A citation surface is any place an AI assistant can retrieve information about your business while it assembles an answer. Most of those places belong to someone else. This chapter covers which surfaces matter, why the set changes by trade and by place, and why most of the work on them is correction rather than publication.
Where do AI engines look when they assemble a recommendation?
Your site is one input among several. An assistant assembling a recommendation draws on directories, review platforms, comparison and roundup sites, forums, local press, association and supplier pages, and your own domain. For commercial queries, a large majority of the retrievable material is content the business does not own.
That reframes the work. A plan consisting entirely of improving your own pages is addressing the smaller part of the surface, and it explains the common experience of a well-built site that assistants never mention.
Is there a universal list of citation surfaces?
The set differs by trade and by place. There is no universal list, and assuming one is how effort gets spent on surfaces nobody consults. The directories that matter for a plumber in Georgia are not the ones that matter for a specialty retailer in the UK.
We identify them per market and per category by looking at what the assistants actually cite when asked the twelve buying questions, which is a measurement problem before it is a content problem.
What does the work on citation surfaces actually involve?
Most of the work is correction, not publication. Wrong specifications on a marketplace listing. An outdated description on a comparison site. A review platform profile nobody ever completed. A supplier page listing a phone number you stopped using.
Unglamorous, slow, and frequently the highest-return work available, because it removes contradictions rather than adding volume.
How do you build a citation trail?
The trail is the path from a general question to naming your business specifically: the mentions, links and corroborating references that let an engine confirm you are a real, established, appropriate answer.
It is built deliberately and it takes quarters, not weeks, because it depends on sources you do not control updating on their own schedule.
Chapter last reviewed 2026-07-26. Platform behaviour changes, so read every chapter of this guide against the date it carries.
Read in order, or jump
Twelve chapters, sequenced the way the work actually runs. The opening four carry the structure; the platform chapters close with two surfaces we choose not to score, and the reasoning for that.
01 How AI search picks local businesses 02 Eligibility: can AI even see your site? 03 Robots.txt: block AI crawlers or allow them? 04 Measuring AI visibility honestly 05 Entity foundations for local businesses 06 Reviews as AI input 07 Citation surfaces: where AI engines look 08 Getting ChatGPT to recommend your business 09 Showing up in Perplexity 10 Showing up in Google AI Overviews 11 Showing up in Gemini 12 Copilot, and why we don't sample itSee where you actually stand
Everything measured in this guide, run against your business, free.
Get my free Visibility Check