Why AI in Marketing Isn’t What You Think
We're drowning in ads. Between 3,000 and 10,000 commercial messages hit us every day, yet most vanish from memory
We’re drowning in ads. Between 3,000 and 10,000 commercial messages hit us every day, yet most vanish from memory within seconds. Meanwhile, marketers are losing their minds trying to keep up with an explosion of channels, shrinking attention spans, and the impossible math of personalization at scale.
I saw a video involving Raja Rajamanar from Mastercard and Stefan Pretorius from WPP that pulled back the curtain on how AI in marketing might actually solve this mess, and it’s not what you’d expect.
The Problem Nobody Talks About
Here’s something that’ll blow your mind: Pretorius casually mentioned scrolling through 13,000 ad variations for a single campaign. Thirteen thousand. “I literally have to scroll to page 1300 to get to the end of this,” he said, almost matter-of-factly.
This isn’t some extreme edge case. It’s Tuesday for most big brands now. Every product needs dozens of versions for different audiences, platforms, times of day, weather conditions, you name it. The old Mad Men playbook of crafting one perfect 30-second spot? Dead and buried.
Marketing departments are basically running digital sweatshops, churning out endless variations of the same message. As Rajamanar put it, current marketing practices are still based on “theories and principles from more than six decades ago.” Ouch.
The Real Impact of AI in Marketing
The fear that AI will replace human creativity misses the point entirely. Pretorius made a distinction that should be tattooed on every creative director’s forehead: there’s commercial creativity and there’s pure creativity.
“In the world of marketing, we are not creativity for creativity sake,” he explained. Translation: most of what we call “creative work” in advertising is actually systematic grunt work—resizing images, swapping out headlines, maintaining brand consistency across thousands of touchpoints.
Nobody got into advertising to spend their day in Photoshop creating the 8,000th variation of a hand soap ad. That’s assembly line work masquerading as creativity. AI can handle that stuff in seconds, freeing humans to do what they’re actually good at: thinking strategically about what makes people tick.
The Data Surprise That Changes Everything
Here’s where things get interesting. Everyone assumes better targeting means knowing more about individual consumers: their age, income, browsing history, favorite coffee order. Wrong.
Pretorius dropped this bomb: “Very often non-personal data, contextual data about where you are, what’s the weather like… are far more useful in terms of predicting behavior than who you are.”
Think about it. Knowing someone is a 34-year-old software engineer tells you less than knowing they’re standing in an airport at 6 AM on a Monday morning when their flight just got delayed. Context beats identity almost every time.
This is huge for privacy too. Instead of building creepy detailed profiles of individuals, smart marketers focus on moments and situations. It’s less Big Brother, more helpful friend who notices you look tired and suggests coffee.
Synthetic Focus Groups: Brilliant or Terrifying?
WPP’s demo showed something straight out of science fiction: AI personas that act like focus group participants, trained on mountains of consumer research data. You can literally have conversations with synthetic consumers about your marketing ideas.
“What should the packaging be?” Pretorius asked the AI. It responded with detailed feedback based on eco-conscious Gen Z preferences.
The speed is incredible. Traditional focus groups take weeks to organize and cost thousands. This happens instantly and costs pennies. But here’s the catch—when Rajamanar asked about validation against real focus groups, Pretorius admitted, “I wish I had a greater answer for you.”
We’re creating “superhuman” research tools that might be completely wrong. That’s either the future of market research or a very expensive way to make terrible decisions. Probably both.
The Job Question Everyone’s Dancing Around
Let’s be honest about what everyone’s really worried about: jobs. When AI can create 13,000 ad variations in minutes, what happens to the humans who used to do that work?
Pretorius’s answer was refreshingly direct: “AI shifts tasks.” Instead of eliminating jobs, it pushes people toward more complex, strategic work. The person manually creating thousands of product shots gets freed up to think about brand strategy or consumer psychology.
But he also said something that should scare every executive: you can’t just create an “AI department” and call it a day. This transformation has to happen “from the CEO all the way down.” Companies that try to bolt AI onto their existing processes are going to get steamrolled by competitors who rebuild from the ground up.
The Brutal Truth About Getting Started
Want to know the difference between companies that will thrive with AI and those that will crash and burn? Pretorius summed it up in five words: “Start doing things right now.”
Forget waiting until next quarter. Skip putting it off until the budget meeting. Don’t hold back until the strategy feels perfect.
“Stop talking about it. Stop waiting. Stop watching others do it,” he said. AI is weird that way, you can’t learn it theoretically. You have to use it to understand how it changes your thinking.
Companies still holding planning meetings about their AI strategy while their competitors are already shipping AI-powered campaigns? They’re toast. This is the reality of AI in marketing today: you adapt or get left behind.
The Uncomfortable Questions We Need to Answer
This whole AI transformation raises questions that keep marketing executives up at night:
If AI generates thousands of brand messages, who’s responsible for brand authenticity? When synthetic focus groups replace real consumers, are we optimizing for what people actually want or what our algorithms think they want? As personalization reaches infinite scale, do we risk creating echo chambers that make society even more fragmented?
There’s also the practical stuff nobody wants to discuss. What happens to creative agencies when production becomes push-button simple? How do you manage quality control across 13,000 ad variations? Who’s liable when an AI-generated campaign goes sideways?
Where This Is Really Heading
The marketing industry is about to get weird. We’re moving toward a world where the limiting factor isn’t production capacity or media spend. It’s our ability to understand what actually works and why. This transformation represents the true future of AI in marketing.
Imagine campaigns that adapt in real-time to weather, news events, local culture, individual moods, and a thousand other variables. Imagine testing hundreds of messaging strategies simultaneously and automatically doubling down on what works. Imagine never again launching a campaign that flops because you tested everything beforehand with synthetic focus groups.
That’s either marketing paradise or consumer hell, depending on how we handle it.
The Real Revolution
The biggest change isn’t that machines are getting creative. It’s that humans are getting smarter about creativity. AI forces you to think more clearly about what you’re actually trying to accomplish and why.
As Pretorius put it, working with AI makes you “more aware of how you think.” That meta-cognitive awareness (understanding your own decision-making process) might be the most valuable skill marketers can develop.
The companies that figure this out won’t just be more efficient; they’ll be fundamentally different. They’ll move faster, test smarter, and connect with consumers in ways that feel almost psychic.
The revolution isn’t coming. It’s here. The only question is whether you’re ready to stop making excuses and start making changes. AI in marketing isn’t just about new tools—it’s about rethinking everything we thought we knew about reaching customers. stop using “AI” as a catch-all solution and start thinking critically about what each task actually requires.
