Marketing AI ROI is rising—but clarity is not
Marketers aren’t short on budget—they’re short on certainty.
That’s the uncomfortable truth lurking beneath the headlines about rising spend and expanding AI investments. For all the talk of transformation, what many marketing leaders are actually navigating is something far less glamorous: ambiguity dressed up as progress.
In a recent conversation, Liz Miller, vice president and principal analyst at Constellation Research, cut through the noise with unusual clarity. The issue isn’t whether marketing is changing. It’s whether anyone can confidently say what’s working.
The illusion of “more with more”
The narrative around marketing budgets 2026 suggests cautious optimism. Budgets are ticking up, if only modestly. But as Liz explained, that increase comes with strings attached.
“We’re getting this budget with an asterisk,” she said. “Our CEOs and boards are looking at us and saying, ‘Where’s the AI?’”
More money, in other words, hasn’t simplified marketing’s job. It’s complicated it. The expectation isn’t just to do more—it’s to prove more, faster, and with greater precision.
That shift has quietly changed the nature of scrutiny. Marketing has always been accountable, but now it’s being asked to justify not just outcomes, but the path taken to get there. Experiments are questioned. Exploration is scrutinized. Even hesitation—once a sign of strategic thinking—can look like indecision.
It’s as if marketing has been handed a larger canvas, but with a spotlight trained on every brushstroke.
ON-DEMAND WEBINAR
Marketing’s 2026 Reality Check: Budgets, AI, and the Regrets No One Talks About
The AI question no one can answer
At the center of this pressure is Marketing AI ROI, a phrase that has become both a promise and a problem.
Everyone wants it. Few can define it.
Part of the issue is that AI has exposed a deeper flaw in how marketing measures success. For years, teams have optimized for outputs—clicks, opens, conversions—because they were easy to track. AI, with its ability to scale those outputs, has only accelerated the pattern.
But scale doesn’t equal value.
“The right answer is an output,” Liz noted. “The outcome is whether that answer actually changed the trajectory of the business.”
It’s a subtle distinction, but a critical one. An AI system that generates better-performing emails is useful. One that helps reshape a product, refine positioning, or unlock growth—that’s transformative.
Too often, organizations celebrate the former and assume they’ve achieved the latter.
Feeding the beast
If AI is the engine, data is the fuel. And here, marketing faces a paradox.
There is more data than ever—campaign metrics, behavioral signals, customer feedback. Yet much of it is static, historical, or repetitive. AI systems trained on this data can optimize patterns, but they struggle to innovate beyond them.
Liz offered a vivid analogy: AI is “a hungry, hungry beast.” And like any creature, it doesn’t thrive on the same meal repeated endlessly.
The problem isn’t volume. It’s freshness.
Without a continuous stream of new, qualitative, human insight—how customers think, why they choose, what they reject—AI simply reinforces yesterday’s decisions. It becomes a mirror, not a lens.
This is where many marketing teams falter. They invest in AI tools but neglect the inputs that make those tools meaningful. They automate before they understand.
The result is diminishing returns: smarter systems producing increasingly predictable outcomes.
The cost of playing it safe
There’s another, quieter consequence of this environment: a growing reluctance to experiment.
Marketing has long depended on testing—trying, failing, refining. But under heightened scrutiny, failure can feel less like a step forward and more like a liability.
“We’re getting scrutinized over trying,” Liz said. “That’s a very weird position to be in.”
The irony is hard to miss. At the very moment when AI enables faster, more efficient experimentation, organizations are becoming less tolerant of the process that drives innovation.
It’s a bit like asking a scientist to guarantee results before running the experiment.
The teams that break through this tension aren’t the ones that avoid failure. They’re the ones that reframe it. They treat experiments not as risks to minimize, but as signals to interpret—data points in a larger system of learning.
Are you an insight seeker?
Join UX, research, and design leaders to push your craft further.
From efficiency to effectiveness
Early conversations about AI in marketing focused heavily on efficiency: faster content creation, automated workflows, reduced manual effort. And those gains are real.
But efficiency alone is a dead end.
“Efficiency always points to more inefficiency,” Liz observed. “The more efficient you are in one task, the more it shows you the ten tasks you’re not doing well.”
The next phase of AI adoption isn’t about doing the same things faster. It’s about doing better things.
That means shifting the focus from efficiency to effectiveness—and ultimately, to growth. It means asking not just how AI improves a process, but how it changes the outcome of that process in a meaningful way.
And it requires a different kind of measurement. Not just what happened, but why it mattered.
Velocity, not speed
If there’s one idea that captures this moment, it’s velocity.
Speed is easy. Anyone can move quickly, even if they’re going in circles. Velocity requires direction—an understanding of where you’re headed and why.
“What feels fundamentally different is that we have to work with velocity, not just speed,” Liz said. “If you don’t have direction, you’re just going fast in a circle.”
Marketing today is moving faster than ever. The risk is that it’s also spinning—chasing trends, adopting tools, optimizing tactics without a clear sense of purpose.
AI doesn’t solve that problem. It amplifies it.
The organizations that succeed won’t be the ones with the most advanced technology. They’ll be the ones with the clearest direction—and the discipline to align their tools, data, and teams around it.
The uncomfortable path forward
There’s no neat resolution to this story. No framework that neatly ties together budgets, AI, and ROI into a predictable formula.
What there is, instead, is a set of uncomfortable truths.
Marketing leaders must ask harder questions. They must tolerate uncertainty longer. They must invest not just in technology, but in understanding—of their customers, their data, and their own assumptions.
And perhaps most importantly, they must resist the temptation to confuse activity with progress.
Because in the end, the difference between motion and momentum is the difference between spinning your wheels and actually moving forward.
Or, as Liz put it: “Don’t confuse an output with an outcome.”
GUIDE
