Guide
To get cited by ChatGPT, your brand needs to appear in the sources it retrieves and trusts — authoritative third-party pages, listings, and reviews — while your own pages answer buyer questions in plainly extractable language. Here is the seven-step process we use with clients, applicable to Perplexity and Google AI Overviews as well.
The process
List the real, full-sentence questions buyers in your category ask AI tools — not keywords. "Best CRM for a 5-person real estate team" behaves differently from "CRM software". Cover the whole journey: discovery, comparison, and validation questions.
Ask ChatGPT (with search enabled), Perplexity, and Google AI Mode each question and note which sources are cited and which brands are named. This is your citation baseline — the sources on that list are the ones you need to appear in.
Restructure key pages so the direct answer appears in the first one or two sentences, one question per page. Add FAQ, Article, and Organization schema in static HTML — structured data rendered only by JavaScript is invisible to most AI crawlers.
Check robots.txt allows GPTBot, OAI-SearchBot, PerplexityBot, and Google-Extended (if you want AI Overviews inclusion). Add an llms.txt file summarizing who you are and what your key pages cover.
Work down your citation baseline: get accurately listed in the directories, review platforms, comparison articles, and communities the engines cited. An accurate unlinked mention in a trusted source beats a backlink from a page AI never retrieves.
Make your name, offer, pricing, and location read identically across your site, listings, and profiles, and refresh important pages at least quarterly — AI engines demonstrably favor recently updated sources.
Citations shift as models and indexes update. Re-run your query set on a schedule, log who is cited, and treat it as a trendline rather than a one-time score.
Everything above is doable in-house if you have the hours. The step teams most often get stuck on is the second one — building an honest citation baseline — because it means running dozens of queries across multiple engines and cataloging the results by hand. That baseline is exactly what a Lumecite GEO audit delivers, along with the prioritized gap list of which sources to go win. If you're newer to the topic, start with what GEO is and how it differs from SEO.
Most brands that do the work see first citations within four to eight weeks. On-page changes like schema and answer-first rewrites can be picked up within days of a recrawl when ChatGPT browses live, while earning third-party mentions and having them propagate takes one to three months. Citations tied to model training data move slowest, on the cadence of model updates.
No. There is currently no way to pay OpenAI, Perplexity, or Google to be recommended inside organic AI answers. Citations are earned through retrievability, third-party corroboration, and consistency. Anyone selling guaranteed AI placements is either doing legitimate GEO work under a bolder label or overpromising.
Almost always because your competitors appear in the sources ChatGPT retrieves from — comparison articles, review sites, directories, and forums — and you do not, or because their pages state what they do in plainly extractable language while yours bury it. A citation audit that maps exactly which sources get cited for your buyers’ questions will show the specific gaps.
Both. ChatGPT answers from its training data by default and runs live web searches when the question needs current information, citing sources when it browses. Training data determines whether the model already "knows" your brand; live retrieval determines whether you get cited with a link. GEO improves both: consistent presence across the web feeds future training runs, while extractable pages and trusted mentions win live citations now.
Get started
Send us your site and the questions your buyers ask. We'll show you where AI tools cite your competitors today, and exactly what it takes to get your brand named instead. No pitch, no obligation.