94: Are LLM Visibility Audits Total BS? (Emily Zertuche)

Every vendor in the industry suddenly has an "AI visibility score" to sell you. Stuart Butler and Adam Stoker dig into whether GEO (Generative Engine Optimization) is real measurement or an expensive mirage, joined by repeat guest Emily Zertuche, CDME, founder of Zertura (formerly Visit Corpus Christi), in her first episode as a vendor.The trio unpacks a Fast Company piece questioning whether GEO is real. Emily explains why most "visibility scores" fall short of basic statistical rigor - testing a prompt three times isn't a sample, it's noise. She cites IAB's May guidance acknowledging there's no real standard yet for measuring non-deterministic LLM outputs, and lays out what rigor actually looks like: repeated sampling, confidence intervals, margin of error - not a single score.Stuart argues the industry chases feel-good numbers that don't move demand or economic impact, while Adam pushes back that writing off an entire vendor category discourages the experimentation destinations need. All three land on a better path: quality scores, sentiment tracking, net promoter score for visitors and residents, and treating visitor experience as the real driver behind citations and LLM recommendations.Bottom Line: There's no silver-bullet AI visibility score today. Ask vendors to show their methodology and confidence intervals before you buy in, and keep measuring real experience and sentiment - that's what actually earns citations.