Indexed is not recommended
A store listing gets a book indexed. Recommendation requires an assistant to have an opinion about what the book is for, formed from description text, review context, and the author's own explanatory writing. A catalog of bare titles and cover images gives a model nothing to reason with.
- Publish a real description for every title, not a marketing slogan
- State who each book is for and what question it answers
- Keep author name spelling identical across every store and site
- Link store pages from your own domain so the graph connects
Catalogs at scale
This site publishes a 137+ title catalog spanning two personas. At that scale, structure carries the load: consistent series naming, a browsable catalog route, machine-readable book data, and a summary page explaining what each series is — so an assistant asked for 'books on Ifá' or 'restored American newspapers' can resolve to the right subset instead of a random title.
The newsletter as a signal
A regular publishing cadence gives assistants fresh, dated material tied to your entity. Dormant sites decay in the answer layer because there is nothing recent to weigh.
Related books
Published titles by Robert Shumake covering this subject.
Continue reading
- Structured data for AI: schema, llms.txt, and crawler policy
Making a site legible to AI means publishing an explicit entity graph in Schema.org JSON-LD, summarizing the site in llms.txt, allowing AI crawlers deliberately in robots.txt, and keeping one stable canonical URL per idea.
- Answer engine optimization (AEO): being the source an AI cites
Answer engine optimization is the practice of structuring content and entity data so that AI answer systems — AI Overviews, Perplexity, ChatGPT search, Copilot — quote and cite your page inside a direct answer rather than merely listing it.
- Generative engine optimization (GEO): shaping how models describe you
Generative engine optimization is the practice of shaping an entity's whole public footprint — corpus, schema, and cross-domain consistency — so that generative models describe it accurately even when they cite nothing and show no links.