I spent more than a decade in global advertising at Leo Burnett and Saatchi & Saatchi, working on brands people buy in a supermarket aisle in seconds. Then I moved into deep technology: AI infrastructure, data platforms, cybersecurity, and products whose buyers take months to decide.

People assume those worlds have nothing in common. In my experience, they have everything in common.

Lesson 1: People buy feelings and justify with facts

In consumer marketing, nobody pretends a shopper is purely rational. We study the emotional trigger first, then give the rational reasons that make the choice feel smart.

Enterprise buyers are no different. The head of data isn't only buying throughput. They're buying relief from being the bottleneck, and credibility with their CFO. When tech marketing leads with specs, it answers a question the buyer hasn't emotionally asked yet.

Lesson 2: One idea, said many ways

Great campaigns are built on a single, simple idea that can live in a TV spot, a shelf wobbler and a sampling event. Deep-tech companies often try to say ten true things at once and are remembered for none of them.

For an agentic AI product, the discipline is the same. Find the one sentence your buyer should repeat to their boss. Then let your content and outreach agents express it a thousand ways, for a thousand accounts, without losing it.

Lesson 3: Insight is research, not guesswork

Agencies invest heavily in understanding the consumer before a word of copy is written. Many startups skip this step because they are close to the product and assume they know the buyer.

Today we can do the research faster than ever. Agents can analyze calls, reviews, forums and win-loss notes. But someone still has to read the patterns with a trained eye and ask what's really going on.

Lesson 4: Trust is the product

A consumer brand lives or dies on trust built over years. Agentic AI asks buyers to hand real work to software, which is an enormous act of trust. Your marketing has to earn it: clear language, honest limits, proof from peers.

The technology is new. The human on the other side is not.