How to help your marketing team thrive with AI

Last Sunday, I was listening one of my favourite podcasters, Nathaniel Whittemore (The AI Daily Brief), who gave an obvious advice, but bluntly put and good: For teams to thrive in AI removing tedious work is the first step, however achieving work that was out of reach before is the real game changer.

The obvious success criteria of marketing AI use cases are hours saved, cost reduction or faster campaign cycles.

All of it is real and worth having.

But a business case built on "saved hours" will never tell you what your team could become…

I would start by asking this question: "What can my team do now that it could never have done before?"

Here are the three answers I would give, and a four-step plan to act on them.

1. Rehearsing the campaign before spending the budget: Your market simulation

Marketing has never had a rehearsal room. We planned, we tested a concept with a focus group if the budget allowed, and then we went live and found out. The learning happened in flight, and by then the money was spent.

Market simulation gives us that rehearsal room. You build a simulated version of your market, with your segments, your competitors and your price points, and you run the scenario before committing to the media plan. What happens to share if we cut the discount? What does a likely competitor response do to the launch?

2. Interviewing a thousand customers in one go: Your AI moderator

Customer research has always forced a choice between depth and scale. You could have eight people in a room and real listening, or ten thousand surveys and tick-box answers.

An AI moderator can hold an unscripted conversation with a thousand real customers in a week, notice the unexpected answer, and ask the follow-up a good researcher would ask. Researchers writing in Harvard Business Review (April 2026) describe teams going a step further and keeping a study permanently open, so listening becomes part of the operation instead of a project with an end date.

3. Persuading a buyer who is not human: AI assistants

This one still feels strange to write. A growing share of the visitors arriving at brand and retail sites are sent there by AI assistants, because customers now ask their assistant to research, to compare, and increasingly to buy. Adobe Analytics measured AI-referred traffic to retail sites growing 393% year on year in early 2026, and these visitors convert at a higher rate than the average visitor once they arrive. Persuading a machine that shops on a human's behalf is a new marketing job, and nobody had it in 2023.

How ready are marketing teams for this?

Not very, if the adoption data is honest. Section's AI Proficiency Report (July 2026) surveyed over 5,000 knowledge workers: 69% say their organisation has taken action on AI agents, yet only 16% use the tools they were given. The tools are everywhere, and the capability to use them is still rare.

Which is why the leader's job here is bigger than tool selection. I have laid out a four-step plan to point a busy team above the floor.

Step 1. Write the out-of-reach list.

Ask every team lead one question: what did we say no to in the past year because of budget, headcount or time? The answers will be the research that never got fielded, the segments that never got a campaign, and the launch questions left to the post-mortem. That list, and nothing else, is your brief.

Step 2. Match one item to one of the three jobs above.

Unanswered launch questions point to market simulation. Unheard customers point to AI-moderated interviews. And if your analytics already show assistant-referred visitors, start by making your site readable to the machines doing the shopping.

Step 3. Pilot with the ground rules set first.

Decide who owns the output, what the threshold is, and when a simulated or AI-collected result is strong enough to act on.

Step 4. Reinvest the proof.

When the pilot works, show the wider team what it did and how it worked.

The next two years

The teams that win the next two years will run two lists side by side: the efficiency list that funds the operation, and the out-of-reach list that grows it. The teams that lose will be very busy, very fast, and exactly where they started.

Share this with one marketing leader whose AI business case still stops at saved hours.

← All insights