Services
Answer Engine Optimisation (AEO)
Answer engine optimisation (AEO) is the practice of making your website readable, quotable and trustworthy to AI answer engines such as ChatGPT, Perplexity, Claude and Google’s AI Overviews, so that they cite your business when customers ask questions you can answer. The channel is no longer speculative. Adobe measured a 1,200% rise in traffic reaching US retail sites from generative AI sources in under a year, and Semrush found the average AI-referred visitor converts at 4.4 times the rate of one arriving from traditional organic search. It is also a channel many businesses are already losing with no way of noticing, because most AI crawlers do not execute JavaScript and simply never see a site that builds its content in the browser. RL Enterprise is an international AEO agency with Canadian roots, working across North America and Europe. We audit how AI systems currently see your site, fix what blocks them, and restructure your content so answer engines can lift it as the answer.
What is answer engine optimisation, and how is it different from SEO?
Traditional SEO earns you a ranking. AEO earns you a citation.
When someone asks ChatGPT or Perplexity a question, the engine does not show ten blue links. It composes one answer and cites a handful of sources. Either you are in that answer or you do not exist for that customer. The same is now true at the top of Google, where AI Overviews answer the query before the first organic result appears.
This is not a cosmetic change to search. The evidence says AI search is a separate distribution channel. Studies report that roughly 28% of the pages ChatGPT cites most often have zero organic visibility in Google. Ranking well in Google does not automatically get you cited, and being invisible in Google does not exclude you. AEO is the discipline of winning this channel deliberately, and it is also called generative engine optimisation (GEO) or AI search optimisation. Same work, different labels.
| Traditional SEO | AEO | |
|---|---|---|
| The prize | A ranking position on a results page | A citation inside a generated answer |
| Who reads your site | Googlebot and Bingbot, which render JavaScript (Bingbot imperfectly) | GPTBot, ClaudeBot, PerplexityBot and their search variants, which do not execute JavaScript at all |
| Unit of retrieval | The page | The chunk: a heading and the passage under it |
| What wins | Relevance, authority, links, user signals | Direct answers, cited sources, quotations, statistics, machine-readable structure |
| What loses | Thin content, slow sites, bad links | Client-side rendering, blocked crawlers, vague copy, keyword stuffing |
| Who it favours | Established sites with deep link equity | Peer-reviewed research shows the biggest gains go to lower-ranked, smaller sites |
Why should a small business care about being cited by ChatGPT?
Because this is the rare moment in search where being small is an advantage.
The foundational peer-reviewed research on this field (the GEO paper by Aggarwal et al., presented at KDD 2024 by researchers from Princeton and Georgia Tech) tested which page changes actually lift visibility in generative engines. Three tactics each produced gains in the region of 30 to 40%: adding citations to sources, adding quotations from credible sources, and adding statistics. Keyword stuffing performed at or below baseline. Most importantly, the gains were largest for lower-ranked sites. Generative engines disproportionately reward smaller players who structure their content well.
That maps exactly onto who we work for: small and mid-sized businesses that need to outrank larger competitors, regionally or internationally. In classic SEO, a bigger competitor’s decade of backlinks is a wall. In AI search, a precise, well-evidenced answer from your site can be cited ahead of a vague page from theirs.
Can ChatGPT and Claude actually read your website?
For a large share of modern websites, the honest answer is no, and the owners have no idea.
Independent crawler research (the Vercel/MERJ study of AI crawler behaviour) confirmed that OpenAI’s GPTBot, OAI-SearchBot and ChatGPT-User, Anthropic’s ClaudeBot, Claude-SearchBot and Claude-User, and PerplexityBot do not execute JavaScript. Googlebot and Applebot do render it. Bingbot does, imperfectly.
If your site is a client-side rendered React, Vue or Angular application that ships an empty HTML shell and paints content with JavaScript, the AI crawlers see the shell. Your pricing, your services, your expertise: invisible. We are a development shop as much as an agency, so fixing this with server-side rendering, static generation or pre-rendering is core work for us, not a referral to someone else.
We regularly find robots.txt files that block these bots by accident, usually copied from a template during a redesign. One wrong line can remove you from ChatGPT search entirely.
You do not have to take our word for any of this. We built a free tool that checks your site in about a minute: it tests whether GPTBot, OAI-SearchBot, ClaudeBot, Claude-SearchBot and PerplexityBot can actually read your pages, whether your robots.txt allows them, whether structured data exists, and whether your error pages return real 404s.
| Crawler | Operator | Executes JavaScript? | What it controls |
|---|---|---|---|
| GPTBot | OpenAI | No | Training data collection for OpenAI models |
| OAI-SearchBot | OpenAI | No | ChatGPT search. OpenAI documents that opted-out sites “will not be shown in ChatGPT search answers” |
| ChatGPT-User | OpenAI | No | Live page fetches when a user’s ChatGPT session visits your site |
| ClaudeBot | Anthropic | No | Training data collection for Claude |
| Claude-SearchBot / Claude-User | Anthropic | No | Claude’s search index and live user-triggered fetches |
| PerplexityBot | Perplexity | No | Perplexity’s answer index |
| Googlebot | Yes | Google Search, AI Overviews and AI Mode. This is the control that matters for AI Overviews | |
| Google-Extended | n/a (robots.txt token, not a crawler) | Gemini training only. Google states it “does not impact a site’s inclusion in Google Search”, and blocking it does not remove you from AI Overviews | |
| Bingbot | Microsoft | Imperfectly | Bing, which also feeds ChatGPT search results |
| Applebot | Apple | Yes | Siri and Apple Intelligence surfaces |
How do you get a page cited by AI answer engines?
Our methodology, in the order we execute it:
- 01Baseline and technical audit.We record where you currently appear (and do not appear) in ChatGPT, Perplexity, Claude and AI Overviews for the questions your buyers ask. Then we audit rendering, robots.txt, crawlability, structured data and response codes. Everything after this is measured against the baseline.
- 02Fix the machine-readability layer.Server-rendered or pre-rendered HTML so non-JavaScript crawlers get the full content. Correct robots.txt directives for each bot, chosen deliberately rather than by template. Clean HTTP status behaviour, because soft-404s poison AI indexes with pages that should not exist. Schema.org structured data so machines can parse what you are without guessing. One honest note on schema: Google removed FAQ rich results from Search in May 2026, so FAQ markup no longer earns a visual rich result. We still implement it, because the point is machine parsing, not the snippet.
- 03Restructure content for retrieval.Answer engines retrieve chunks, not pages. So every important page leads with a direct, complete, liftable answer. Headings are phrased as the questions buyers actually ask. Each section stands alone, because it may be retrieved alone. This is exactly how this page is written, deliberately.
- 04Add the evidence that engines reward.Citations to sources, quotations from credible sources, and statistics: the three tactics the KDD 2024 research found lift generative visibility by roughly 30 to 40% each. We build them into your content rather than sprinkling them on top.
- 05Build presence where engines already look.Wikipedia and Reddit together account for over 25% of US ChatGPT citations. We assess, honestly, whether your business has a legitimate route to visibility in the communities and reference sources your engines draw from, and we do it without astroturfing, which gets detected and burned.
- 06Monitor and report.AI answers change constantly. We re-test your target questions on a fixed schedule and report citations gained and lost against the baseline, so you can see the channel moving.
Does llms.txt work? A note on what we will not sell you
You may have read that adding an llms.txt file makes AI systems understand your site. Google’s John Mueller has said plainly that no AI system currently uses it. We will add one if you want it, since it costs nothing and may matter later, but we will not bill you for it as a deliverable or pretend it moves anything today. An AEO agency that leads its pitch with llms.txt is telling you something about the rest of its pitch.
The same honesty applies to Google-Extended. Some agencies frame blocking or allowing it as an AI Overviews lever. It is not. Google is explicit that Google-Extended only governs Gemini training; Googlebot is what controls AI Overviews and AI Mode. We configure both correctly and tell you exactly what each one does.
Our guarantee
We do not guarantee rankings or citations. Nobody controls what ChatGPT says, and anyone who guarantees it is guessing with your money. Here is what we do guarantee: a documented baseline of your AI visibility before we start, a defined scope of deliverables with dates, and a report against that baseline at every interval. If we miss a committed deliverable in a given month, you do not pay for it. That is a promise we fully control, which is why we can put it in writing.
What you get
Every engagement includes
- AI visibility baseline report. Where you appear today in ChatGPT, Perplexity, Claude and AI Overviews for an agreed question set, documented before we change anything.
- Full technical AEO audit. Rendering, robots.txt per-bot analysis, structured data coverage, HTTP status behaviour, crawlability. Prioritised, with effort estimates.
- Rendering fixes. SSR, static generation or pre-rendering implemented by our own developers where JavaScript is hiding your content.
- Structured data implementation, validated, across your key entity, service and content pages.
- Answer-first content restructuring: pages-per-month figure for the standard engagement tier
- Evidence integration. Citations, credible quotations and statistics built into your priority pages.
- Scheduled re-testing and citation reporting. Monthly reports showing citations gained and lost against baseline.
Questions
The things people ask before they call.
›What does AEO stand for?
AEO stands for answer engine optimisation: the practice of optimising a website so that AI answer engines such as ChatGPT, Perplexity, Claude and Google’s AI Overviews can read it, understand it and cite it in their answers. The same discipline is also called generative engine optimisation (GEO) and AI search optimisation.
›Is AEO a replacement for SEO?
No. It is a second channel that shares foundations with SEO but rewards different things. Google rankings still drive substantial traffic, and Googlebot governs AI Overviews, so classic SEO remains part of the answer there. But roughly 28% of ChatGPT’s most-cited pages have no Google organic visibility at all, which proves the channels are not mirrors. We run SEO and AEO as one programme because the technical foundations overlap and the content work compounds across both.
›How long does AEO take to show results?
The technical layer moves fastest: rendering fixes, robots.txt corrections and structured data are typically crawled again within days to weeks, and OAI-SearchBot, ClaudeBot and PerplexityBot revisit sites regularly. Citation changes in live answers follow as engines refresh their indexes. We do not quote a universal timeline because it varies by site and query; what we commit to instead is a dated baseline and re-testing at every reporting interval, so you see real movement rather than a promise.
›Can you guarantee my business will be cited by ChatGPT?
No, and you should walk away from anyone who says yes. AI engine outputs are controlled by OpenAI, Anthropic, Perplexity and Google, not by any agency. What we guarantee is the work: a baseline, a defined deliverable scope, measured reporting against that baseline, and money back on any deliverable we miss.
›My site is built in React. Is that a problem for AI search?
It can be a serious one. GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-SearchBot, Claude-User and PerplexityBot do not execute JavaScript, so a purely client-side rendered React app is invisible to all of them. The fix is server-side rendering, static generation or pre-rendering, and because we build production React and Next.js applications ourselves, we implement the fix rather than just recommending it.
›Should I block AI crawlers to protect my content?
That is a legitimate business decision, but make it with accurate information. Blocking OAI-SearchBot removes you from ChatGPT search answers, which OpenAI states explicitly. Blocking GPTBot only affects model training, not ChatGPT search. Blocking Google-Extended only affects Gemini training and does not remove you from Google Search or AI Overviews. Most robots.txt files we audit get at least one of these wrong. We configure each directive deliberately and document what each choice costs and protects.
›Does FAQ schema still matter now that Google removed FAQ rich results?
Yes, for a different reason than before. Google removed FAQ rich results from Search in May 2026, so the markup no longer produces the expandable snippet in results. But structured data’s real job in an AEO context is machine parsing: it tells crawlers unambiguously which text is a question and which is its answer, which is exactly the format answer engines retrieve. We implement it for the machines, not the snippet.
Find out what AI engines see when they look at your site
Most site owners have never checked, and your Google rankings will not tell you: roughly 28% of the pages ChatGPT cites most often have no Google organic visibility at all. The check takes about a minute and costs nothing: our free tool tests whether GPTBot, OAI-SearchBot, ClaudeBot, Claude-SearchBot and PerplexityBot can read your site, whether your robots.txt lets them in, whether you have structured data, and whether your 404s are real. If the results worry you, or if you already know you want AI search handled properly, talk to us. We work with clients across North America and Europe.