When a buyer asks, the answer names a few brands in your category. Six factors carry most of the weight in whether yours is one of them.
When someone asks ChatGPT for the best CRM for a small business, they get a short list of names and a reason. Those names are not random. What is weighed: who you clearly are, the structured data you publish, corroboration from other sites, and how quotable your pages are. This guide takes the six factors one at a time, shows the order that moves the needle, and includes the free survey that shows which competitor is holding the slot you want. It cannot buy you a citation — the factors are readable and the gap is measurable, which is the useful part.
Which factors decide whether AI cites you?
Six, in our rough order of leverage: organization schema with sameAs, FAQPage and HowTo markup, content depth and structure, external citation patterns, author attribution, and freshness signals.
- Organization schema with sameAs: the clearest identity signal a small site can publish — it tells a machine who you are and links your site to profiles you already own. Google notes that structured data is not required for its generative AI features, so we treat it as a strong helper, not a guarantee.
- FAQPage & HowTo schema: when an assistant answers a "how" or "what" question it can lift a clean question-and-answer pair from a page that publishes one. Google states structured data is not required for generative AI search, so we count this as a helper — not a requirement.
- Content Depth & Structure: AI extracts meaning from heading hierarchy and list structure. Poorly structured pages produce garbled excerpts, so they get skipped.
- External Citation Patterns: Sites linking to authoritative sources signal participation in a legitimate information ecosystem. We score this as a trust signal.
- Author Attribution: In the citations we check, content with a named author and a real bio is cited more than anonymous content. A named author is read as a credibility signal.
- Freshness Signals: Content updated this month starts ahead of content from 2023. Modification dates are read, and we score them.
Why do most scores come back low?
In the audits we run, the same three gaps come up first: no JSON-LD, no sameAs links, and poor heading structure. All of it is fixable in one sitting.