Everything long-form on this site, in one place, with what each piece actually answers and how long it is. If you only do one thing, run the free scan first and let your score tell you which guide to read.
Why this page exists, said plainly. The four guides below are Shopify pages rather than blog posts, so the Articles index has never listed them. That is about 5,511 words of the most useful material on the site sitting where nothing links to it as a set. Rather than move four URLs that are already earning, this page is the set. Word counts were measured from the page bodies on 2026-08-20, not estimated.
Paste your URL and get a score for whether an AI assistant can read your site. The score is free and needs no email. The full prioritised fix list is behind a newsletter signup, and that is the only gated thing on this site.
The map. Which guide to read first, what to do after your scan, and what order the work actually goes in. Read this if you do not know where you are.
Generative Engine Optimization in plain English, and how it differs from SEO. What an AI assistant is actually doing when it names a business, and the moves that made one real shop get named.
Four checks you can run yourself with nothing to buy: ask the engines about you, read your robots.txt, test your schema, and read one page the way a machine reads it.
The four free tools to install before you write anything: Search Console, Bing Webmaster Tools, Microsoft Clarity for AI citations, and GA4. Step by step. If you cannot measure it you cannot claim it.
Nineteen straight answers to the questions people actually send me. What cited by AI means, how it differs from ranking, and what nobody can promise you.
The whole method run on my mom's real Shopify store, in the order I ran it: entity, passages, schema, crawler access, catalog hygiene, measurement. It keeps its original figure with its original window, and it carries its own correction log, including the part where the citations sent almost no traffic.
The same shop tags almost every order as direct, which hides real search wins. Includes the fake win I nearly published and the method that costs nothing to run.
The working out, as well as the settled answers. Studies where the result went against me are published with the ones that did not.
Bangermeter, Inkwash, HUMANIST, Assay and Candor. All MIT licensed, all on GitHub, and every claim they make is checkable at the source.
Everything above is free and it is the same work I do for money. If you would rather not run it yourself, the eight services are billed by the hour at $75 with a one hour minimum, and you can buy hours without a sales call.