AI Is Forcing Marketers to Redesign Their Role, Rethink What Truly Matters, and Let Go of What Machines Now Do Better
The environment in which marketing operates has fundamentally changed. We are entering a period of profound redefinition — arguably a revolution.
Brands still matter. Creativity matters more than ever. And customers remain firmly at the center. What has changed are the conditions under which people discover, evaluate, and choose. And, importantly, the ways marketing teams can now operate within those conditions. Agentic AI has opened new paths for marketers, allowing us to create and operate with at an unprecedented level of speed and intelligence.
This is not an incremental shift. It is a structural one. It’s changing how marketing teams work, how content is designed, how software creates value, and where human judgment still matters most. And it demands a different kind of thinking and operating from marketers.
My perspective is shaped by conversations with marketing leaders across the industry, and insights shared at recent events, including GEO in Comms by Notified, the Marketing Brew event, and many candid discussions happening with clients and fellow marketing and communications leaders trying to make sense of what comes next.
Marketing’s Agentic Turn
Marketing is moving from AI as a tool to AI as a system of agents. We are at the inflection point where marketers are moving beyond experimenting in pilots and beginning to deploy agentic AI in more meaningful, scalable ways. This shift will redefine the marketer’s role over the next 6–12 months.
The fastest-moving teams are already making that transition. They’re shifting away from task execution toward system orchestration, using AI agents to automate repetitive work, surface insights, and act across channels—with humans remaining in the loop for judgment, creativity, and trust.
A practical way to approach this shift is through a crawl, walk, run progression.
Crawl starts small and personal: using AI to take one repetitive or frustrating task off your plate, running it manually, and building trust in the output while maintaining full human control. Walk begins when those early wins are automated and scheduled. AI starts running in the background, integrated into existing workflows and producing consistent outputs like reporting, summaries, or content repurposing, while humans shift from doing the work to overseeing it. Run is when multiple agents operate together across tasks and channels, monitoring performance, surfacing insights, and taking action at scale. At this stage, marketers are no longer focused on execution, but on orchestration, setting priorities, applying judgment, and ensuring creativity and trust remain distinctly human.
Flipping Content Strategy: From Brand-Out to Answer-In
Discovery no longer works the way it used to. Zero-click search is increasingly being adopted. People ask questions and receive synthesized answers without ever visiting a website. Increasingly, those answers are not just informational; they are directional. They influence decisions before a human ever considers clicking through.
If answer engines are forming decisions earlier, content can no longer be designed primarily as a vehicle for brand messaging or downstream conversion. It has to meet people at the moment they are trying to understand something, compare options, or make sense of a choice.
Traditionally, the content strategy began with the brand: What do we want to say? How do we position ourselves? Today, it has to begin with the audience: What questions are people actually asking when they prompt LLMs?
This flips traditional content strategy.
Looking closely at prompts, how questions are phrased, what intent they signal, what uncertainty or urgency they express, becomes the compass. Marketers must think in pockets of understanding: clusters of questions that reveal how people think, decide, and buy. And then, let that inform their content strategy. One that is not static, but one that is continually evolving as intent shifts and new questions emerge.
Similarly, campaigns haven’t disappeared, but they no longer anchor discovery. What anchors discovery now are patterns of intent: how people phrase uncertainty, comparison, evaluation, and readiness. Questions reveal far more than personas ever did. They show what someone is wrestling with and how close they are to making a decision. AI systems recognize these patterns quickly, and they reward content that aligns cleanly with them.
In this environment, the marketers gaining ground aren’t speaking louder, they’re listening better. They’re studying prompts closely and using those signals as a compass for developing content and building campaigns.
We’re watching marketing move from the age of the Big Idea (one perfect campaign, locked for a year) to the age of many small generations: continuous iteration, constant learning, and adaptive execution. Ideas are no longer scarce. Judgment is.
The SaaS Apocalypse: When Software Stops Being the Story
For a long time, SaaS gave marketers something solid to work with.
There was a product. It had features. It had a workflow. Marketing’s job was to explain what it did, why it was different, and how customers should adopt it. As part of their tech stack, marketers had to quickly learn and adapt to it—learning new tools, bending processes around platforms, and operate within rigid stacks.
That assumption no longer holds.
Agentic AI is quietly undoing the idea of software as a fixed thing. As AI systems assemble workflows dynamically, pulling together tools, data, and actions based on intent. Software stops behaving like a product you learn and starts behaving like something that adapts to your needs. Two people can use the same platform and experience entirely different outcomes, paths, and value.
For marketers, this is a meaningful shift.
When software becomes fluid, feature-led narratives lose their grip. It becomes harder to anchor messaging around a static set of capabilities, because that’s no longer how value is experienced. What matters instead is whether the system solves a problem in context, at the moment it’s needed.
The so-called “SaaS apocalypse” isn’t about collapse. It’s about disappearance of edges. Products blur into experiences. Features dissolve into outcomes. And marketing shifts from explaining software to shaping how it is understood, experienced and selected.
What Shouldn’t Be Automated (Yet)
As agentic AI becomes more capable, the instinct is to automate aggressively. But one of the clearest lessons emerging from teams already working with these systems is that speed without judgment erodes trust. AI should not run unchecked, not because it isn’t powerful, but because marketing operates in environments where brand, creativity, values, money, and reputation are always on the line.
Trust is built by designing systems where humans retain control over the decisions that matter. The most effective teams are intentional about keeping people in the loop, especially before content is published, budgets are committed, or messages go out into the world. Guardrails around brand, tone, and compliance aren’t constraints; they’re what make scale sustainable.
At the same time, this doesn’t mean creativity is being squeezed out. In practice, the opposite is happening. By automating the mechanical work—reporting, slide building, versioning, optimization—agentic AI is creating space for the work marketers are constantly struggling to get ahead of: strategy, narrative, big creative ideas, and perspective. This is a shift from production to purpose, from filling channels to shaping meaning.
One boundary remains clear. Human connection cannot be automated. At least not yet. In person events, relationship building, deep collaboration, and trust-based work only become more valuable as AI-generated noise increases. As content scales effortlessly, presence stands out. As automation accelerates, authenticity is a luxury.
The goal is to be deliberate about where automation belongs. Humans remain essential where judgment, trust, and connection are the currency. The strongest marketing systems are designed around that division of labor, by choice, not by accident.