Michael Treff | CEO, Code and Theory | Why Your AI Strategy Needs to Be More Than Tools & Efficiency - An Agency Perspective

0:00The CMO Confidential Podcast is a proud member of the I Hear Everything Podcast Network. Looking to launch or scale your podcast? I Hear Everything delivers podcast production, growth, and monetization solutions that transform your words into profit. Ready to give your brand a voice? Then visit IHearEverything.com. Welcome to CMO Confidential, the podcast that takes you inside the drama, decisions, and choices that go with being the head of marketing. Hosted by five-time CMO Mike Linton. Marketers have been working on personalization since the dawn of the internet, and actually even earlier than that.

0:43Now with AI, you can deliver ads made for each of your customers. Publicis Sapient is committed to faster, better personalization, and is proud to introduce Slingshot and Bodhi AI. Now, these AI platforms are designed to customize every key touchpoint and reduce creative timelines from weeks to days. With Publicis Sapient, marketing isn't just fast, it's personal. Smarter marketing, happier teams, wow-worthy customer moments. Upgrade your marketing at publicissapient.com.

1:25welcome marketers advertisers and those who love them the chief marketing officer confidential cmo confidential is a program that takes you inside the drama the decisions and the politics that go with being the head of marketing at any company in what is one of the most scrutinized jobs in the executive suite i'm mike linton the former chief marketing officer of best buy ebay Farmers Insurance and Ancestry.com here today with my guest Michael Treff. Today's topic, why your current AI strategy needs to be a lot more than tools and efficiency, an agency perspective. Now, Michael is the CEO of Code and Theory, an agency that combines creatives with engineers of all things to bring clients answers at the intersection of technology and marketing.

2:14Under his leadership, the company has grown to over 2 ,000 people and worked with many clients, including JPMorgan Chase, Black & Decker, Microsoft, and Qualcomm. They were also just named ad agencies, B2B agency of the year, and were called out by Fast Company as an innovation leader. Full disclosure, we met at the ANA B2B conference in 2024, and this is his second time on the show. I also learned that he played in a punk band called Baby Teeth that came on the last show. So who knows what's going to come out this time. Welcome, Michael. Hey, thank you so much for having me. Thank you for having me back.

2:54It's always such a pleasure. Awesome. All right. Well, let's have fun like we did last time. You know, you see the B2B marketplace writ large all day. There's so much going on, political change, cultural change, technology change, and what you think is the most transformative change ever, like a lot of people, AI. How are companies dealing with this? And then we can talk about agencies and other things. All right, great. I mean, it's a big question. Thanks for opening us. You know, from our perspective, it kind of goes across the spectrum. There are so many seismic changes that are happening all at the same time.

3:30You mentioned a bunch of them. There's political, you know, lack of certainty. There's massive transformation in technology and AI. There's also huge disruptions in consumer behavior. And what everybody forgets in B2B is that these are people too. They're just customers, just like anyone else. And so we're seeing these huge waves of disruption really affect the entire spectrum of our client base. Now, specifically in B2B, people think it's unique and it's different. And in some ways it is. The procurement cycles can be very different. The deal sizes can be different. The scrutiny can be different.

4:06The need for alignment can be very different across a client organization. Oftentimes, the feedback responses that you get from activities that you do to drive revenue in the B2B category take longer. So you don't have as much to optimize. But that said, at the end of the day, our approach has always been, these are people. These are people that need help making decisions. These are people that have needs, and how can we best service them? And in today's world, A lot of that work, a lot of that work is on really understanding where people are, where are they engaging, what are their behaviors, what would be useful to them in their decision journey, and how you can be more of a holistic partner to your customers in the B2B space.

4:47Because that's going to cut across technology. That's going to cut across your operations. That's going to cut across marketing decisions. And so we're pushing our clients to look at the problems they can solve for their customers and then align around that. So implicit in that answer is that what the business world needs is a much more holistic look than they've maybe taken before us. And B2B is now looking actually at kind of customer problems and how to fix them versus just give me some marketing to sell this thing I made. Is that a fair comment? I think it is. I think it's a nice summary. I mean, and it's unfortunate because you can do so many amazing things in the B2B space.

5:32Like there are more complicated decision cycles. That should be an opportunity. You know, there's so many more ways you can add value into that funnel. There's so many different ways you can bring content and thought leadership and information and whatnot to your customers. The touch points can be very different. And oftentimes more and more of them are inhuman in person than they were, you know, in a traditional consumer funnel because these decisions can be a lot bigger. oftentimes they involve partners go ahead yeah and it's also a marketplace where a lot of times you know all your buyers are all your potential customers and right you know i mean depending on the product so so this is interesting we'll talk about that in a little bit um i want to i want to ask you about you know there's been an awful lot of mergers lately with the biggest one being you know omnicom ipg there's people that are saying and they've been on our show like rashad and other the folks saying this is going to be, you either get really big or you get really little and the middle are very specialized and the middle is no place to be.

6:32Tell us about that. That's a great question. So I think there's a lot of truth in that. I think the middle is hard. I also don't think that's analogous to like agencies and professional services businesses. Like look at the world. The middle is getting squeezed to the polls, right? That's in any category you want to look at. So that's definitely happening and that is dangerous. So what does that mean? For us, when you look at these mega mergers, we see it as an opportunity. We're part of Stagwell. We're definitely a challenger holding company, nowhere near the size and scale of some of our competitors, but overperforming.

7:02And why are we overperforming? Because we're 100 % focused on our customers, on our clients. We're less focused on figuring out how IPG and Omnicom come together and who's going to do what job and what the governance is. So I look at it like a massive opportunity for us. We can step change by solving problems and staying focused. Like I ask our clients all the time, you know, do you want to spend the next couple of years figuring out what your team looks like and what the contracting is going to look like? Or do you want your agencies like completely focused on driving performance and solutions for your business?

7:35And that seems to be a good sell. Now that said, I should, there's a big caveat here. I don't want to lose the caveat. All right. Let's go with the caveat. Let's hear it. Thank you. Thank you, sir. So there's no question, though, that scale has advantages, sometimes in data. And if we're going to move into an AI conversation later on, the gold that allows AI to be the most effective is the data that you put into it. And so certainly when you deal with these mega mergers and the sort of polarity that's going to happen in our professional services category, there are massive advantages to that. And they shouldn't go unrecognized.

8:09The counter is it's much harder to make that data useful across an enterprise when that enterprise is five, six, seven times the size of other enterprises. And so there's a catch 22 there. Okay, super interesting. Let's talk about, you know, when we were talking before the show, you talked about the concept of leadership to win in an anxious market. There's no doubt it's an anxious market. We actually have done a couple shows on just how anxious the market is and how to get through it. But what does that mean in your mind for leaders of today? And then let's talk about what it means for code and theory.

8:48Okay. This week's episode of CMO Confidential is brought to you by Publicis Sapient. Visit them at www.publicissapient.com. At Publicis Sapient, we're not just talking about AI. We're delivering real results. Today. With Sapient AI, we speed up software development and save clients thousands of hours. We revolutionize how travelers discover their perfect getaway. and even turn a simple fridge scan into a personalized meal plan. Making AI work for you, it's what we do. Now back to our conversation with Michael Treff, CEO of Code & Theory. I mean, look, what I'm going to say here doesn't sound like revelatory, but I rarely see it in practice.

9:37And we have a philosophy, and what we believe the best leaders in the organizations that are going to win or doing is they're making disruption and change their offense. So instead of reacting to all the uncertainty or freezing because of that uncertainty, which you see a lot of, and you can see that in the deal numbers of what's happening from a new business perspective across the industry, there's a lot of stagnation, wait and see, we're not so sure, similar to what happened in AI in the very beginning, right? That paralysis is the absolute opposite of what leaders should do in uncertain and shaky times.

10:12You should make this opportunity your offense. Because what can you do when things are uncertain that you otherwise couldn't do? You're not necessarily pulled down by the day-to-day. You're not pulled down necessarily by the aggressive growth that you're experiencing. You're just trying to keep up and service and keep the quality high and build the leadership and scale. So we put our clients to make moments of disruption, moments of uncertainty, moments of change the offense. Take that opportunity to change the things that you don't like because everyone will accept it's an uncertain time. We have to make a lot of changes.

10:42The cultural bar of acceptance is so much lower in those moments. You have a real opportunity to do that. So you asked what we do also at Code and Theory. I mean, we spent the last year doing that at Code and Theory. We replatformed our management team. We looked at how we go to market, how we productize our services, how we package them. We built new capabilities and new practices that we felt were the future-leaning practices. We invested in technology, so we could also have technology solutions that we can bring to our clients that are productized, that are going to accelerate their success.

11:15And we changed the way we work because there's more scrutiny. But there's a whole cultural and a mindset shift in there, which I think is really important actually for marketers and agencies as well, which is you have to be on offense to win. But you want a whole bunch of your culture and your company wants you to be on defense because offense you can actually lose. But there's this old general quoted gentleman to secure people to succeed. You have to ride towards the sound of battle. And that is you got to go do something or you won't get good enough at it to actually develop any skills. And eventually defense will fail you, I think.

11:57Yes. And yes. And old improv motion there. You have the ability to. Yes. And you have the ability to experiment in a different way. when things are uncertain. You can place a few little bets and see what happens. You can be more liberal in terms of where you're prioritizing your offensive plays because you can't force clients, at least in our category, in this professional services category, if people are uncertain and they're not sure what's going to happen with tariffs or their budgets, et cetera, they're not sure what the impact of all this smart tech investment and AI investment is going to be, you can't control necessarily that.

12:36They're going to appropriate money and make decisions when they're going to make decisions, you can help them, you can counsel them, of course. But you have the ability to do things in an unencumbered way in these moments of uncertainty, and you should experiment and use that to your advantage. And to be clear, yes and is the improv thing where you get the line from somebody, you cannot reject the line. You have to take the line and build on the line. You can't say, that was a terrible feed, Michael, I need a better thing and you have to accept the line and i think that that's the other thing i think you're saying here which is you have to go on offense in some way even though it's uncomfortable or you will never win you will just play defense but there's a huge amount of companies playing defense um and how do you talk people into this offense for real when they don't want to do it or they want to wait it's really hard and look that's why you know oftentimes we're able to change ourselves faster than we're able to change our clients because we built a culture of always changing at coding theory, right?

13:38We're used to it. We want it. We crave it. You know, for clients, it's harder to do. And, you know, a lot of our work has changed from, and I think this is probably true of the broader industry, from just executing the RFPs and executing the work and executing the deliverables to really being the change agent across the C-suite to drive the thinking change, the sea change in approach. We work with PMOs because we do a lot of marketing services work. We work with CTOs because we do a lot of technical and infrastructure work. We work with CIOs because data is the key of all of this. We work with CEOs because business priorities and chief revenue officers, et cetera.

14:14And when you sit there and you get that view, what you see is that stasis hurts. The wait and see allows an opportunity for others to leapfrog you. And that is what you don't want. You don't want to come through this to a more stable, more growth-oriented time saying, oh, I wish I had made these changes when we had the opportunity. Well, the other thing, since no one actually knows exactly how it's going to play out, you have no practice if you're just playing defense. And if you don't have any practice, you don't have any experience, and that will probably kill you. I want to dive into this a little more because it was, I think over a year ago, we talked about our AI in the marketplace.

14:54You know, your thoughts, winners and losers. And then talk about like, we'll talk about AI strategy and why, you know, the title of the show. But tell us what's happened in the last year. And I know you also have something you must say about prompt engineers. Yes, of course. Obviously, I have to go back to that theme. But look, I'm sure you've seen it too. So, you know, your perspective is very valuable as well on this. But what we've seen, at least from my perspective, is a massive acceleration in the efficacy of the software. It's just gotten so much better. And it doesn't matter if it's like a small fit for purpose tool or if it's larger enterprise systems.

15:32Like these tools, these AI built tools are just getting better and better and better. So much so that they're accelerating at far greater leaps and bounds than anyone had imagined. Every time, you know, whatever it is, Gemini or GPT, it doesn't matter who it is, releases the new version. you hear like, oh my God, it's so much better than we could have imagined. And that's not going to slow down. Like the acceleration curve will continue to go. And it's faster than any other historical adoption curve. Like that's really important. And it requires no infrastructure to go faster either. Like I don't have to have an install base or users or nobody, you know, to do mobile.

16:10People actually needed mobile devices. AI It means nothing. It's just going as fast as it can go. Yeah, so we're seeing that, and we're seeing massive spend. As you know, we're seeing massive, massive – see, I don't want to overhit you with stats, but I'm going to hit you with a couple of stats. You knock yourself out, Michael. Go ahead. Thank you. Thank you. Here comes the math lead in me. So it's clear everybody is investing tremendously in MarTech and AI. 80%, I mean, according to like IBM and Deloitte, 80 % of CEOs say that AI will transform their business, but like 28 % of them have a vision on how it will drive value for the brand other than efficiency and cost savings.

16:51Another stat, 70 % of C-suite leaders say AI is critical. This is according to us, Ensure and McKinsey. But 28%, 28 % say it's delivering strategic value currently. So what we're seeing is massive adoption of software under the banner of efficiency, cost savings, do more with less, scaled content, whatever it may be. But we're seeing dissatisfaction with the ROI of that spend, with the value extraction from that. And partly, and this is where I think the winners and the losers to your question, the winners are going to be the people who understand that having a plan for how your entire enterprise orchestrates AI, what you want to do with it holistically, how it will benefit customers, how it will benefit value generation you can create for them, and then what you want your humans to do.

17:43That's the next wave of this. It's orchestration, intent, imagination, and purpose. Well, let me distill that. And by the way, I wish you could generate a little more enthusiasm about this. Like what I hear you saying is, and it's in the title of the show is a lot of people are thinking their AI strategy is the application of tools and saving money. And we just saw a bunch of consulting companies lay off people, you know, and the jobs are disappearing and that AI is having a super impact there. You're saying it's a lot more than that. And I don't know if you ever actually talked about the prompt engineer.

18:19So let's talk about it's more than that and also why the prompt engineer was a terrible strategy last year. So let's start there. It's such a flashpoint for me because there was such a focus for like a year, 18 months. There was like mania and chaos about we're going to need to hire prompt engineers and they're$450 ,000 a year to hire and there's not enough in them and no one knows how to do it. And I got to find the kid who's 19 at MIT who can like lead my department. It's like, there's some truth in that. Like, it's good to have AI native people who are now growing up with this the same way, you know, we grew up with the internet.

18:55but like the software is going to figure that out. The whole idea is that it's artificial intelligence. It's meant to be more and more and more human. It's meant to operate with us as a co-pilot in a human capacity. That means if you have to really hire people to engineer prompts, to get the outcomes that you want, you're either, you don't have the wrong software or you don't have a strategy around how to orchestrate all of it. And so that's where it goes for me. So I think this is probably and why prompt engineers are featuring so largely in this little discussion is that was viewed by some people, I think what you're saying as a strategy, I will dominate prompt engineers and I will hire a bunch and effectively those tools and then they will save me money and this will be great.

19:40And what you're saying is that's not really a strategy. That's an execution of, you know, that it's better than nothing, but it's not really a strategy. It's not like writing it into your workflows or writing it into your job specs or anything else. It's just, it's bolting on something. And I think that's what you're calling out here about the losers are doing it the wrong way. Everyone's doing it. Is that what you're saying? That is what I'm saying. So most of the focus is on, let me get the right software in, let me get these tools in, and let me make sure my people adopt them so I can show my CFO the cost savings and the efficiency savings.

20:16I know I'm saying that's not important in a challenged business climate, but yes, and there needs to be more than just the tool identification. Because at the end of the day, the value of AI is connecting people, tools, products, the weight workflows, and data. it can connect those things, those three things in a different way than anything we've ever experienced before. It can become the integration and orchestration engine that allows end-to-end enterprises to do better. So like, what am I talking about? I'll give you like a real example. You have CMOs on this show all the time. And if you ask them, what is your biggest frustration in the way your organization uses data?

20:59Generally, the answer is we have so much of it. We don't know what to do with it. It's all not connected. And I can't really unlock the value of all the insights because it's not orchestrated across our team. So maybe this team has like this little sliver over here and this team has this little sliver over here. And everyone moves to the data that favors them the most. Exactly. So that is a perfect example of what AI can solve. So we're building this, like for all of Stagwell, we're building a system, it's called the machine. it is an orchestration operating system to bring people data tools together fine it's going to really significantly drive value for ourselves and the way we work and our clients no question about it but that's just a perfect example on that now let's let's take another example from the other side it's called the machine it's called machine like florence and the machine sorry correct no i appreciate the reference thank you i mean you know in my in my band circles i'm not sure that she gets as much credit as she should have but we'll discuss that another time yes but that's the idea.

21:54So another example, how many of your CMOs that you have on, and this is another example where the AI strategy of orchestration wins. They have agencies, they have agencies all over the world. They have specific agencies, they have AORs, they have IATs, but really at the end of the day, are I am looking at performance completely end to end? Because I've never seen that. I've never seen a performance dashboard and real-time analytics that I can agentically say, go optimize my segmentation, my targeting, and this creative, and it just happened in real time. I've never seen that connectivity. And that is the power of AI.

22:27So when I sit here and say, yes, you have to get the tools right, but the software is going to solve that. The real challenge is, okay, where can it add value in connecting my people, my tools, and my data to drive the enterprise forward with its products, its services, its ways of working for its customers? And no one's talking about that. Well, thank you for ramping up the enthusiasm. I think that's a great point. I want to flip over now to the creativity thing. I think one of the things you may have said is creativity is being democratized. Tell us more about that. Great question. So as you mentioned, I was in a punk rock band.

23:04And one of the things that happens when you're in a punk rock band is you really don't know how to play your instruments when you start. And that's true for me too. I still don't know how to play any instruments, really. And like, that's what was amazing about that and like getting in a band and starting it was that like, you just did it. You just did it and you figured it out, right? When we talk about creativity being democratized today, think about how you were raised in school. If you could draw, you were creative. You had a craft skill. If you could write, you were creative. You have a craft skill.

23:34So the craft skill was what segmented you into the creative bucket or the non-creative bucket. But having an idea about a business, having an idea about a financial thing that could completely change the balance sheet, those are creative functions too. Strategy, insights, the ability to tie it all together, all of that is very creative. But those people who didn't have a craft skill were relegated to the non-creative workforce historically, unless you have like very understanding partners and, you know, companies who can embrace that. Today, the tools solve that. I don't need to, I'm not a coder.

24:06I'm a very bad engineer, but I can code full websites through prompts and software today. I can do it right now. So am I now a creative in the traditional definition? So that's what I mean. Like the craft execution, anyone's going to be able to do. Now certain people do it better and taste matters and skill matters. But the execution of things is going to software. So the question becomes today, not like, what creatives am I hiring? You should expect everyone to be creative in what they do. You want full stack creatives, people who understand strategy, execution, design, communications, and technology.

24:44You want that because the tools will enable that. But you've got to ask yourself as an organization, and this goes back to your previous question, what do I want my humans doing? Well, I do think there's a case to be made here. And we've had it on a couple other shows of the person that actually can think creatively is going to be super valued. And that is different than the person that can draw a great picture or do great film. It's the person that can see the field and see the opportunity. And that will be where the creativity goes. You know, I want to pile onto this because one of the things we both agree on, I think, is this endless search where you put people in a room where you challenge them to come up with the big idea.

25:32And tell us your theory on the big idea and why it's often a waste. I love you for bringing that up. Thank you. So like about a decade ago, I started to write this book and I totally failed because there was no chat GPT and I'm a terrible writer. It was called Death to the Big Idea. Because I was so sick and tired of getting these briefs from – I will not name names – of clients where it was like, we need the big idea. It's a$200 ,000 social campaign. And it's like, this is absurd. Like what are we talking about here? And also, what is even the definition of a big idea? because you can make an argument that sure, it might be the brand platform.

26:10It might be the communications platform. It might be the campaign. No argument. That's valid sometimes, but sometimes it's actually the product itself and how to make the product better. And sometimes the big idea is actually who your audience should be because maybe you're trying to sell the thing to the wrong audience. So like, I believe, and I started to write this book that this concept of the big idea, the big idea has distracted the entire industry into a singular focus on traditional platform creativity, which left so much opportunity off the table. Look, I agree with this, and I know you're going to talk to my class at Case this year, but one of the things about the big idea is there's this assumption if you demand and interrogate the room long enough, it's going to show up when, in fact, it doesn't show up like that.