The AEO Playbook: How to Get Cited by ChatGPT & Google AI in 2026

Answer engine optimisation (AEO) is the practice of structuring your content so AI tools. ChatGPT, Gemini, Perplexity and Google AI Overviews. Quote and cite you as a trusted source. You win by giving clear, self-contained answers, marking them up with schema, reinforcing who you are as an entity, and earning authoritative mentions.

That is the whole game in one paragraph. The rest of this guide is the playbook. The exact, repeatable moves we run to get pages cited inside AI answers. It builds on the foundations in our AEO vs SEO explainer; here we go deep on the how.

What is AEO (and how is it different from GEO and SEO)?

AEO is optimising to be the source an answer engine quotes when someone asks a question. GEO. Generative engine optimisation. Is often used interchangeably, with a slight lean toward influencing the generated text itself across all LLMs, not just search-attached ones. SEO is the older discipline of ranking in classic organic results. In practice you run them together: AEO and GEO sit on top of solid SEO, because the models still pull heavily from pages that already rank and carry authority.

The difference that matters for your day-to-day work: a search engine shows ten options and lets the user choose. An answer engine synthesises one response and credits a handful of sources. To make that shortlist, your content has to be easy to lift, easy to attribute, and easy to trust.

Why AEO matters right now

Two shifts make this urgent. First, AI answers are becoming the default surface. Google's AI Overviews now appear in roughly a quarter of US searches. Conductor's 2026 AEO/GEO Benchmarks Report measured AI Overviews on around 25.11% of Google searches, with informational queries far higher at roughly 57% (Conductor, Nov 2025). With AI Mode rolling out as a default tab, the share of journeys that start. And sometimes end. Inside an AI answer keeps climbing.

Second, the citations are not random. AI answers still over-index on pages that already perform in classic search. Ahrefs' 2026 analysis of AI Overview citations found a meaningful share are pulled from pages already ranking in the top results, even as the exact percentage shifts over time (Ahrefs, 2026). The takeaway is honest and unglamorous: ranking and authority feed AEO. You cannot skip the fundamentals and hope a model adopts you anyway.

The AEO playbook: seven moves, in order

Run these in sequence on any page you want cited. They compound. The later moves work best once the earlier ones are in place.

1. Lead with an answer-first block

Put the direct answer to the page's core question in the first paragraph, in under 40 words, written so it makes sense lifted out of context. No throat-clearing, no "in today's fast-paced world." State the answer, then earn the rest of the page by expanding on it. This single move is the highest-leverage thing you can do for AEO.

2. Add a Key-takeaways box near the top

A five-to-six-bullet summary high on the page gives models a pre-digested set of claims to quote. It also helps human skimmers. Which improves engagement signals that feed back into ranking. Keep each bullet a complete, standalone statement.

3. Structure the body for extraction

Use descriptive H2/H3 headings phrased as the questions people actually ask. Break complex points into numbered steps and short lists. Use one clear comparison table where a decision is involved. Tables are exceptionally easy for models to parse and reproduce.

4. Mark it up with schema

Structured data tells a machine what your content is, not just what it says. The high-value types for AEO:

Schema type	Use it when	What it earns
FAQPage	You answer 3+ distinct questions on the page	Quotable Q&A blocks AI loves to lift
HowTo	You give an ordered, step-by-step process	Step extraction into AI answers
Article / BlogPosting	Any editorial page	Author, date and publisher clarity
Organization / Person	Sitewide, on your about/author entities	Entity identity and trust
Product / Comparison	You compare options or list specs	Cleanly parsed decision data

Keep the schema honest. It must match what's visible on the page. Marking up an FAQ that isn't there is the kind of mismatch that gets pages ignored, not promoted.

5. Reinforce your entity

An AI recommends you only when it's confident who you are. Make that unmistakable. On your Organization and Person schema, use sameAs to point at your verified profiles (LinkedIn, Crunchbase, Wikipedia/Wikidata where it exists, authoritative directories) and knowsAbout to declare your areas of expertise. Then keep your name, address and phone (NAP) and your core claims identical everywhere. Site, footer, profiles, listings. Inconsistency is how a model loses confidence and drops you.

6. Publish an llms.txt and clean technical signals

An llms.txt file at your domain root is a plain-text map that points AI crawlers to your most important, well-structured content. Think of it as a robots-friendly table of contents for machines. Pair it with a healthy site: fast load, clean HTML, accessible content (not buried in client-side rendering), and a current XML sitemap. Models can only cite what they can read.

7. Earn authoritative mentions and links

This is the move people skip. And it's why their content never gets cited. Models trust what multiple credible sources corroborate. Being mentioned and linked by trusted sites lifts both your ranking and your AI trust at once. This is exactly why disciplined link earning still matters; it's the engine behind our Backlink VaultThe ArmoryThe Edge, and we go deeper in our guide to link building in 2026.

How to measure AI visibility

You can't optimise what you don't watch. AEO measurement is younger and messier than rank tracking, but a practical loop works:

1. Prompt-test manually. Ask ChatGPT, Gemini and Perplexity your target questions on a fixed schedule and log whether. And how. You're cited.
2. Track AI Overview presence for your priority queries; note when you appear as a cited source versus a competitor.
3. Watch referral traffic from AI surfaces in your analytics (ChatGPT, Perplexity, Gemini and Google AI referrers are increasingly distinguishable).
4. Use an AI-visibility tracker to scale the manual checks. See our roundup of the best AI SEO tools for 2026.
5. Review monthly, tie gains back to the specific pages and moves above, and double down on what's getting cited.

People also ask

Is AEO different from SEO, or a replacement for it?

It's an extension, not a replacement. AEO sits on top of SEO because answer engines pull heavily from pages that already rank and carry authority. Do strong SEO, then add AEO discipline so the same content earns citations as well as rankings.

What is an llms.txt file and do I need one?

It's a plain-text file at your domain root that points AI crawlers to your most important, well-structured content. It's an emerging, voluntary standard. Not yet universally honoured. But it's cheap to add and signals intent. Treat it as a useful complement to a clean sitemap, not a magic switch.

Does schema markup guarantee I'll get cited by AI?

No. Nothing guarantees a citation. Schema makes your content easier to understand and lift, which improves your odds. It works alongside answer-first writing, entity clarity and authority. Schema that doesn't match the visible page can hurt you, so keep it honest.

How do AI engines decide which sources to cite?

The exact ranking is proprietary, but the observable pattern is consistent: relevance to the question, how directly the page answers it, the source's authority and how many credible sources corroborate it. Pages that already rank well are disproportionately likely to be cited.

How long does AEO take to show results?

On-page moves. Answer-first blocks, schema, takeaways boxes. Can show up in AI answers within days to a few weeks once recrawled. Entity and authority gains are slower, often a few months, because they depend on the wider web catching up. Measure monthly and be patient with the authority layer.

Can small businesses win at AEO against big brands?

Yes, especially on specific, intent-rich questions where you have genuine expertise. The winners aren't the biggest budgets. They're the ones who made their answers stupidly easy to quote and backed them with real authority in their niche.