How Banking CMOs Can Play the AI Long Game and Win
To avoid repeating 1999's internet bubble mistakes, marketing leaders need a strategic three-layer approach to AI: strong first-party data foundations, thoughtful technology implementation, and human creativity that delivers differentiated value.
By Justin Estes, Contributor at The Financial Brand
Simple Subscribe
Subscribe Now!
In an evolving banking ecosystem, CMOs must approach AI as a transformative force while avoiding the pitfalls of short-term thinking and hype cycles.
On a recent episode of the Banking Transformed podcast, host Jim Marous spoke with Eric Fulwiler, co-founder and CEO of marketing consultancy Rival, about how true marketing innovation will come not from AI tools themselves, but from the strategic application of customer data, technology, and human creativity.
Q: You’ve compared the current AI revolution to the internet of 1999. What similarities and differences do you see between these major transformations for CMOs?
Eric Fulwiler: It’s not too hard to draw this correlation because, essentially, it is the disruption cycle just being played out in a different way. The disruption cycle, if you’re not familiar with it, there is a hype bubble in the beginning, then you go down what’s called the trough of disillusionment, and then you end up on the slope of enlightenment to eventually what the future state is going to be of any new technology or any disruption within an industry.
The connection to the internet in 1999 is a key point, and I think it’s helpful for people to consider. Much of the conversation right now is dominated by elevated expectations. I believe that unrealistic expectations exist within the hype surrounding AI in the marketing world.
That bubble will burst because history repeats itself; these are socioeconomic cycles driven by human nature, which has remained unchanged over the last 20 years. And many businesses will get caught up by investing in short-term, shiny AI objects instead of considering the fundamental impact on their business, customers, members, and the industry as a whole.
For me, what we’re working with clients on is what the future state 10 years from now looks like as opposed to the next 12 months. You don’t want to be a Pets.com in 1999. You want to be an Amazon; you want to be a business built on what AI is going to do to disrupt the marketing industry over the next 10 years, not just chasing the hype cycle that everybody else is pursuing over the next 12 months.
Q: Why are most marketers still "playing checkers with AI while the winners are playing chess"?
Fulwiler: I always go back to first principles. All growth comes from being able to add differentiated value to the end customer, the end consumer, the end member. And right now, I feel like so many CMOs in this category, but also in others, are focused on the internal dynamics, the career dynamics of "we need to be doing something with AI, what is it" as opposed to "what is the value that it is actually creating for the end customer?"
Fundamentally, I’m not an AI skeptic; I believe this will be the most transformative technology that our industry, and probably the world, has ever seen. But sometimes I feel like it’s a situation where the emperor has no clothes. Everybody in the industry is talking about these amazing things that AI, ChatGPT, and LLMs are doing within the marketing world right now.
When I use them, when I utilize a lot of these technologies, it makes things easy, is how I think about it. If you don’t know anything about something, if you’re like an intern or a junior marketer, it’s going to replace a lot of the stuff that you would do.
But if you’re a CMO, and if you are a business and a brand that wants to be differentiated, that wants to be ahead of the curve, that wants to be leading within your industry and against your competitors from a marketing perspective, right now, a lot of the AI technology especially around LLMs, does not do great and it certainly doesn’t do differentiated.
Even without the technical understanding, if you understand what these LLMs are doing, they are fundamentally statistical prediction machines. You ask it for something, and it looks at all the data that has been trained on, as well as the context from human beings, to say, here’s the most likely thing, and the most likely thing to make you happy. These applications are designed to make the user happy. It’s telling you what it thinks you want to hear, and it’s also providing similar answers to those that have come before in your industry.
And for us, all the research we’ve done on challenger brands reveals that, at the foundation of what challenger brands do differently, they’re differentiated within their category. And so, if everybody is using the same tools, which are trained on the same data and getting similar answers, how is that helping you to be differentiated in what you do in marketing?
The Three-Layer Model for Marketing Innovation
Q: You’ve described a three-layer model of first-party data, AI tools, and human creativity. How should CMOs allocate resources across these layers?
Fulwiler: We recently finished the research on a short documentary about the future of the CMO role. We interviewed Raja from MasterCard, as well as a number of current and former CMOs, venture capitalists, private equity firms, and industry bodies, to gain a 360-degree perspective on the current state of the CMO role.
The short version is that there are really three things the CMOs of today need to do to be effective in their role: first, they need to be able to speak the language of the board, not marketing-speak, but business and commercial language at the board level. Second, they need to own and consistently deliver commercial results. One of my biggest pet peeves in our industry is what I call marketing for the sake of marketing.
When people ask me, "What do you think of this campaign?" "I don’t know." "What were the results that it delivered?" Marketing is a means to an end, and CMOs need to own and deliver the commercial results in the end.
But the third piece is data and technology, of which AI is a part. And what they need to do, returning to my marathon example, is to take the 10-year approach and get in shape right now. Since last year, I have spent 30 minutes every working day getting a taste of the tools available, experimenting with them, and trying to do my job more efficiently with them. It’s like most things – I don’t think the answer is complicated, but it is hard to do consistently over time.
That’s really where it starts — you need to start prioritizing this to gain a first-party expert’s understanding of this technology and its implications within your business. And broaden the lens. It’s not just about ChatGPT and LLMs; it is about the data layer, the tech layer, and also the human and talent layer on top of it, and what you can do to create a whole that’s greater than the sum of its parts with talent and technology together.
Balancing Short-Term Wins and Long-Term Strategy
Q: How can marketers balance quick AI wins with long-term brand-building initiatives?
Fulwiler: You’re probably undervaluing the long-term and overvaluing the short-term. But I understand the reality of what I call "the work around the work." There’s the marketing job you need to do, and there’s also the job you need to do to sell it through to the board and other stakeholders.
There is enough low-hanging fruit. If I examine our business and where we’re working with clients right now, we’re looking at how technology can enhance effectiveness or efficiency. For example, our creative team is beginning to generate a significant amount of basic content using AI. It is not replacing the creative director who comes up with the ideas or the art designer, but when it comes to "we need 80 different variations of this one concept," then we’re using AI to make those changes.
Dig deeper:
- Stop Marketing Like It’s 1999
- Break Through the Noise: Reinventing Bank Marketing for Today’s Consumer
- The Art of Standing Out in Financial Service Marketing
We’re also developing tools to leverage AI for consumer journey mapping, a process that typically takes six months and costs around half a million dollars. We can achieve this in about six weeks for a fraction of the cost by leveraging the data these AIs have been trained on, combined with first-party data from our customers.
The first and most important thing is to place greater emphasis on the long-term opportunities and risks. However, the second step is to examine what you’re doing now and ask, "Where can AI enhance effectiveness or efficiency in our current processes?"
Measuring Marketing Success in the AI Era
Q: How can CMOs shift from being cost centers to revenue centers through their approach to marketing?
Fulwiler: So much of marketing is about thinking and acting like a media company because a media company’s business model is attention. How do you attract, retain, and then monetize people’s attention?
CMOs need to think and act like heads of media companies, focusing on how to earn attention and then monetize it through the products they’re selling. At 11:Fs, marketing was a revenue center, not a cost center for the business, because we built up so much attention with the content and the experiences in the community that we were able to monetize that through sponsorship like a media company would and cover the costs and then some of the whole marketing activity.
That’s an extreme example, but directionally, even from a mindset perspective, I think it’s interesting to consider what it would mean to take that approach within your business.
Learning from Marketing Innovators
Q: What daily practices can marketing leaders adopt to prepare for an AI-driven future?
Fulwiler: Three things. First, step outside of your bubble and get different perspectives. Listen to people like Gary Marcus, who’s a bit too much of an AI skeptic, or Ethan Mollick, who’s a professor at Wharton who wrote a great book called Co-Intelligence.
Second, the "a little bit every day" approach is probably the most important thing. Ask your team for an audit — it doesn’t have to be formal. Just say, "Where are we in terms of our readiness and our fitness? Where are we using AI now? Where should we be using it? What should we be doing differently?"
And finally, put aside a bit of a budget for AI testing and learning internally. The way I would tie it off, going back to the marathon metaphor, is it’s just like getting in shape for something physical — you just have to do a little bit consistently over time. No amount of training that you can do in a month is going to get you in shape to run a marathon next month; you need to start doing a little bit consistently over time.
