No Hacks

The episode delivers substantial, non-obvious ideas packed throughout: the counterintuitive finding that lower domain authority sites get more ChatGPT citations, the 2.3% LLM recommendation rate as a maturity model, the distinction between citations/mentions/recommendations, and the critical gap in agentic browsing tooling. However, there are stretches of softer conversation and some repetition of themes that dilute density slightly.

“websites earning the most Chat GPD citations are the ones which, the lower domain authority, which is the opposite of what every mainstream tool has been telling you for a decade”

“LLMs recommend a brand only 2.3% of the time”

The defensive vs. offensive framing for AI visibility is fresh and contrarian to the mainstream chase for AI search rankings. The brand accuracy audit as a concrete methodology is novel. However, some framings echo earlier Seer thinking (Will Reynolds episode mentioned), and the core point about measuring brand outcomes vs. vanity metrics, while sound, is not entirely new to sophisticated practitioners.

“everything we just spent the last 10, 15, 20 years learning is now doing you a disservice”

“Citations are akin to being on page two or page three of Google, a leading indicator, not a business outcome”

Alisa Scharf is exceptionally well-calibrated for this topic: Chief AI Officer at a major SEO agency with direct responsibility for AI practice across all client accounts, conducting research at scale (15,000 prompts, 500k+ pages), and actively advising on strategy in real-time. She speaks from hands-on implementation, not theory. Her willingness to downgrade popular metrics (citations) publicly demonstrates practitioner credibility, not career-guest opportunism.

“chief AI Officer at Seer Interactive”

“Her research at scale 15,000 prompts and more than a half a million pages”

Strong specificity throughout: named studies with concrete numbers (2.3% recommendation rate, 15,000 prompts, 500k+ pages), specific examples (Supabase/Claude Code agent provisioning, Seer losing 80% organic traffic, agencyspotter.io citations), domain authority ranges (20-40 vs 80-100), and real tools mentioned (CHEQ, SparkToro). Few hand-wavy claims; most assertions are anchored to research or client work. Minor deduction for some abstraction around internal politics and culture fit discussions.

“LLMs recommend a brand only 2.3% of the time”

“We're still heavily associated with San Diego. And then there's another category or degree of I hate to call it misinformation, but at a certain point it kind of is that”

Slobodan asks sharp, clarifying questions and pushes back productively (e.g., 'what does that mean to you?' framing, probing on failure modes between enterprise and startup, challenging the tooling gap). Good follow-ups and willingness to disagree gently (e.g., on the importance of agentic readiness). However, some questions feel softer or more affirmational than truly challenging; Alisa's claims about metrics or strategy are rarely interrogated hard. The host's own predictions sometimes derail focus from the guest.

“What's the first question you fire back?”

“What are the failure modes? What's the difference in those failure modes that you notice?”