The Mad Men of Marketing Are Back (And They're Wearing Sneakers)

Andrew Holland spent 17 years catching criminals as a police officer before he started catching something arguably more elusive: the attention of AI systems. Now, as Director of SEO at JBH, a UK-based digital PR agency, he's become one of the most provocative voices arguing that the SEO industry has been measuring the wrong things for years.

His thesis is simple, uncomfortable, and increasingly hard to ignore: the era of content-as-traffic-machine is collapsing, and what's replacing it looks a lot like Madison Avenue circa 1962.

The Traffic Illusion

Here's the thing about digital marketing in 2026: we've spent two decades optimizing for a metric that was always a proxy for something else. Traffic became the goal because it was measurable, not because it was meaningful. Holland has been writing at Search Engine Land about this disconnect for years, but the AI revolution has turned his contrarian position into something closer to consensus.

The numbers are brutal. BrightEdge data shows AI Overviews now trigger on approximately 48% of all tracked search queries, up 58% year-over-year. Seer Interactive's research found organic CTR plummeted 61% for queries where AI Overviews appear. Google search referral traffic to publishers declined globally by about a third in the year to November 2025.

The old model worked like this: publish content, capture keywords, drive traffic, call it growth. The problem? Traffic was never the product. For most B2B companies, content was supposed to be a vehicle for positioning, not a destination. Somewhere along the way, we forgot that.

SEO Doesn't Own AI Search

Holland's most controversial claim, and the one that's earned him both admirers and critics, is that SEO doesn't own AI as a marketing channel. Copywriting and digital PR do.

All the data, every shred of it, shows that SEO doesn't own AI as a channel. Copywriting and digital PR do.

Andrew Holland

This isn't just professional turf warfare. It's a fundamental reframing of what visibility means when the interface between your brand and your customer is a language model, not a search results page. When ChatGPT or Perplexity recommends a solution, it's not ranking documents by keyword density. It's synthesizing reputation, proof, and positioning from across the web.

Holland calls this fame engineering, a term that sounds like marketing jargon until you realize what he's actually describing: the systematic construction of brand presence across enough touchpoints that AI systems recognize you as the obvious answer to a buyer's problem.

The GEO Divide

The industry has been wrestling with terminology. GEO (Generative Engine Optimization), AEO (Answer Engine Optimization), AI SEO. Holland's position is that the naming debate misses the point. In a video interview with James Dooley

SEO equals rankings. GEO equals AI recommendations. Consumer survey data will reveal the impact of AI on behavior. I predict it's big.

Andrew Holland

The distinction matters because it changes what you optimize for. Traditional SEO rewards technical execution: page speed, structured data, keyword placement. GEO rewards something closer to brand marketing: consistent messaging, third-party validation, presence in the conversations AI systems are trained on.

Semrush data shows that roughly 60% of searches now yield no clicks. When AI Mode is present, that number climbs to 93%. The click, that fundamental unit of digital marketing measurement, is becoming an endangered species.

The Mad Men Parallel

Holland's recent piece at Search Engine Land draws an explicit parallel to the advertising industry of the 1950s and 60s. Back then, agencies grew brands through persuasion, positioning, and earned trust in a world of scarce media channels and powerful gatekeepers. If you wanted attention, you bought your way in, then made your product the obvious choice.

AI is exposing everything SEO has neglected. Brands that win recommendations from AI systems won't do so by publishing more content. They'll win through positioning, persuasion, and corroborated proof.

Andrew Holland

This is where the sneakers come in. The modern version of Don Draper isn't wearing a three-piece suit and drinking Old Fashioneds at lunch. He's probably in a Henley and dark jeans, scrolling through AI citation reports and wondering why his competitor keeps showing up in ChatGPT's recommendations while his brand remains invisible.

What This Means for B2B Marketing

For marketing executives, Holland's framework suggests several uncomfortable pivots:

Stop measuring traffic as a primary KPI. E-commerce sites reported a 22% drop in search traffic due to AI-generated suggestions. The visitors who do arrive are more educated, more qualified, and closer to a decision. Volume is down; intent is up.

Invest in digital PR like it's 2006. Holland points out that 90% of SEO agencies don't have PR staff, and most retainers aren't expensive enough to fund good digital PR. Yet PR is what builds the third-party validation that AI systems use to determine credibility.

Think about AI availability, not just visibility. Holland's concept of AI availability asks a different question than traditional SEO: not "can people find us?" but "will AI recommend us when it matters?" That requires presence across media, content, and communities, not just your own website.

Treat your brand like a media company. In his piece on traffic diversification, Holland argues for world-building, creating multiple entry points (newsletters, podcasts, books, interviews) that allow people to discover your brand and become part of your ecosystem.

The Uncomfortable Truth

Holland's work is gaining traction because he's saying what a lot of marketing leaders have suspected but couldn't articulate: the content marketing industrial complex was always a bit of a shell game. We measured what was easy to measure, built dashboards that showed lines going up, and called it strategy.

The AI disruption isn't killing marketing. It's killing the parts of marketing that were never really working in the first place. The brands that built genuine authority, that invested in positioning and proof rather than just keyword volume, are finding that AI systems recognize and recommend them. The brands that optimized for traffic without building trust are discovering that their content was always just noise.

Marketing is like dating, as I've said before. You don't propose on the first ad impression. But you also don't get a second date if you spent the first one reciting keywords and hoping the algorithm would notice.

Holland's message to the industry is essentially this: the algorithm noticed. It just wasn't impressed.