How to be in the consideration set on day one. The committee query map, the citation audit, machine-readable trust, the source graph, and the paid-media flywheel that feeds AI discovery.
84-page A4 PDF · $149 · 14-day refund
Most B2B content plans still assume the buyer’s first touch with a vendor happens on the vendor’s website. In the analytics we run for B2B vendors, that stopped being true some time ago. A few years back roughly 30% of account-based first touches on vendor sites landed on educational content. That share has collapsed. The first thing an identified account does now is look for proof: pricing, a demo, a comparison, a security page. The learning still happens. It happens inside ChatGPT, Gemini, Perplexity and Google’s AI Mode, where you can’t see it.
The market research says the same thing from the buyer’s chair. G2’s 2026 survey of 1,076 buyers found 51% now start research in an AI chatbot more often than Google, and AI chatbots have become the number one influence on the shortlist. 6sense’s study of nearly 4,000 buyers found the winning vendor was on the shortlist on day one 95% of the time. Forrester puts 13 internal stakeholders and nine external ones on the average purchase, each asking the engines different questions in different words.
This playbook is the method for dealing with that. Not “optimise for AI” as a slogan, but a repeatable way to find out, question by question and role by role, whether an engine names you, what it says, who it cites instead, and what to do about it.
What’s inside (84 pages, A4 PDF)
Part One explains how answer engines actually decide what to cite: the two-memories model, query fan-out, passage-level selection, and what the 2026 citation studies say each engine prefers.
Part Two is the audit. The Committee Query Map, a 7x5 grid of buying-committee roles against decision moments, with a worked example. The citation audit protocol, with a logging schema, six metrics, variance rules and a gap register. The AI Visibility Scorecard, seven weighted metrics and a committee heat grid that fits on one page.
Part Three builds the signal: the twelve proof pages a vendor needs, the citeable-unit writing standard, crawler and robots.txt policy, an honest position on llms.txt, and how to earn a presence on Reddit, LinkedIn, YouTube, the review sites and the trade press without doing anything you’d be embarrassed to explain. Plus creators, executives and employee advocacy designed for retrieval rather than reach.
Part Four is the angle you won’t find in SEO-led guides. The factors most strongly correlated with AI visibility are brand mentions, branded search and third-party conversation, and those are exactly what a brand-to-demand media programme produces. How to run paid media upstream of the engine, what each surface contributes, the real state of ads inside ChatGPT, AI Mode and Copilot as of late 2026, and how to use account graphs and intent data to decide which committee questions to win first.
Part Five covers measurement (the AI referrer regex, log analysis, a three-tier dashboard) and the operating model (roles, RACI, a 90-day plan).
Part Six is the toolkit: three teardowns, 40 prompts and nine templates.
Who it’s for
Content, SEO and growth leads who need a method rather than a mood. Demand and ABM teams who want AI discovery connected to accounts. Paid media teams who suspect the money they already spend is doing more for AI visibility than anyone gives it credit for. CMOs who need a scorecard that survives a finance review.
Who it’s not for
Anyone looking for a list of hacks. There’s a whole chapter on why the hacks don’t work.
How the evidence is handled
Every number carries a label: Measured (a study with a published method), Reported (a vendor or press figure), Our data (FunnelFuel’s analytics), or Working rule (a judgement from doing this). 34 numbered sources. Where I don’t know, the book says so.
Subscribers get $30 off
Subscribing to The B2B Stack is free, and subscribers get $30 off the playbook and the bundle. Your code is in the Subscribers’ Toolkit, along with the free report and the other subscriber tools.
The companion: The B2B Signal Optimisation System
The playbook gets you named in the answer. The System scores what those buyers do when they land: a 35-page playbook with the 100+ high-value actions mapped to funnel stage, scoring templates, a GTM setup guide and an example dashboard. It’s $97.
Want both? The playbook and the System together are $199 as a bundle.
About the author
Mike Harty has been building B2B programmatic platforms since 2011, first at PowerLinks and now as co-founder of FunnelFuel, a B2B advertising specialist built on a living account graph. He publishes The B2B Stack, read by teams at dentsu, WPP, Informa, Salesforce and the large vendors and agencies who run B2B media for a living.
Updates
Platform details change quarterly. Buyers get every revision of this edition free.
Refunds
If it isn’t useful, email within 14 days for a full refund. No form, no questions.
Mike