Essays built on original data — what actually moves growth, and how to stay visible as AI rewrites search.
The field guide: what GEO actually is, how AI answers get built, GEO vs SEO vs AEO, and a working glossary — from an operator with 192,924 AI citations of proof.
Read the field guide →Twelve weeks, 192,924 AI citations, one real website. What actually makes ChatGPT, Perplexity, and Google's AI cite you — and why most of what the new “GEO experts” are selling didn't survive contact with the data.
Read the field notes →The nine moves that actually earn citations from ChatGPT, Perplexity, and Google's AI — with an interactive readiness checklist that scores your page as you tick them off.
Read the playbook →The one thing nobody tells you — ChatGPT reads Bing, not Google — the six moves that earn citations, and how to check if it cites you.
Read the playbook →Same foundation, different unit of competition — SEO earns a ranking, GEO earns a citation. What transfers, what breaks, and how to split your effort.
Read the comparison →The retrieval mechanic that quietly runs every AI search engine: one prompt becomes 8–16 hidden sub-queries — and you get cited for searches nobody ever types.
Learn the mechanic →The three jobs a GEO tool actually does, what each tier honestly costs — Otterly, Peec, Profound, LumenGEO & the suites — and the $0 stack that covers most teams.
Compare the tools →The first-party case study: 87 experiments, 25,000+ users in 4 months, zero ads or PR — what on-site GEO actually did, and the honest caveats about what the data does and doesn't prove.
Read the case study →The numbers that matter for GEO — AI Overviews reach, citation patterns, what actually moves visibility — every stat attributed and linked. Built to be cited.
See the stats →MMM, attribution, and incrementality each answer a different question and carry a different bias. The operator's move isn't picking one — it's giving each a job and letting them calibrate each other. With an interactive measurement triangle.
Read the field guide →You don't need a six-figure vendor — or the quarterly refresh — to know which way budget should move. How to model the mix scrappily, fast, and without fooling yourself. With the rigor ladder.
Read it →A point forecast is a wish with a decimal. Triangulate three models that should disagree — bottoms-up funnel, top-down trend, and cohort/LTV — into a defensible range, then track your error so it compounds.
Read it →Real incrementality tests are scarce. Choose them like a portfolio — scored by decision value × uncertainty ÷ cost — and run the agenda as the maintenance schedule for your whole measurement stack.
Read it →One user is three identities — web, iOS, Android — measured by three systems that never agree, under privacy rules that keep moving. The 5-layer stack that holds cross-platform attribution together, with an interactive channel × platform matrix of what's actually measurable.
Read the playbook →0→1 isn't scaled-down 1→n: no budget, no brand, no team. The honest sequence for finding the one channel that works — with GrantCompass (25,000+ users, $0 ads) as the measured-not-caused case, and a first-90-days plan.
Read the playbook →Everyone has the same models — the gap between teams isn't access to AI, it's whether they build with it. The 5-level AI-native maturity ladder, where AI really creates leverage, and what it still can't fix.
Read the field guide →Positioning isn't a tagline — it's the upstream choice that makes every channel and message easy or impossible. A startup positioning framework, the wedge, and a live positioning-statement builder.
Read the field guide →Set a saturation curve per channel and watch the optimizer split a fixed budget to maximize output — the same logic an MMM uses to allocate spend. Interactive, illustrative, free.
Open the tool →Paste any URL and get a 0–100 score for how likely ChatGPT, Perplexity, and Google's AI are to cite it — with the exact fixes, ranked. The automated companion to the playbook checklist.
Scan your page →