, but not everyone can become a technical wizard. So I think the solution to that is you've gotta have at least one technical wizard per, let's just say, ten employees and a hundred bots.

And the technical wizard inside of the right combination of creative strategic problem solvers is gonna be terrifically helpful, because the strategic folks on your team can think about, "Oh, wow, we could become a more customer-centric organization. We could deliver this, but I don't know how to build it or buy it or even evaluate a tool if I were to buy it."

So I think moving forward, we're gonna be combining curious marketers who are AI power users and technical builders. I think you can be a curious marketer and a power user, but you may not be a builder, so we need that technical person on the team as well. So the goal is to turn ideas into workflows, if you will, without requiring everyone to become an engineer.

So again, maybe this practical structure could work for you. One power user supports a pod of marketers. A pod identifies use cases tied to business goals. Power users help build a prototype. Marketers own the business judgment and the quality. I think that's a pretty good framing moving ahead. So again, we don't need five point guards — going back to the basketball theme. You need a system where everybody knows their role and can work together to solve these problems.

Drew: All right. So Drew, you have made it to the finish line. Oh — it's overtime, for heaven's sakes. What's the final word for CMOs?

Drew: Being a Knicks fan may still be harder than being a CMO, but both require resilience. You don't need to be down 22 points to change the game. AI experimentation was the warmup, and God bless you for doing that. The next move is to turn these experiments into repeatable plays, repeatable plays into systems, and systems into an operating model.

CMOs who do this well won't just use AI more. They'll build marketing organizations that learn faster, execute smarter, and create more leverage. I kinda like that. I think we could do this. I did have one little last thought about reorgs — this is gonna be harder than anybody imagines, so the main thing I want you to focus on is just making sure that your employees are ready and you know who they are.

Drew: All right, Drew, bring us home — preferably without fouling anybody.

Drew: All right, CMO, your 30-day assignment is simple. Pick one business problem, pick one CFO-friendly metric, pick one AI use case, assign a buddy pair, set the guardrails, and run the play. Then check the scoreboard. If it worked, run it again. If it didn't, adjust. That's how experiments become systems.

Drew: And Drew, is this where you mention the CMO Super Huddle?

Drew: Yeah, well, good. Smooth transition. Nicely done, Drew. So yes, join us October 22nd and 23rd at the CMO Super Huddle, where we help Huddlers join the top 5% of AI-empowered marketing leaders. We have 10-plus AI experts, including Noah Brier, who I mentioned earlier. We've got fresh research on AI adoption and a possible maturity model, and the findings from the FOMO — as in Future of Marketing Org Design Task Force. Okay, and so —

Drew: And Drew, will Spike Lee be invited?

Drew: No, not yet, but hope is a strategy if you're a Knicks fan. For everyone else, build an operating model. Until next time, keep those renegade thinking caps on and strong.

 

For more interviews with innovative marketers, visit renegademarketing.com/podcast and hit the subscribe button.

Show Credits

Renegade Marketers Unite is written and directed by Drew Neisser. Hey, that's me! This show is produced by Melissa Caffrey, Laura Parkyn, and Ishar Cuevas. The music is by the amazing Burns Twins and the intro Voice Over is Linda Cornelius. To find the transcripts of all episodes, suggest future guests, or learn more about B2B branding, CMO Huddles, or my CMO coaching service, check out renegade.com. I'm your host, Drew Neisser. And until next time, keep those Renegade thinking caps on and strong!