The old playbook is dead. Eyeballs → emotions → clicks was the architecture of a world where humans did their own browsing. In the agentic economy, the buyer isn't human — it's an AI. Your customer's AI shops, compares, and decides for them. Your brand isn't chosen by a person; it's selected by a machine based on cold, hard parameters. The brands that win the next decade will be legible to machines and authentic to humans. Everything else is optimising for a world that no longer exists.
The tactical companion to Letter 66 — The End of the Attention Economy. Four pillars for the AI-first era. Not investment advice.
Letter 66 established the structural case: the $258 billion attention economy was built on the assumption that humans browse, scroll, and click. That assumption is becoming false at speed. When you ask an AI assistant to find the best noise-cancelling headphones, book a restaurant, choose a supplier, or compare insurance policies — the AI does not see your ads. It does not respond to emotional creative. It does not experience urgency or aspiration or social proof. It reads structured data, evaluates parameters, and returns a recommendation.
The implication is not subtle. Every dollar spent on creative that moves humans emotionally is misallocated if the human is increasingly delegating the decision to a machine. Every investment in SEO that ranks pages for human browsers is strategically misaligned if the browser is increasingly an AI agent that synthesises answers rather than returning blue links. The marketing profession is not facing a tool upgrade. It is facing a customer substitution. The customer is being replaced — partially, then increasingly, then dominantly — by an AI proxy that operates on entirely different principles.
You cannot win an AI agent with a Super Bowl ad. You can win it with a structured product data feed, a credible review profile, a competitive price, and an API that responds in under 200 milliseconds.
Search Engine Optimisation was the art of ranking in Google's top ten results so human browsers would click your link. Generative Engine Optimisation is the art of being cited, referenced, and recommended inside AI-generated answers so AI agents surface your brand when a human asks a question you should own.
The mechanism is fundamentally different. Google ranked pages by authority signals — backlinks, keyword density, page speed, domain age. AI models cite sources by credibility signals — factual accuracy, source authority, content structure, data richness, and the degree to which the information is cross-referenced across trusted sources. You cannot game GEO with link-building schemes or keyword stuffing. The only strategy that works is actually being authoritative. This is, paradoxically, what SEO always claimed to reward but rarely did.
Practical GEO: publish structured, deeply accurate content on the topics you want to own. Get cited in sources that AI models weight as authoritative — quality journalism, academic papers, industry bodies. Ensure your Wikipedia presence is accurate and comprehensive. Build schema markup into every page so your content is machine-parseable. Maintain factual consistency across every channel — AI models notice contradictions.
When an AI agent evaluates your product against competitors, it does not feel aspirational about your brand story. It does not respond to packaging aesthetics or celebrity endorsement or the emotional arc of your television commercial. It evaluates the parameters its human principal has specified: price, delivery time, review score, return policy, carbon footprint, product specifications, API availability, and structured data quality.
Your product data is now a competitive asset in the same way your creative used to be. The brand with cleaner structured data, faster API response times, more authentic review profiles, and more machine-parseable product attributes will win the AI agent's consideration set — regardless of how good their advertising creative is. This requires a complete reorientation of where marketing investment goes. The budget that funded the brand film now needs to fund the data infrastructure team.
The specific investments that compound: product data standardisation and enrichment, review strategy that generates authentic customer advocacy at scale, pricing intelligence that keeps you competitive within AI agent parameters, logistics optimisation that improves the delivery metrics AI agents weight, and API-first architecture that makes your inventory and pricing available to agent-accessible commerce platforms.
As AI generates competent content at near-zero marginal cost across every channel simultaneously, the scarcity — and therefore the value — of genuinely original human creativity increases sharply. Consumers are already developing finely tuned sensitivity to AI-generated content. They can feel its uncanny smoothness. They trust it less. They share it less. They remember it less.
The creative hierarchy is bifurcating at speed. At the bottom: high-volume, algorithmically competent content production — entirely replaceable by AI, and the agencies building business models around this are already experiencing margin compression. At the top: strategic creativity, cultural insight, genuine originality, and the kind of creative work that produces real cultural moments — more valuable than ever, commanded by the small number of humans who can genuinely deliver it.
The strategic imperative is not "use AI to produce more content." It is: use AI to produce the same amount of extraordinary content at far lower cost, by concentrating human creative effort at the strategic and cultural level where AI cannot substitute. More content is not the answer. Better content, distributed more intelligently, is. The brand that floods every channel with AI-generated mediocrity is training its audience to ignore it. The brand that deploys human genius sparingly, at culturally resonant moments, is building the kind of attention that AI cannot manufacture.
Third-party cookies are gone. Apple's App Tracking Transparency framework has severely degraded mobile targeting. GDPR enforcement has tightened across Europe. The tracking infrastructure that underpinned two decades of digital advertising precision — following a user across 500 websites, building a behavioural profile, serving contextually relevant ads at the perfect moment — has been structurally dismantled.
What replaces it is not better third-party data. What replaces it is the direct customer relationship — the email subscription, the loyalty programme, the owned app, the community, the repeat purchase history. The brand that has built genuine direct relationships with customers, with consent, at scale, holds the only targeting data that the post-cookie world validates. Everything else is rented data from platforms that can revoke access at any time, for any reason, at any cost.
First-party data strategy requires long-term investment in exchange mechanisms — reasons customers willingly share data because they receive genuine value in return. The loyalty programme that actually delivers. The personalisation that actually improves the experience. The community that actually connects people with shared interests. These are not marketing tactics. They are relationship infrastructure. The brand that owns the relationship owns the future's most valuable marketing asset.
Long-horizon thinking on capital, technology, and the forces shaping the next decade of wealth creation. Written from first principles. Not consensus. Not noise.