AEO

How does AEO work? The 5-layer framework

How AEO works: the short version

AI engines, ChatGPT, Perplexity, Google AI Overviews, Gemini, decide what to cite by synthesizing signals from multiple sources: structured data on your website, content published on trusted platforms, brand mentions across the web, and the authority signals of your domain.

AEO is the practice of building and strengthening all of these signals simultaneously.

This is not a single tactic. It is a layered system. Businesses that implement only one or two layers see marginal results. The compounding happens when all five layers are working together.

Layer 1: Technical foundation

AI crawlers operate differently from Google's Googlebot. They cannot reliably execute JavaScript, do not wait for client-side rendering, and have less tolerance for slow load times.

The technical requirements for AEO: server-rendered HTML (content must be in the HTML response, not loaded via JavaScript), semantic HTML (using the correct structural tags), clean heading hierarchy (one H1 per page, H2s for major sections, H3s for subsections), fast load times, and mobile-first design.

A quick diagnostic: right-click your homepage and select View Page Source. If the body content is not visible in the raw HTML, AI crawlers are not seeing it either.

Layer 2: Structured data

Structured data is the clearest signal an AI engine can receive. JSON-LD schema tells the engine exactly what your business does, where it operates, what it costs, and what questions it answers.

The highest-impact schema types for AEO:

FAQPage: each FAQ entry becomes a candidate for direct pull-through into Google AI Overviews. Deploy with 4 to 6 question-answer pairs per page. The question goes in the `name` field and the answer goes in `acceptedAnswer.text`. Keep answers under 60 words for highest extractability. This is the single highest-impact technical change most businesses can make.

Service: describes your service offering with name, price, area served, and provider.

LocalBusiness: name, address, phone, hours, geo coordinates. Essential for any business with a physical location or service area.

Organization: company name, URL, social profiles, description. Builds the entity recognition that AI engines need to confidently identify and recommend your business. Include `sameAs` links to your LinkedIn company page, your X profile, and any other verified external profiles.

Use Google's Rich Results Test (free) to validate every JSON-LD block before publishing. Merkle's Schema Generator is useful for producing clean boilerplate for each schema type.

Layer 3: Answer-first content

AI engines prioritize content that answers questions directly. The most common mistake in on-site content is burying the answer in the fourth paragraph after three paragraphs of introduction.

The answer-first structure: heading that matches the question, direct answer in the first 1 to 2 sentences, supporting detail and context below. Specificity over generality. "Our AEO service typically shows initial citation improvements within 4 to 6 weeks" is extractable. "Results vary depending on your industry" is not.

For a London hospitality group we work with, restructuring service pages to lead with direct answers before expanding into detail produced their first AI citations within 3 weeks of publishing.

Layer 4: Platform seeding

This is the layer most businesses miss, and it is where the compounding happens.

Medium accounts for 14.3% of ChatGPT citations. Publishing detailed, authoritative articles on your core topic with consistent brand name references builds a direct citation signal. Target a minimum of 1,200 words per article, structured with clear H2 headings that match buyer questions.

Reddit accounts for 46.7% of Perplexity citations. Authentic participation in relevant subreddits, answering questions, contributing to discussions, builds citation authority. Promotional posts are downvoted and filtered. Genuine contributions over months build the authority that Perplexity trusts.

LinkedIn sends a consistent authority signal across all engines. Long-form LinkedIn Articles (not just posts) published consistently on your core topic area build topical expertise recognition. Aim for one article every two weeks.

YouTube has a 0.737 correlation with AI citation authority, the strongest correlation of any single platform. Businesses with consistent YouTube presence see dramatically higher citation rates across all AI engines. AI models index video transcripts and subtitles, so publishing full transcripts with each video compounds the effect.

The content seeding work is ongoing. One Medium article does not build citation authority. Consistent publication over months does.

Layer 5: Citation monitoring and gap analysis

Citation monitoring means running queries across ChatGPT, Perplexity, Google AIO, and Gemini and recording which businesses get cited, how prominently, and with what language.

The gap analysis identifies queries where competitors are cited and you are not, queries where you are mentioned but not recommended, and queries where no business is recommended (opportunity). Each gap maps to a specific content intervention.

Tools for monitoring: Ahrefs and SEMrush for on-site signals, DataForSEO for citation tracking at scale, and a simple spreadsheet for running 30 to 50 queries manually each month. Manual monitoring takes about 2 hours per month but gives the most reliable data on what AI engines are actually saying.

How long does AEO take to work?

Technical fixes (structured data, site architecture): 1 to 2 weeks to implement, days to weeks to register with crawlers. On-site answer content: 2 to 4 weeks to write and publish, 2 to 4 weeks to be indexed. Platform seeding: 4 to 8 weeks to begin registering as citation signals. Sustained citation share improvement: 3 to 6 months.

The technical mechanics of AEO implementation

AEO implementation has a specific sequence that matters. Skipping steps or completing them out of order produces fragile results that deteriorate when competitors catch up.

Step one is always a technical audit. Before adding schema or publishing content, you need to know whether AI crawlers can parse your site at all. A site that relies on client-side JavaScript rendering for its main content may be largely invisible to AI crawlers that do not execute JavaScript. Server-rendered HTML is the baseline requirement.

Step two is schema implementation. JSON-LD blocks are deployed in the site head or body. FAQPage schema, Service schema, Organization schema, and LocalBusiness schema cover the most common business types. Each block makes explicit what the page is about, who runs it, and what it offers. This is information AI engines would otherwise have to guess.

Step three is on-site content structure. Pages organised with the answer before the elaboration. H2 and H3 headings that match the exact questions users ask AI tools. Short, extractable answer paragraphs (2 to 4 sentences) followed by supporting detail. This structure makes the AI's job easier: it can find the answer, extract it, and cite the source.

Platform seeding: the highest-variable step

Platform seeding is where AEO results vary most between businesses. The mechanics are straightforward. The execution is where most providers cut corners.

Medium requires long-form, substantive articles (typically 1,200 to 1,500 words) that reference your business and your industry knowledge. The articles need to be genuinely useful to readers, not marketing copy. ChatGPT cites Medium at 14.3% of responses and has clear quality signals it uses to distinguish useful content from promotional content.

Reddit requires authentic participation in relevant subreddits. Promotional content is downvoted and ignored. Genuinely useful answers to specific questions build reputation and citation authority. Perplexity cites Reddit at 46.7% of responses precisely because Reddit content reflects real-world user experience rather than brand messaging.

LinkedIn content correlates with AEO authority for professional service businesses. Articles and posts from the founder or practice leads build the entity verification trail that AI engines use to confirm a business is real and credible.

YouTube has a 0.737 correlation with AI citation authority across all AI engines. A consistent video publishing strategy on your core topic cluster compounds faster than almost any other AEO investment over 6 to 12 months. A Manchester recruitment firm we work with saw ChatGPT citations triple within 4 months of starting a weekly short-form video series on hiring challenges.

Monitoring and iteration

AEO without monitoring is like running an ad campaign without tracking conversions. The monitoring setup involves running a defined set of queries across ChatGPT, Perplexity, and Google AI Overviews on a weekly or monthly basis and recording which businesses are cited. Share of voice is the metric: out of 50 queries in your topic cluster, how many cite your business?

The iteration loop is: identify queries where competitors are cited and you are not, create targeted content addressing those specific queries, publish on the platforms the relevant AI engine trusts, and monitor for citation changes over the following 4 to 8 weeks.

This is why AEO is a continuous programme rather than a one-time project. The competitive landscape shifts. New queries emerge. AI engines update their training data. The businesses that maintain citation dominance are those that maintain the programme.

Frequently asked questions

How is AEO different from SEO?
SEO gets your website to appear in Google's organic search results. AEO gets your business cited in AI-generated answers on ChatGPT, Perplexity, and Google AI Overviews. The signals are different (brand mentions vs backlinks), the platforms are different (Medium and Reddit vs third-party sites), and the measurement is different (citation share vs ranking position).

What schema type should I implement first?
FAQPage JSON-LD on your key service pages. Add 4 to 6 question-answer pairs using buyer language. This is the fastest technical change that produces measurable AI citation improvements.

Does AEO replace SEO?
No. The two disciplines share a technical foundation (server-rendered HTML, clean markup, fast load times) and reinforce each other. AEO extends beyond your website into the platforms AI engines trust. Run both in parallel.

How do I know if AEO is working?
Run 30 queries your target customers ask AI tools. Record which businesses are cited. Run the same queries 60 days later and compare. Citation share improvement across those queries is the primary metric.

Can a small business do AEO without an agency?
Yes. The minimum viable programme (FAQPage schema, one Medium article per month, monthly citation monitoring) can be executed without specialist budget. The constraint is consistency, not cost.

Related reading