TL;DR: AI recommendations are driven by five controllable signals: presence in cited sources, content citability, structured data, AI crawler accessibility and consistency of company information.
1. The category source map
Every product category has its "source map" — rankings, media, directories and forums AI cites most often. If a competitor appears in three rankings cited by Perplexity and you in none, you lose before the user even asks.
2. Content citability
AI cites content that is easy to extract: definitions in the first two sentences of a section, numeric data with sources, comparison tables, authorship with bios and update dates.
3. Structured data
Schema.org (Organization, Product, FAQ, LocalBusiness, Review) is the cheapest way to hand AI facts about your company in a format it does not have to guess.
4. AI bot accessibility
GPTBot, ClaudeBot, PerplexityBot and Google-Extended must be able to fetch and render your content. A robots.txt block or JS-only rendering excludes you from part of the answers.
5. Entity consistency
Mismatched data between your site, Google profile, directories and LinkedIn lowers the model's trust in facts about your company — and wrong data in AI answers is a straight path to losing a customer.