You know what AEO is. You've added FAQs, installed schema, tidied up your Google Business Profile — and yet when you ask ChatGPT who the best provider in your category is, someone else's name comes back.
Here's the part most guides skip: "AI search" isn't one engine. ChatGPT, Gemini and Google AI Overviews each build their answers from different sources, weigh different signals, and reward different tactics. Optimising for "AI" in general is like buying one ad and expecting it to run on TV, radio and billboards.
In this follow-up, I'll break down how each platform actually selects the businesses it recommends, give you a specific playbook for each, and show you how to test your AI visibility in the next ten minutes.
The short answer
To get cited by AI engines, match your optimisation to how each one sources answers: ChatGPT relies on web-wide brand mentions and its search partner's index; Gemini leans on Google's index and Knowledge Graph entity data; and Google AI Overviews lifts answers from pages that already rank well and answer questions directly. Strong businesses cover all three — because buyers use all three.
How to appear on ChatGPT
ChatGPT
Sources: training data · live search · web-wide mentionsChatGPT builds recommendations from two layers. Its model knowledge — what it learned about your brand from articles, directories, reviews and discussions across the web — and its live search, which pulls current pages when the question needs fresh information. That means ChatGPT citations are less about your website alone and more about how often and how consistently your brand is described across the wider web.
This is why the New Jersey personal injury firm I work with gets named in ChatGPT responses: we didn't just optimise their site — we built consistent descriptions of who they are, where they operate and what results they get across directories, legal platforms and third-party mentions.
- Seed the web with consistent brand descriptions — same positioning sentence across directories, profiles, PR and guest content. Repetition is how models "learn" who you are.
- Get onto the pages ChatGPT's search surfaces — "best X in [city]" listicles, industry roundups and comparison articles are cited constantly. Being featured in them matters more than outranking them.
- Make your money pages retrievable — fast, crawlable, with a clear one-paragraph summary of who you serve and why you're credible near the top.
- Don't block AI crawlers — check robots.txt isn't excluding the bots that feed AI search (a surprisingly common self-inflicted wound).
How to appear on Google Gemini
Gemini
Sources: Google index · Knowledge Graph · Business ProfileGemini has a home advantage: it draws directly on Google's index and Knowledge Graph — Google's structured map of entities, their attributes and relationships. If Google's systems clearly understand your business as an entity (name, category, location, services, reviews), Gemini can recommend you with confidence. If your entity data is thin or contradictory, you're a risk it won't take.
- Treat your Google Business Profile as a ranking asset — complete every field, keep categories precise, post updates, and answer the Q&A section yourself before customers do.
- Strengthen entity signals with schema — LocalBusiness/Organization markup with sameAs links to your profiles ties your website to the Knowledge Graph entry.
- Win reviews with substance — Gemini's local recommendations lean on review volume, recency and the specific words reviewers use. Reviews mentioning your services and location are worth more than star ratings alone.
- Fix NAP inconsistencies first — conflicting addresses or business names across the web directly weaken the entity Gemini reads from.
How to appear in Google AI Overviews
Google AI Overviews
Sources: top-ranking pages · direct-answer copy · structured contentAI Overviews is the most SEO-adjacent of the three: it overwhelmingly cites pages that already rank well for the query or closely related ones. But ranking isn't enough — it lifts from pages whose copy is structured to be quotable: a question heading, a direct answer in the first one or two sentences, then detail. This is the same mechanic behind the 35+ featured snippets I've won for clients, and it transfers almost one-to-one.
- Keep your traditional SEO sharp — you generally need top-10 visibility for the topic before an Overview will cite you. AEO doesn't replace this; it converts it.
- Write "liftable" passages — 40–60 word direct answers under question-phrased H2/H3s. If Google can quote it cleanly, it will.
- Target question queries deliberately — "how much does X cost", "X vs Y", "is X worth it" trigger Overviews far more often than head terms.
- Refresh dates and facts — Overviews visibly favour current pages; stale content gets rotated out.
The three engines, side by side
| ChatGPT | Gemini | AI Overviews | |
|---|---|---|---|
| Primary source | Model knowledge + live web search | Google index + Knowledge Graph | Top-ranking pages for the query |
| Heaviest signal | Consistent brand mentions across the web | Entity data + Business Profile + reviews | Rankings + quotable direct answers |
| Best first move | Get featured in "best X" roundups & directories | Fully optimise Google Business Profile | Add direct-answer passages to ranking pages |
| Time to impact | Slowest — authority builds over months | Medium — entity signals update steadily | Fastest — can move in weeks |
| Biggest mistake | Optimising only your own website | Inconsistent NAP / thin profile | Burying the answer mid-page |
Test your AI visibility in 10 minutes
Before optimising anything, establish your baseline. Open each platform and run the questions your buyers actually ask:
The visibility test
Run each prompt in ChatGPT, Gemini & Google searchScore each response: Named (you're recommended), Mentioned (you appear but aren't the pick), or Invisible. Repeat monthly — AI answers shift as your signals build.
Two things surprise most business owners the first time they run this. First, how often a smaller competitor is being recommended purely because their content answers questions better. Second, how differently the three engines treat the same business — being cited on one platform tells you nothing about the other two, which is exactly why a platform-specific approach matters.
Where to start (if you can only do one thing per platform)
- AI Overviews first — it's the fastest win. Add direct-answer passages to the pages that already rank, and you can see citations within weeks.
- Gemini second — a fully built-out Google Business Profile plus consistent entity data compounds with everything else you do on Google.
- ChatGPT third — it's the slowest burn but the deepest moat. Web-wide brand mentions take months to accumulate, which is precisely why competitors can't quickly copy them once you're established.
For the bilingual law firm I mentioned in Part 1, this exact sequencing — Overviews, then entity work, then mention-building in both English and Spanish — is what produced citations across three AI platforms alongside 120% organic growth.
Frequently asked questions
Do I need different content for each AI platform?
Why does ChatGPT recommend my competitor but Google AI doesn't?
Can I track AI citations automatically?
Does Perplexity or Bing Copilot need a separate strategy?
How long until I'm cited on each platform?
The bottom line
"Getting cited by AI" isn't one job — it's three. Rank and structure for AI Overviews, build entity and review strength for Gemini, and earn consistent web-wide mentions for ChatGPT. Do all three and you're not chasing an algorithm — you're building a presence every answer engine independently agrees on.
And if you haven't read Part 1 yet, the foundations live here: What is Answer Engine Optimization (AEO)?
Want your visibility test done for you?
I'll run your money queries across ChatGPT, Gemini and Google AI Overviews and send you a plain-language scorecard — as part of a free growth audit.
Instant reply on WhatsApp · No pitch call required · 48-hour audit turnaround