Why the Marketing Funnel Must Die (and What Comes Next)

The classic marketing funnel—awareness at the top, conversion at the bottom—no longer reflects how people search, discover, and buy. In 2025, the funnel is breaking apart, and AI is leading the disruption.

According to a SparkToro and Datos study, nearly 60% of Google searches now end without a click. Users are finding their answers directly inside AI Overviews or other generative engines like ChatGPT and Perplexity. This “zero-click reality” means the days of stuffing the top of the funnel with as many people as possible are over.

The new game is intent. Marketers must pivot from blasting broad audiences to recognizing the signals of high-fit, high-intent users—and focusing resources on them.

Zero-Click Reality

• 58.5% of U.S. searches and 59.7% of EU searches no longer result in clicks (SparkToro + Datos).
• AI Overviews now capture 34–37% of clicks, replacing the “rank zero” spots SEO once targeted.
• Bain & Co. estimates 15–25% declines in organic web traffic due to zero-click behavior.

The “open jaw” pattern shows impressions going up but website traffic going down. Visibility is still there, just not in your analytics dashboard.

The Conversion Paradox

Less traffic doesn’t mean less business. In fact, it often means more.

• ChatGPT referrals convert 5–15x better than Google organic traffic.
• HubSpot attributes 13% of new revenue directly to AI-driven leads, which convert at ~3x higher rates.

Why? Because users reaching you after an AI answer are already far down the journey. Instead of “spray and pray,” you’re talking to people who already know what they want.

The New Playbook: Intent-Driven Marketing

Stop interrupting. Start listening.

Traditional funnel math (“300 leads → 100 conversations → 3 buyers”) no longer applies. AI lets you spot the 3 real buyers early by reading engagement signals like:

• Email opens and click-throughs
• CRM behaviors and past deal patterns
• Event attendance or community interactions

This requires new disciplines:

• AEO (AI Engine Optimization): Make your content machine-parsable with clear Q&A blocks, schema markup, and passage-based formatting.
• GEO (Generative Engine Optimization): Optimize for how LLMs synthesize and cite content, not just how search engines rank it.
• Search Everywhere Optimization: Diversify beyond Google. AI leans heavily on Reddit (40.1%), Wikipedia (26.3%), YouTube (23.5%), and other trusted sources.

New KPIs for the AI Era

If clicks are down, how do you measure success?

• Impressions in AI Overviews
• Citations in LLMs (Reddit/Wikipedia mentions, YouTube content, LinkedIn authority)
• Conversion quality, not lead quantity

The Human Advantage

AI is great at automation. Humans are great at judgment.

As HubSpot’s Dharmesh Shah put it: “Automate the mundane, amplify the magic.”

• AI handles admin, note-taking, and research.
• Humans bring taste, curiosity, and strategy.
• The best teams are “hybrid”: humans leading, AI accelerating.

The SCET Design Lab Advantage

At Berkeley SCET Design Lab, we sit at the intersection of innovation, marketing, and technology.

We help startups, labs, and organizations:

• Prototype at loop speed: shifting from cruise-ship campaigns to rapid “jet-ski” experiments.
• Implement AEO strategies: from schema markup to proprietary data publishing that boosts citations.
• Build hybrid teams: where people lead with creativity and AI accelerates execution.

The future of marketing isn’t about bigger funnels. It’s about finding fit, recognizing intent, and combining human insight with AI acceleration. That’s the work we do with our partners every day.