Buyer’s guide
How answer engines pick what to cite, the four numbers that actually mean something, how to tell a real AEO partner from a rebranded SEO retainer, and how to brief one when your own team does the work.
Answer engine optimization, or AEO, is the practice of making your content the source an AI assistant uses when it answers a question. Generative engine optimization, or GEO, describes the same work. The two terms are interchangeable in practice, and nothing useful turns on the difference.
Search engines return a ranked list of links. The user scans the list and chooses. Several results share the demand for a query, so more than one page can win something from the same search. Answer engines return one synthesised answer. The model retrieves a small set of sources, reads them, and writes a single response in its own words. There is no list to appear on. Your page is either inside the material the model used, or it is absent from the answer.
In search, position four still earned impressions, clicks and traffic. The list rendered every result, so a lower position meant less traffic, not zero traffic. An answer has no position four. The model names a small number of options, cites a short list of pages, and stops. Everything outside that set gets nothing: no mention, no citation, no visit. Visibility in an answer engine is closer to pass or fail on each query than to a sliding scale of rank.
AEO covers ChatGPT, Gemini, Perplexity, Copilot, Google AI Overviews and Google AI Mode. Each runs its own retrieval and its own source selection, so a page quoted in one can be invisible in another. Treat them as separate surfaces with separate numbers, not as one channel. One rule holds across all of them: answer engines read HTML. Content behind a paywall, or assembled in the browser by JavaScript, is often not read at all.
An answer engine handles a commercial question in three steps. Retrieval: it runs one or more searches and pulls a candidate set of pages. Grounding: it reads those pages and builds the answer out of what they say. Citation: it names or links the sources it leaned on. Each step is a separate filter. A page can be retrieved and never grounded. A page can be grounded and never cited.
Commercial prompts, the ones with buying intent, trigger a live web search rather than a recall from training data. The answer is assembled from pages that exist today. That makes this winnable now. You do not have to wait for a model to be retrained to show up. You have to be in the retrieved set, and you have to be readable and quotable once you are there.
Retrieval and grounding run on the HTML a crawler receives. Content behind a paywall, a login, or a consent wall is not in that HTML. Content that only appears after JavaScript executes in a browser is often not there either. If the claim you want cited is injected by a script, or sits inside a tab that loads on click, treat it as invisible. Put it in the served HTML, in plain text, under a heading that names the question it answers.
A model weighs whether independent sources corroborate a claim. Your own page states something. Review platforms, directories, partner pages, press coverage, and industry write-ups either support it or they do not. When the outside record agrees with the owned page, the claim survives into the answer. When only the owned page makes the claim, it reads as marketing and gets dropped. Third party corroboration is not a supporting tactic. It is part of what the model reads before it answers.
They fail in different ways, and each failure points at a different fix.
Track both. They move independently, and combining them into one number hides which problem you actually have.
Four numbers carry most of the meaning in answer engine reporting: visibility, share of voice, position and sentiment. Each answers a different question. Most reporting mistakes come from swapping one for another.
Visibility is your mentions divided by total responses. Count the responses in a run that mention your brand, then divide by every response in that run. It is absolute. It does not move because a competitor was added or removed from tracking. Visibility is the correct metric when the question is simply whether the brand shows up at all.
Share of voice is your mentions divided by all tracked brand mentions in the same responses. It is relative by construction. It tells you how much of the counted conversation is yours, not how present you are. Two reports can show the same share of voice while one brand is mentioned far more often, because each report counts a different set of brands. The denominator is a choice, not a fact.
Position records where your brand falls inside an answer when it is mentioned. Position 1 beats position 5. This inverts the usual reading of a chart. A position number moving from 2.4 to 3.1 is a decline, even though the line goes up. Write it as “position improved to 2.4” or “position slipped to 3.1”. Never write “position rose”.
Sentiment describes the tone of the mention: favourable, neutral or critical. It only covers responses that mention you, so it is meaningless in isolation. Strong sentiment on low visibility means very few people are hearing the good version.
The first is treating share of voice as if it were visibility. A share of voice figure describes a competitive set, not the market. Report visibility when the question is presence, and share of voice only when the tracked set is stated alongside it.
The second is reading a rising position number as progress. If your reporting template says a metric went up, the reader assumes improvement. For position, that assumption is backwards.
Add one competitor to tracking and your share of voice falls, because the denominator grew. Remove one and it rises. Nothing real changed in either case. Keep the tracked set stable for the whole of a reporting window. If the set has to change, change it between windows, state the change, and do not compare across the boundary as though it were like for like.
Answer engines do not return the same answer every time. Daily runs on a small prompt set produce swings that are sampling noise, not movement. Use 7 or 28 day windows for anything shown to a client. Compare each window to the immediately preceding period of equal length, not to a fixed launch baseline.
mentions / total responsesDo we show up at all?Absolute. Does not move when a competitor is added to tracking.your mentions / all tracked mentionsHow much of the counted conversation is ours?Relative. The denominator is a choice, so it moves when the tracked set changes.rank inside the answerHow early are we named?Lower is better. A number going up is a decline.tone of the mentionHow are we described?Sits in a narrow band. Small moves are usually noise.An AEO partner should be judged on measurement discipline, not on claims. Use these eight criteria, then compare proposals on the work they actually commit to.
Ask the agency to explain, in its own words, how the work changes when the goal moves from ranking a page to being cited in an answer. A team that does this every day will talk about retrieval, the sources an assistant reads before it answers, mentions against citations, and prompt sets built from real buyer questions. A team selling AEO as an add-on to an SEO retainer will talk about keywords, rankings and content volume under new labels. Then ask who does the AEO work, how much of their week it fills, and what they changed on their own site.
The same test settles whether to hire a specialist or extend your SEO contract. An SEO partner can often fix technical gaps, such as pages assistants cannot read. If assistants name competitors, cite review sites you are missing from, or describe your product wrongly, the work sits off your site and needs different skills and reporting.
Ask to see the agency’s own visibility across answer engines, not only client screenshots. An agency that sells AI visibility should be able to open its own tracking and show where it appears and where it does not. Client case studies are useful, but they are selected. Own data is not.
Branded prompts contain your company name. They mostly measure demand you already have, and they return flattering numbers. Non-branded prompts are the ones buyers type when they do not know you yet. A partner who reports one blended visibility figure is hiding the number that matters. Ask for the two figures side by side, every time.
Answer engines cite pages. A serious recommendation names the specific page a model cited and states what would have to change for your page to be cited instead. If the reasoning is “add more content” or “build authority”, there is no source behind it. Ask which URL the model cited, and what the fix is against it.
Assistants use different retrieval and different sources, so a result in one does not transfer. Ask the agency to name every assistant it tracks for your category, for example ChatGPT, Gemini, Perplexity, Copilot, Google AI Overviews and AI Mode, and to report each one separately.
Then ask for the prompt list itself, not a count, and the markets it runs in. Each prompt should trace back to a real buyer question from sales calls, support tickets or on-site search. This is also the test for case studies: if an agency cannot show the prompts behind one, you cannot tell whether it reflects questions buyers ask or questions picked because the brand already wins them. For reference, we ask six AI tools the same buyer questions every day, across the US, the UK and Germany (our AEO case study).
Check that your partner uses the terms exactly as defined in the four numbers above: visibility is absolute, share of voice moves when the tracked set changes, and a lower position is better. Then check the window: 7 or 28 days, compared to the immediately preceding period of equal length, not a fixed baseline.
Mentions are an early signal. The business case is visits, trials and enquiries. Ask for a sample report before you sign: it should group AI assistant referrals as their own channel in your analytics, with a conversion rate next to your site average. Monitoring without that link is the cheapest part of the work to sell and the hardest to judge.
Our own numbers show why it matters. Between 1 March and 5 September 2026, ChatGPT sent 2.7% of our visits and 12.4% of our leads, which makes a visitor from ChatGPT 4.6 times as likely to contact us as our average visitor (GA4, AEO case study). The case study also states its limit: these are leads on our own website, not app installs.
Mentioned but not cited is fixed on your site. Cited but not mentioned needs third-party corroboration: directories, review platforms, editorial coverage, and content on the sources models already read. Ask whether the partner executes that work or only edits your site, and confirm they check that your key content is served as HTML.
The two measure different things, and a report that blends them hides both.
Proposals that look alike on price often commit to very different work. Compare technical fixes, content and third-party citations, then test depth with one question: how would they correct a wrong claim about your product, such as an old price or a plan you no longer sell? A deep answer traces the claim to its source, then puts a correct, dated statement in HTML on your site and on outside sources the assistant already reads.
crawler access, HTML, structureDoes it name the pages and the fix?Deep: lists the pages assistants cannot read, why, and the fix. Thin: “technical optimisation”.pages built on buyer questionsWhich questions, which pages?Deep: maps each priority prompt to a page and the answer it needs. Thin: a monthly article count.sources assistants already citeWhich outside sources, and who does the work?Deep: names the review sites, publishers and lists cited for your prompts. Thin: “digital PR”.inaccurate product answersHow do they correct the record?Deep: traces each wrong claim to its source and publishes evidence to correct it. Thin: no plan.Admiral Media answers these on a first call, using our own tracking rather than selected screenshots. Talk to our team.
An SEO audit asks whether search engines can crawl, index and rank your pages. An AEO audit asks whether assistants name you when buyers ask about your category, which sources they cite, and what they say about you. The crawlability checks overlap. Almost everything else is new.
An SEO audit starts from your URLs and keywords. An AEO audit starts from a prompt set: the questions buyers ask when they compare products in your category. Each prompt runs on each assistant, for example ChatGPT, Perplexity and Google AI Overviews, and the audit records whether you are named, where, and which pages are cited. Answers change from run to run, so one screenshot is not an audit. Ask for repeated runs and a stated window.
This is the part an SEO audit does not have. For every comparison prompt, the audit should list the domains and URLs the assistant cited, then sort them by type.
That gap analysis is the core of the audit. It shows whether the problem is your own pages, a review site you are missing from, or a conversation you are not part of. Every finding should carry the prompt, the assistant, the date and the cited URL.
Answers can differ by market. A UK subscription business should ask for prompts run from the UK, in British phrasing, with prices in pounds and the providers UK buyers actually compare. A prompt set built for the US tells you how you look to US buyers. For example, in answers captured on 21 September 2026, Perplexity and Google AI Overviews in the UK both named Admiral Media first when asked for a specialist app user acquisition agency (AEO case study).
Some teams want an agency to diagnose and direct while their own developers and writers do the work. That works if the brief makes the agency’s output usable by your team. Cover four things.
Give the agency read access to what it needs to diagnose, and keep write access with your team.
Ask that every technical recommendation cites the platform’s own documentation. Google’s guide to AI features and your website says a page must be indexed and eligible to show with a snippet, and that no special structured data is needed to appear in AI Overviews or AI Mode. OpenAI’s crawler documentation separates OAI-SearchBot, which decides whether a site can be shown in ChatGPT search answers, from GPTBot, which covers model training, and each setting is independent. Perplexity’s crawler documentation describes PerplexityBot as the crawler that surfaces and links sites in its search results. If an agency asks your developers for special markup “for AI”, ask where the platform asks for it.
A short weekly check-in covers fixes in flight. Report AI visibility on 7 or 28 day windows, because daily runs are noise. A monthly review should end with what shipped, what moved, and the next fixes in order. Revisit the prompt set each quarter, and do not compare across a change to it.
Most of these fixes fit a short spec. On our own site we wrote one page that tells AI tools who we are, put our case study results into the HTML so tools that skip scripts can read them, cleaned up hidden page labels that contradicted each other, and rewrote titles and headings around real buyer questions (AEO case study). If your team will ship the fixes, talk to our team about scoping the audit and the briefs.
The first work is the record and the technical surface. Correct how your company is described wherever models can read it, and make sure the pages you want quoted are served as HTML. A page that assembles itself in the browser is invisible to the systems you are trying to influence. This stage moves fastest because it sits entirely inside your control.
Citations come next. Earning third-party references depends on other publishers, so it runs on their calendar, not yours. Once both stages hold, gains compound: a correct record that independent sources agree with gets picked up more often.
Answer engines return different answers to the same question. Daily runs produce daily noise, and one day is not a trend. Use 7 or 28 day windows for anything client-facing. Change indicators compare to the immediately preceding period of equal length, not a fixed baseline, so the second full window is the first thing you can honestly compare against. A movement chart in week one is sampling variance presented as progress.
A report is useful when it points at a specific page or a specific source. Not “visibility improved”, but: this prompt returns this answer, this domain is cited, this page of yours is not, so here is what to change. Every figure should carry an owner and a next action.
Admiral Media has run creative performance work for mobile apps since 2019, across 150+ brands. Talk to our team.
Sources: GA4, 1 March to 5 September 2026. Google Search Console, June to August 2026. From our AEO case study.
What this does not prove yet: these are leads on our own website, where a lead is one contact form. We have not yet measured the same effect on App Store visits and installs.
Source: Bing Webmaster Tools, AI Performance report, 1 June to 30 August 2026.
The pattern worth stealing: the pages doing that work are guides and benchmark posts, not service pages. Reference content is what an answer engine reaches for.
We run answer engine optimization for apps, games and ecommerce brands, and we publish our own numbers. Talk to our team.