The Marketing Chaos Tax: We Built Marketing for a World That No Longer Exists

Part 2 of The Chaos Tax series

When GenAI first entered the mainstream, we marketers were ecstatic. We'd spent years talking about 1:1 customer relationships. Now we had the ability to personalize at scale, produce content at a fraction of the time and cost, and reach customers more precisely than ever before.

Three years later, we got what we wanted. Everything got worse.

AI slashed the cost of content creation,[1] yet with customer acquisition costs up 40–60% in two years,[2] the combination of cheaper tools and greater speed has only made results harder to achieve.

In Part 1, we defined the Chaos Tax as the accumulated cost of operating in a market that reinvents itself every quarter. Marketing is paying a double tax: first for the chaos that already existed, and now for AI compounding on top of it.

Technology isn't new to marketing. Every era has been shaped by it: TV. Radio. The web. Mobile. Social. Streaming. Marketers didn't invent these channels but were always tasked with figuring them out after they arrived. What's different now is the tempo. TV took three decades to mature; social took five years. Each wave started before the last one had fully landed.

Then AI arrived, and brought its chaos with it. Marketing, built for a different tempo, was still chasing the last wave.

AI didn't break marketing, it exposed it. It revealed how much strategic judgment had been hollowed out, how fragile the operating model had become, and how much of what we called marketing capability was actually project management. This exposure event is amplifying what was there before: a marketing function optimized for volume and reporting, not judgment or growth, and unprepared for the wave now hitting it.

Truth #1: Efficiency has come at the expense of differentiation

Content demand has exploded fivefold in two years, and 71% of marketers expect another fivefold increase by 2027.[3] We used to joke that every startup website looked alike. Now everything does: ads, emails, landing pages, social posts. AI didn't create sameness, but it certainly industrialized it. Nearly two-thirds of customers now say most brands seem the same.[4] The industry's response? Make more. The customer's response? No thanks.

Half of consumers now prefer brands that avoid GenAI in consumer content.[5] No wonder "slop" (low quality, AI-generated content) was Merriam-Webster's 2025 Word of the Year.[6]

Marketing is supposed to be the home of customer insight, imagination, and connection. Our industry spent decades celebrating it. Then, when budget pressure came, we stopped protecting it. We ceded customer understanding, creative thinking, competitive rigor, and go-to-market judgment to other functions, platforms, and agencies. What largely remained was a coordination layer of marketers managing internal and external partners doing the work.

Strategic marketing fundamentals were sidelined in favor of technology, scaled operations, and dashboards. We built ambitious content infrastructure. In the process, we turned marketing into a content factory, and called it efficiency.

Marketing AI tools can invent strategy, fabricate understanding, validate bad ideas, and eliminate creative friction. Or they can deliver on the promise of customer understanding and connection. Whether they create value or amplify mediocrity depends entirely on the judgment directing it.

"Human in the loop" is the industry's answer, a polite way of saying we've given AI a babysitter. But if we don't train marketers who think critically and strategically, "human in the loop" is just a rubber stamp with extra steps. Validation replaces challenge, independent judgment atrophies and accountability disappears. The result is weaker work and less engaged customers.

Truth #2: Measurement is distorting value

Another challenge is marketing's ability to evaluate itself clearly. Measurement systems that govern modern marketing were designed to answer one question: what can we prove worked this quarter? The KPIs we ended up building around this have become destructive. Often, what gets protected inside organizations isn't always what creates the most value. It's what can be more easily measured and defended in a quarterly budget meeting.

Finance has been accelerating this for years by demanding quarterly ROI proof on every line item.[7] Yet McKinsey interviewed more than fifty senior marketing leaders and found that none of them could clearly articulate the ROI of their MarTech stack.[8] And it's not the marketers' fault; existing pressure on marketing to meet goals designed for robotic performance versus meaningful customer interaction has only added to the mounting chaos.

What's easy to measure is activity. But action gets disguised as progress disguised as outcomes. The actual capability that drives the outcome is invisible to the system that determines funding. Once human contribution becomes invisible, organizations start assuming it's easily replaceable, because the system never learned who contributed in the first place.

What measurement can't see, it can't protect, and if it can't protect, it eventually cuts. That's how marketing gets measured into mediocrity. The current approach to AI in marketing only exacerbates this. Gartner found that 84% of marketing organizations are trapped in this cycle.[9] Underfunded measurement leads to unclear impact, which breeds C-suite skepticism, which leads to tighter budgets. Companies stuck in this loop are half as likely to exceed their growth targets.

Truth #3: The next wave is already here

Everyone is currently selling a variation of "agents," which realistically is often just process automation with better UX and an underlying LLM. The real thing (autonomous agents) is just around the corner, with AI systems completely reshaping the brand-buyer relationship, capturing behavioral data you need and filtering messages marketing used to control.

The most dramatic shift is in discovery. AI is rapidly creating new ways for customers or agents on their behalf to find, evaluate, and buy. ChatGPT handles 2.5 billion prompts a day,[10] Google's AI Overviews are rewriting search results, and nearly 60% of searches already end without a click to any website.[11] When someone asks AI about your category, your brand is being shaped through these interactions.

But decisions about what to shape, and why, are not optimization problems. They're strategic ones. McKinsey estimates AI agents could orchestrate $3–5 trillion in consumer commerce by 2030.[12] Your brand's presence in these systems is being formed right now, and this is where we need to pay attention.

No matter the wave, good marketing demands strategic judgment, creative instinct, customer understanding, and sharp competitive sense. AI can write copy, run a journey, and personalize messages. But it can't tell you whether you're solving the right problem or whether the answer it comes up with is the right one to solve those problems. That's still a human job, and the job we need most.

How we fix it

This isn't a critique of marketer judgment. Senior marketers were handed a function that needed to be built for a future nobody had a crystal ball for, given budgets that continued to shrink, told to prove ROI on every line item, and asked to ride never-ending technology waves. The hollowing out was the predictable outcome of an almost impossible job description. But now the people who were forced to make tradeoffs that weakened marketing are being told to use AI to "fix it."

But a brand's fundamental questions haven't changed: Who are you? What makes you not just distinctive, but irreplaceable? How do you want the world — and now, its agents — to see you?

We're constantly reminded that technology matters more than ever before. The question is which technology, pointed at which problems, and chosen by people with the expertise to know where it works and where it doesn't. AI's real promise is straightforward: making good marketers better marketers. This isn't a bleeding-heart argument about job protection. It's a strategic argument about protecting the function that creates growth and value. But it only works if we stop treating AI as an outsourcing mechanism designed to replace marketers and start using it to amplify human talent. Otherwise, we're just funding the machine while starving the people who direct it.

The chaos isn't going away. Redesigning marketing to absorb the volatility is the only strategy that survives what's coming next.

Bringing judgment back

If I were sitting across from you, here are the questions I'd start with.

When did your team last produce something that genuinely surprised you? Something that made you stop and think, "I've never seen this before. A machine didn't make that." If you can't answer that quickly, it tells you something about how little independent judgment is left in the function.

That leads to a harder question: what's irreplaceable about your organization? Instead of focusing on your positioning statement or brand manifesto, go deeper into actual understanding, competitive differentiation, and internal judgment that make your marketing organization a great one. And if you can't name what's unique, go figure it out.

Try this exercise: Open ChatGPT or Claude (name your LLM of choice, I won't judge you) and ask the questions your customers ask before they buy from you. What comes back is increasingly your brand's front door. Is it accurate? Favorable? Is it even you?

What are you actually trying to do with AI? Do you have a goal? How would you know it worked?

Finally: are you prepared for the version of the job that's coming? Gartner found that 65% of CMOs expect AI to dramatically change their role in the next two years. Only 32% think they need new skills to meet it.[13]

Marketing didn't break because of AI. It broke because it was optimized for systems that no longer exist, and in the process, stripped out the capabilities it now needs. The marketing leaders who come out stronger will be rebuilding a marketing function that absorbs volatility and creates real differentiation and value. Over the rest of this series, we will show how. But the Chaos Tax is compounding as we speak. The hard questions won't wait, and the answering is yours to lead.

Jess Lin is co-founder of Atmos Partners, a strategic advisory firm helping PE-backed and global enterprises design for AI value. Previously Global GenAI Lead for Marketing and Global Marketing Transformation Offering Lead at Accenture Song. She writes about what happens when the marketing and growth function meets AI. Get in touch at [email protected].