Answer engine optimisation for UK and UAE businesses: what it is, how AI picks who to name, and what moves it
BrandurePublished Last updated
Answer engine optimisation (AEO) is the work of getting your business named and cited inside AI-generated answers from ChatGPT, Perplexity, Gemini, Google AI Overviews, Claude and Copilot. Each engine decides who to name from a different source layer, so the levers that move you are mostly outside your own website: directories, trade press, reviews, Google’s own feeds and a consistent business identity.
What is answer engine optimisation, and how is it different from SEO?
SEO is about ranking in a list of links. AEO is about being the business an AI engine names when someone asks “who’s best for X”, and being the source it cites to back that up.
The two overlap heavily. Most AI engines search the web before answering commercial questions, so search visibility still feeds them. But being ranked is no longer the same as being named, and the click is shrinking. Pew Research found that users clicked a traditional result in 8% of visits when a Google AI summary appeared, against 15% without one, and clicked a link inside the summary itself in just 1% of visits (Pew Research, July 2025).
If the answer is the destination, the question becomes who the answer names.
How does each AI engine decide which businesses to name?
They don’t agree, and they don’t read the same sources. We tested best [category] in Dubai across seven categories on ChatGPT, Perplexity, Gemini and Google, logged out with a UAE location, in August and again in September 2026 (full results). Claude and Copilot weren’t in that test, so for those two we rely on cited third-party research.
| Engine | What it mostly draws on | What that means for you |
|---|---|---|
| ChatGPT | Whatever third-party authority exists in the category: a directory, trade press, or review volume | Your standing in those sources matters more than your own pages |
| Perplexity | Editorial content, plus a separate Places module | Published category content still works here, including a brand’s own |
| Gemini | Google’s own feeds, and which feed depends on the vertical | Feed and profile hygiene (Hotels, Maps, Shopping, YouTube) comes first |
| Google AI Overviews | Pages from the eligible pool, with the cited source rotating | Aim to be eligible, not to own a specific citation |
| Claude | Brave Search results, when it searches at all | Brave visibility is a measurable proxy |
| Copilot | Bing’s index | Bing is the only engine that tells you which of your pages it cited |
ChatGPT: uses whatever authority the category already has
In our September run, ChatGPT grounded its law firms answer entirely on Legal 500, with a citation for all eight firms it named. For B2B SaaS it spread across trade and startup press (eChai Ventures, Analytics Insight, DXBStart). For aesthetic clinics it cited Doctify and Practo for framing only and gave no per-brand citation for any of the 11 clinics it named, falling back on ratings and review counts.
The rule we saw: if a directory exists, ChatGPT uses it. If trade press exists, it spreads across it. If neither exists, reviews decide.
Perplexity: reads editorial content, and still trusts brands writing about themselves
Perplexity leaned on editorial sources in almost every category. It also shows two answer layers on local queries, a Places module and the generated text, and the two often name different businesses.
Self-published content still works here. A Dubai HR software company’s own “best SaaS companies” article, which ranks the company first, was still grounding Perplexity’s answer in September. ChatGPT had stopped citing self-published lists entirely by then (how Dubai companies write the sources AI quotes).
Gemini: routes you into Google’s own feeds
Gemini’s answers mapped onto Google products, and which product depended on the vertical (full breakdown):
- Hotels: the Google Hotels module. The same three hotels, ratings and review counts as Google’s own Hotels module, but in a different order and at different nightly rates.
- Local services: Google Maps.
- Manufacturers: Google Shopping cards.
- Universities: YouTube.
For a local business, that makes Google Business Profile work AEO work, not a separate task.
Google AI Overviews: the citation is a lottery, the shortlist isn’t
AI Overviews fired on 4 of our 7 categories in both August and September: law firms, B2B SaaS, manufacturers and universities. Across three identical manufacturer searches, Google named the same three brands each time but rotated which source it cited, while the organic results underneath stayed identical.
So the realistic goal is being in the pool Google draws from. Anyone promising you a specific AI Overview citation is overselling.
Claude: built on Brave
Claude’s web search uses Brave’s results. One study reported by Search Engine Land found Claude searched on about 36.6% of prompts, mostly “best X”, ranking, location and comparison queries, and that its citations matched ChatGPT’s only 8% of the time on the same prompts (Search Engine Land, June 2026). That’s a single practitioner study, so treat the figures as directional.
Copilot: built on Bing
Copilot grounds on Bing, and Bing Webmaster Tools now shows how often Copilot and partner surfaces cited your pages and the “grounding queries” behind those citations (Search Engine Land, January 2026). If you’ve never verified your site in Bing, that’s the cheapest AEO fix there is.
Why do third-party sources drive AI citations more than your own website?
Because the engines are increasingly picking sources that look independent. The clearest example in our data: the same SaaS company held first place on ChatGPT in both August and September, but in September ChatGPT backed that position with three third-party publications rather than the company’s own article. The position outlived the self-published source. Earned coverage was holding it up.
Where a category has an authority layer, it dominates. Legal 500 was ChatGPT’s only law-firm source. And where it doesn’t, the answer is unstable: ChatGPT’s law firms list still reshuffled month to month, with the previous first-place firm dropping out entirely, even with Legal 500 as the constant source. A directory steadies the source, not the ranking.
Self-published category content isn’t dead. It still works on Perplexity. But it’s fragile in a different way: in our September run, one Dubai law firm’s page-one “best law firms” article had been displaced by a competitor running the same play. Positions taken with your own content get copied.
What actually moves your visibility in AI answers?
Five levers, in roughly this order for most businesses:
- Standing in the category’s third-party sources. The directory, ranking or trade publication the engines already cite in your category. Find out which one first; it differs by sector.
- Earned coverage. Independent articles that name you, in publications the engines pull from. This is what kept a brand’s ChatGPT position after its own article stopped being cited.
- Reviews and profiles. For local and service businesses, review volume and Google Business Profile quality drive ChatGPT’s fallback ranking and Gemini’s Maps answers.
- Google’s feeds. Hotels, Merchant Center, YouTube. If Gemini routes your vertical into one of them, that feed is the lever.
- A clean entity. One Dubai clinic appears under two spellings of its name, with two Google listings. Perplexity treated it as one business; ChatGPT and Gemini each picked one spelling and ignored the other. Inconsistent names and duplicate listings split your visibility in two. That’s a fixable identity problem, not a content problem.
Your own site still matters as the thing these sources point to, and Google’s guidance on its AI features is that they rest on the same fundamentals as normal search, with no special markup required (Search Engine Journal).
Does AEO work differently in the UK and the UAE?
The engines are the same; the source layers aren’t. Which directories, trade titles and review platforms carry weight is local, so a UK accountancy and a Dubai law firm need different source maps even on the same engine.
Our first-party testing is Dubai only, so treat the category patterns above as a guide for UK businesses rather than a measured result, and check them in your own market. One practical difference: Google’s generative AI report in Search Console launched in June 2026 with UK properties first, before going worldwide (Google Search Central).
How do you measure whether AI engines are naming you?
Per engine, repeated, logged out. Three rules:
- Report each engine separately. Low overlap between engines is the norm, so a blended score hides where you’re winning and losing.
- Run each prompt more than once. In our test, ChatGPT asked the same clinics question twice, minutes apart, kept the same first place but changed half the list below it. A single run is a snapshot, not a position.
- Record the source, not just the mention. Who the engine cited tells you which lever to pull.
Then add the first-party data that exists. Search Console’s generative AI report shows impressions in AI Overviews and AI Mode by page, but no clicks or queries (TechWyse). Bing’s report shows Copilot citations and grounding queries. Nothing first-party exists yet for ChatGPT, Perplexity or Claude, so manual sampling fills that gap. Our step-by-step guide to checking whether ChatGPT recommends your business covers how to run it, and how to choose an AEO agency covers what to ask if you hand it to someone else.
FAQs
Is AEO the same as GEO? In practice, yes. Generative engine optimisation and answer engine optimisation describe the same work: being named and cited in AI answers. Neither replaces SEO; they sit on top of it.
Which AI engine should a business prioritise? The one your buyers use, and the one where your category’s source layer is easiest to influence. For local businesses, Gemini and Google AI Overviews are usually the fastest wins because they draw on Google profiles and feeds you already control.
Can I pay to appear in ChatGPT’s answers? Not in the organic answer. What you can influence is the sources it cites: directories, press and reviews.
How long does AEO take to show results? It varies by engine and lever. Profile and identity fixes are quick to ship; directory standing and earned coverage take longer to build. Be wary of anyone quoting a fixed timeline before they’ve measured your starting position.
Will my position in AI answers stay put once I have it? No. Answers change month to month, the same engine varies between runs, and competitors copy what works. Treat it as a position to hold, not a one-off win.
Sources
- Pew Research Center: Google users are less likely to click on links when an AI summary appears (July 2025)
- Search Engine Land: Claude visibility and Brave Search rankings (June 2026)
- Search Engine Land: Bing Webmaster Tools testing new AI Performance report (January 2026)
- Google Search Central: Introducing Search Generative AI performance reports (June 2026)
- TechWyse: Google Search Console now tracks AI search visibility
- Search Engine Journal: Google’s new AI search guide calls AEO and GEO still SEO
- Brandure first-party testing:
best [category] in Dubai, 7 categories, ChatGPT, Perplexity, Gemini and Google Search, logged out, UAE location, August and September 2026.