If AI engines do not cite your business when buyers ask in your category, the root cause is almost always one of five structural issues — and four of them ship the same week. This article is the AdMax diagnostic shortlist. Run through it in order. If you find your problem inside ten minutes of reading, fix it yourself; the playbooks are linked. If you find three of them, book the $25 audit and we will ship the whole stack.
The single most common cause. If your homepage and cornerstone pages do not have JSON-LD schema blocks — Organization, WebSite, Service, FAQPage, Article — the AI models cannot confidently identify what your business is. They treat the page as ambiguous text and either skip you or cite competitors with cleaner markup.
Diagnostic: view-source on your homepage. Search for application/ld+json. If zero blocks appear, this is your problem. If one generic block appears, this is still likely your problem.
Fix: add Organization + WebSite blocks to your root layout. Add Article schema to every blog post, FAQPage schema to every FAQ block, Service schema to every service page. We cover the exact blocks in our citation playbook.
AI models cite brands they have seen mentioned in places they already trust — trade press, podcast transcripts, GitHub READMEs, association directories, Reddit threads, Wikipedia. If your brand only appears on your own domain, no model has reason to elevate you above a competitor with three trade press mentions.
Diagnostic:search ChatGPT for "[your brand name] [your category]" and ask it to cite sources. If the only source is your homepage, you have no off-site presence.
Fix: pitch three podcasts in the next 90 days. Publish one explainer on a public repo or directory. Get one mention in a category trade publication. Five mentions across five trusted sources usually moves the citation rate measurably.
The third most common cause. Your pages tell your story, but they do not answer a buyer question. AI engines extract direct answers in 40 to 80 word paragraphs. If your pages do not contain answer-shaped paragraphs immediately under question-shaped H2s, the model has nothing to lift into its synthesis.
Diagnostic: open your homepage. Are any of your H2s a question? Is the first paragraph below it a direct, standalone answer? If no on either count, this is your problem.
Fix: rewrite the top three pages so each H2 is a buyer question and each opening paragraph is a 40 to 80 word direct answer. The rest of the section can be your usual depth.
llms.txt is a small Markdown file at the root of your domain that tells AI crawlers what your site is, who you serve, and which pages summarize the brand. Anthropic-adjacent crawlers (Claude, ChatGPT in some modes) explicitly look for it. If you do not have one, you are skipping the easiest declaration the LLMs accept.
Diagnostic: open yourdomain.com/llms.txt. If you see 404, you do not have one.
Fix: publish a five-line llms.txt today. Format is documented at llmstxt.org. Five minutes of work.
AI engines pull entity information from knowledge graphs — Wikipedia, Wikidata, Crunchbase, LinkedIn Company. If your business has no entry in any of them, the model has no authoritative description to anchor on, and it will pick a competitor that does.
Diagnostic: ask ChatGPT to describe your brand. If the description is generic, vague, or wrong, your knowledge graph trust is low.
Fix: claim and complete your LinkedIn Company page. Submit to Crunchbase if eligible. Add Wikipedia and Wikidata entries if you have third-party press to cite. Make sure every profile uses the same description, the same logo, and the same URL.
The five causes are independent. You can have all five at once. Fixing one in isolation rarely moves the needle; fixing three or more usually does. If you read this list and recognized your business in three or more of the diagnostics, the $25 audit ships the entire stack in 48 hours.
Want AdMax to do the AEO and GEO work for your brand? Book a thirty-minute call with a senior strategist.