The Shift Nobody Is Talking About Clearly Enough
LLM Optimisation — getting your brand cited when people ask AI systems questions — was phase one. It mattered enormously and still does. But phase two is already here, and it changes the game more fundamentally than most practitioners have acknowledged: AI agents that don’t just answer questions but take actions on behalf of decision-makers.
OpenAI’s Operator, Anthropic’s computer use capabilities, Google’s Gemini agents, Microsoft Copilot actions, Apple Intelligence — these are not research prototypes. They are production tools that enterprises and consumers are adopting now. When a procurement manager asks an AI agent to “research the top five managed file transfer solutions and compare them on security, compliance and pricing,” the agent doesn’t show a list of links. It visits websites, reads documentation, evaluates claims, cross-references sources, and delivers a structured recommendation. When a marketing director asks an agent to “find me an SEO consultant who specialises in healthcare IT and has demonstrable AI visibility expertise,” the agent runs the same pipeline — and either your business makes the shortlist or it doesn’t.
AI Agent Optimisation (AAO) is the discipline of ensuring your digital presence is structured, accessible and authoritative enough that AI agents can find you, understand you, trust you and recommend you. It builds on the foundations of LLM Optimisation, Entity SEO and structured data — but adds specific requirements for how autonomous agents navigate, evaluate and compare businesses that go beyond what conversational AI demands.
This guide explains how agents actually work, what they value, and what you need to change about your digital presence to be selected — not just cited. The implementation examples are drawn from this site, because we believe the strongest argument for any methodology is demonstrating it on your own business first.
The Paradigm Shift: The Buying Funnel Moves Inside the Agent
This is the conceptual shift that matters most, and it is the one that most AAO discussion still skirts around.
For twenty-five years, the buying funnel happened across your website and the search engine together. A potential client searched, clicked through to your site, browsed your services, read a case study, compared you against a competitor, and eventually made contact. The search engine was a traffic source. Your website was the conversion environment. The human did the evaluation.
Under AAO, the entire funnel — awareness, consideration, evaluation, shortlisting — happens inside the AI agent before the human ever sees a result. The agent becomes aware of your business through its retrieval process. It considers you against alternatives by parsing your site and cross-referencing sources. It evaluates your credibility through verifiable evidence. It shortlists you (or doesn’t) based on how well your digital presence matches the evaluation criteria. Then it presents the human with a recommendation — often a single recommendation, or at most a shortlist of three.
The human never visited your website during this process. They never saw your homepage. They never browsed your portfolio. The agent did all of that for them, in seconds, and delivered a verdict. Your role is no longer to attract visitors to a funnel on your site — it is to be the answer when the agent runs its own funnel internally.
This does not mean your website becomes irrelevant. The opposite: your website becomes more important than ever, because it is now the primary data source for an autonomous evaluator that reads everything, forgets nothing, and compares you against every competitor simultaneously. The difference is that the evaluator is a machine, and machines value different things than humans scanning headlines.
How AI Agents Actually Discover and Evaluate Businesses
Understanding the agent’s decision process is not academic — it directly determines what you need to optimise. AI agents follow a multi-stage pipeline that is more systematic than any human buyer.
Stage 1: Query Formulation and Retrieval
The agent receives a task — “find an SEO consultant for a healthcare IT company in the UK” — and decomposes it into search queries. It might search for “SEO consultant healthcare IT UK”, “AI visibility consultant healthcare”, “SEO agency HIPAA compliance experience” and other variations. It retrieves results from one or more search engines, its own knowledge base, and potentially structured data feeds.
This stage is where traditional technical SEO and GEO still matter enormously. If your site doesn’t rank or appear in the agent’s retrieval set, you are eliminated before evaluation begins. The agent can only evaluate what it can find.
Stage 2: Site Visit and Content Parsing
The agent visits the most promising websites and reads their content — not as a human scanning headlines, but systematically parsing page structure, extracting claims, and building an internal representation of each business. It reads your service descriptions, your pricing, your case studies, your team credentials, your structured data. It does this in seconds, across every page it can access.
This is where content architecture and structured data become critical. An agent parsing our site, for example, encounters Organisation schema that declares our entity type (ProfessionalService), our area served (United Kingdom), our founder, our knowsAbout properties (SEO, AI Optimisation, Entity SEO, Healthcare IT Marketing, Legal Services SEO), and our service offerings with structured descriptions. It can extract this information in milliseconds without interpreting marketing prose. A competitor’s site without structured data forces the agent to infer the same information from unstructured paragraphs — slower, less reliable, and lower confidence.
Stage 3: Cross-Reference and Verification
The agent verifies claims by checking multiple sources. It cross-references what your website says with what LinkedIn says, what Google Business Profile says, what review platforms say, what case studies claim, and what industry publications mention. Inconsistencies reduce confidence. Corroboration increases it.
When an agent cross-references our claims, it finds consistent information across platforms: Sean Mullins identified as founder on LinkedIn, Google Business Profile showing Southampton location matching the website, case studies referencing named clients (Coviant Software, Motoring Defence Solicitors, Olliers Solicitors) whose own sites corroborate the relationship, and AI platforms already citing our content on LLM optimisation topics. Each corroborating source strengthens the agent’s confidence in recommending us.
Stage 4: Comparative Evaluation and Selection
The agent compares all evaluated businesses against the criteria it has determined (or been given) for the task. For “SEO consultant for healthcare IT,” it weights healthcare industry experience, AI visibility expertise, verifiable results, pricing transparency, geographic relevance and authority signals. It then ranks candidates and presents a recommendation.
The critical insight: every stage of this pipeline is optimisable. And each stage has specific requirements that go beyond what traditional SEO or even conversational AI optimisation addresses. That is what AAO covers.
Push vs Pull: The Architectural Shift That Changes Everything
For two decades, search worked on a pull model: you published content, search engines crawled it on their schedule, and you waited. Google came to you, read your pages, figured out what they meant even when you made it difficult, and decided where to rank you. This model still exists and still matters — but it is no longer the only game.
The push layer is emerging. Technologies like IndexNow (which Bing has been developing for years), the Model Context Protocol (MCP), and structured data feeds allow you to push information directly to the systems that evaluate your business, rather than waiting for them to come and find it. This is not a future concept — IndexNow is already live across Bing, DuckDuckGo, Yahoo, Yandex and Ecosia, notifying search engines of content changes in real time rather than waiting for the next crawl.
For AAO, the push layer matters because AI agents operate on tight timeouts. When an agent visits your site during evaluation, it has seconds to extract what it needs. If your pages load slowly, if critical information is buried behind JavaScript rendering, if your structured data is incomplete — the agent may abandon the request entirely. GPTBot, ClaudeBot, PerplexityBot and other AI crawlers typically timeout within one to five seconds. A slow site isn’t just a ranking disadvantage — it is invisible to the agents doing the evaluating.
Our site serves pages in under a second, scores 99/100 on desktop performance, and provides comprehensive structured data on every page. That is not a vanity metric — it is agent accessibility. When an AI agent evaluates our site, it gets complete, machine-readable information instantly. When it evaluates a competitor running a page builder at 4.2-second load times with no structured data, it gets partial information slowly and with lower confidence. The agent’s recommendation reflects this difference.
The practical implication: your Core Web Vitals investment is not just an SEO and UX improvement. It is an AAO requirement. Fast, clean, well-structured sites are agent-ready. Slow, bloated, poorly structured sites are agent-invisible.
The Entity Home: Your Anchor in an Agent-Mediated World
Every business needs what we call an entity home — the single page you control that anchors everything AI systems know about you. For most businesses, this is the homepage or an about page. For our site, it is the homepage combined with the About page, which carries ProfilePage schema with an inlined Person entity connecting Sean Mullins to the Organisation entity, the knowsAbout properties, the sameAs links to authoritative profiles, and the credentials that establish authority.
The entity home matters because it is the anchor that agents and LLMs use to disambiguate and understand your brand. When an agent encounters “SEO Strategy” in a search result, a case study mention and a LinkedIn reference, the entity home is where it confirms: this is the same entity, this is what they do, this is who runs it, this is their authority basis. Without a clear entity home, agents may treat fragmented mentions as separate entities or fail to build a coherent picture of your business.
Your entity home should contain or link to: your full Organisation schema with all relevant properties, your founder or principal consultant’s credentials and authority signals, your service descriptions in structured format, your geographic coverage, your industry specialisms, and your proof of expertise (case studies, awards, publications). It should be the most comprehensively marked-up page on your site — because it is the page that agents use to calibrate their understanding of everything else.
The Seven Pillars of AI Agent Optimisation — With Implementation
Each pillar below includes the principle, our implementation on this site as a worked example, and a practical checklist for your own business.
1. Agent-Accessible Content Architecture
AI agents don’t browse — they parse. Your content architecture needs to support efficient extraction of the information agents use to evaluate and compare businesses: what you do, who you serve, what results you’ve delivered, what you charge, and how to engage. Every critical piece of information should be accessible within two navigation steps from your homepage.
Our implementation: The site architecture is built around entity clusters. The LLM Optimisation pillar page links to four child pages — AIO, AEO, GEO and this AAO page — creating a clear topical hierarchy that agents can traverse. Service pages use descriptive URLs (/healthcare-seo/ not /services/service-4/). Case studies are one click from the homepage. Pricing is published on service pages, not hidden behind forms. An agent can map our complete service offering, verify our expertise claims through case studies, and find contact information — all within two clicks of the homepage.
Your checklist: Can an agent find your complete service list from the homepage in two clicks? Are your case studies accessible without navigating through multiple category pages? Is your pricing (even indicative ranges) published and machine-readable? Does your URL structure communicate your topical hierarchy?
2. Comprehensive Structured Data
Structured data is the primary communication layer between your website and AI agents. While traditional SEO uses schema primarily for rich results, AAO treats structured data as the API that agents use to extract information without interpreting marketing prose.
Our implementation: Every page on this site carries contextually appropriate JSON-LD schema. The Organisation entity (ProfessionalService type) declares knowsAbout properties spanning our service areas. Service schema describes each offering with structured descriptions. FAQPage schema provides machine-readable question-answer pairs on every guide and service page. HowTo schema provides step-by-step processes. The ProfilePage on our About page inlines a full Person entity with credentials, sameAs links, and alumniOf properties. The @id graph connects all entities into a coherent knowledge representation that agents can parse as a single interconnected dataset.
Your checklist: Do you have Organisation/LocalBusiness schema with knowsAbout properties? Service schema for each service offering? FAQPage schema on pages with Q&A content? Review/AggregateRating schema for social proof? Does your schema graph use @id references to connect entities? Can an agent extract your services, location, expertise areas and contact details entirely from structured data without reading a paragraph of text?
3. Verifiable Claims and Quantified Evidence
AI agents are increasingly sophisticated at detecting unsubstantiated claims. “We’re the best” means nothing to an agent. Verifiable, quantified evidence is the currency of agent trust.
Our implementation: Our Motoring Defence Solicitors case study states specific, verifiable metrics: 7 position-one rankings, #3 for “drink driving solicitors” (720/mo), £149.83 CPC saved per organic click, ~80 five-star Google reviews. The Coviant Software case study references 200+ enterprise leads and £2M+ pipeline contribution. These aren’t marketing claims — they’re specific numbers that an agent can cross-reference against publicly visible data (Google search results, review counts, ranking positions). When an agent verifies that MDS does indeed rank for “drink driving solicitors” and does indeed have ~80 five-star reviews, it confirms our credibility. Each verified claim increases the agent’s confidence in our other claims.
Your checklist: Do your case studies include specific, quantified metrics? Can the key claims be verified through public data? Do you name clients (with permission) rather than using anonymous references? Are results tied to measurable business outcomes (leads, revenue, rankings) rather than vanity metrics (impressions, “brand awareness”)?
4. Cross-Platform Entity Consistency
Agents verify information across multiple sources. Inconsistencies between platforms reduce trust. Entity SEO becomes critical in an agent-driven discovery model because agents actively cross-reference rather than trusting any single source.
Our implementation: Our business name, address, phone number, founder name, service descriptions and industry specialisms are consistent across our website, Google Business Profile, LinkedIn, Wikidata entry, industry directories, and the sameAs URLs declared in our Organisation schema. The sameAs property explicitly tells agents where to find corroborating sources, making verification efficient. Our About page is the entity home that anchors this consistency — every other platform references back to the same core facts.
Your checklist: Is your business name identical across all platforms — including exact punctuation and legal entity format? Does your sameAs schema point to all your verified profiles? Have you audited your Google Business Profile, LinkedIn, Companies House listing, industry directories and review platforms for consistency in the last 90 days?
5. Transparent Pricing and Engagement Models
When AI agents compare vendors, pricing transparency is a competitive advantage. Businesses that publish pricing give agents the data to include them in comparison tables. Businesses that hide pricing behind “contact us” force agents to either exclude them or flag the opacity.
Our implementation: Our day rate (£600–900) is published on our homepage and service pages. Our AI visibility audit pricing (£1,500–5,000) and monthly retainer ranges (£1,500–5,000+) are published on the relevant pages. This is not just transparency for human visitors — it is structured data that agents can extract and use for vendor comparison. When an agent builds a comparison table of SEO consultants, it can include our pricing alongside our services and evidence. Competitors who hide pricing are omitted from that comparison or flagged as non-transparent.
Your checklist: Do you publish at least indicative pricing ranges? Are your prices included in structured data (Offer schema)? Is your engagement model clearly described — retainer, project-based, day rate? Can an agent extract your pricing without submitting a form?
6. Push Architecture and Machine-Readable Interfaces
The shift from pull-only to push-and-pull architecture is the emerging frontier of AAO. Forward-thinking businesses are making their information actively available to agent systems, not just passively crawlable.
Our implementation: Our XML sitemap is comprehensive and automatically updated. Our robots.txt explicitly permits GPTBot, ClaudeBot, PerplexityBot and other AI crawlers — we do not block them, because blocking AI crawlers is blocking the evaluators who determine whether agents recommend you. Our structured data is compiled server-side (not injected via JavaScript), ensuring it is available to every crawler regardless of rendering capability. Our site loads in under one second, within the timeout windows of all major AI crawlers. We have also built an AI Knowledge Agent on our own site — a live demonstration that we not only optimise for AI agents but build them.
Your checklist: Does your robots.txt allow AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Googlebot)? Is your structured data rendered server-side or does it require JavaScript execution? Is your sitemap comprehensive, current and error-free? Does your site load within two seconds consistently? Have you considered IndexNow integration to push content changes to Bing and its ecosystem?
7. Agent Testing and Continuous Monitoring
Just as AI Persona Testing validates how conversational AI perceives your brand, agent testing validates how autonomous agents evaluate and rank your business in competitive scenarios.
Our implementation: We regularly test our own business against agent-driven scenarios. We ask ChatGPT, Claude, Perplexity and Copilot to “recommend an SEO consultant specialising in healthcare IT in the UK” or “compare LLM optimisation agencies” and document where we appear, how we are described, and how our competitors are positioned. This is not vanity — it is the AAO equivalent of rank tracking. When we notice gaps (a competitor being cited for a service we also offer but the agent doesn’t associate with us), we diagnose whether the issue is content parsability, structured data coverage, cross-platform consistency, or evidence depth — and fix it.
Your checklist: Have you tested how agents describe your business in competitive research scenarios? Do you run these tests monthly? Do you compare results across multiple AI platforms? Have you documented what agents get right, what they miss, and what they get wrong about your business?
The Naming Debate: AAO, GEO, AIEO — Does It Matter?
The SEO industry is fractured across half a dozen acronyms for overlapping disciplines. GEO, AEO, AIO, AIEO, AAO, AI SEO — each has advocates, each covers part of the territory, and the debate about which term is “correct” has consumed more practitioner energy than the actual work of optimisation.
Here is our position: the terminology matters less than the taxonomy. What matters is understanding that AI-powered discovery is not one discipline — it is a spectrum with distinct requirements at each stage. We use a four-layer framework within our LLM Optimisation service:
AIO — Be included. Ensuring your content appears in Google’s AI Overviews. The most immediate layer, because it sits on top of the search engine most businesses already optimise for.
AEO — Be the answer. Owning the featured snippets, People Also Ask boxes and direct answers across every answer engine. The bridge between traditional search and AI-generated responses.
GEO — Be cited. Getting your brand referenced and recommended by generative AI platforms — ChatGPT, Perplexity, Claude, Gemini. The conversational discovery layer where people ask questions and AI cites sources.
AAO — Be chosen. Being discovered, evaluated and recommended when AI agents take autonomous actions on behalf of decision-makers. The selection layer where the buying funnel happens inside the agent.
Each layer builds on the previous one. You cannot be chosen by agents (AAO) if you are not being cited by conversational AI (GEO), which requires being the answer to specific questions (AEO), which depends on being included in AI-generated summaries (AIO). The layers are cumulative. The foundations — entity authority, structured data, content depth, technical excellence — serve all four simultaneously.
Whether you call the umbrella discipline GEO, LLM Optimisation, AI SEO or anything else matters far less than whether you are actually doing the work across all four layers. Incomplete terminology produces incomplete strategy only if practitioners stop at the layer their favourite acronym covers. Our approach covers all four because the AI systems your potential clients use don’t respect the boundaries between acronyms.
Who Needs AAO First?
Not every business needs to prioritise AAO immediately. The businesses that should move first are those in considered-purchase categories where buyers research extensively before engaging: B2B technology, professional services, SaaS, healthcare IT, financial services, legal services, and any category where the buying process involves vendor comparison.
The decision framework is straightforward. If your sales cycle involves discovery → evaluation → shortlisting → decision, AI agents will increasingly mediate the first three stages. The businesses optimised for agent discovery will be on the shortlist. The rest will not.
For impulse purchases or very low-consideration products, AAO is less immediately critical — though it will become relevant as agent-mediated commerce expands into retail, travel and everyday transactions.
If you are unsure where you stand, our AI Visibility Audit includes agent testing as a standard component — we run your business through competitive agent scenarios and show you exactly how AI agents perceive your brand versus your competitors. From there, we build the LLM Optimisation strategy that addresses the gaps across all four layers. Book a free 30-minute consultation to discuss your specific situation.
