How AEOscore works
How the product works in practice. A scan runs your buyer questions through the engines; everything else, fix plans, social targets, rival radar, the monthly report, hangs off what that scan finds.
- What happens in a scan?
- A scan runs your panel of real buyer questions through all six engines, records every answer in full, counts each mention of you and your tracked competitors, and updates your score, your fix plans, and your social and listings targets. Scans run weekly on Monitor and daily on Monitor + Strategist, so movement shows up while it is still fresh enough to act on.
- Which AI engines does AEOscore check?
- All six that matter for UK businesses: ChatGPT, Claude, Gemini, Perplexity, Google AI Overviews and AI Mode. Checks run through the engines' official APIs plus the search surfaces people actually see, and a Claude Opus stream is available opt-in. Every tier gets all six engines; we do not paywall engines against each other.
- What is a prompt panel and who writes mine?
- Your prompt panel is the set of real buyer questions every scan asks: brand questions about you, comparisons against rivals, and open recommendation questions where new customers are won. We seed it at onboarding from your business, your competitors and your buyers' intents, and refine it over time. On the Strategist tier the expert hones it with you every month.
- What do fix plans actually give me?
- Every gap a scan finds gets a free written quick take. On top of that, each month includes an allowance of full fix plans: each one reads the pages currently winning the answer, and your own site, before setting out exactly what to change, and can write a first draft for you. Once a fix is live, AEOscore tracks whether the answers moved and says so plainly: improved, too early to tell, or no clear change yet.
- What are social targets and listings to win?
- Social targets are the Reddit threads, videos and forum posts the AI engines cite instead of you, found by reading the actual citations in your recorded answers. Listings to win are the directories and roundup pages sending rivals their mentions. Both come with every scan, so your effort goes to the specific pages the engines demonstrably read, not a generic outreach list.
- What is rival radar?
- Rival radar watches your recorded answers for names that should not be there: new competitors the engines have started recommending, and brand hijacks, lookalike names trading on yours. Both are flagged automatically. Competitor tracking underneath it uses alias matching, so abbreviations, misspellings and rebrands still count instead of quietly slipping through the numbers.
- Do I get a report I can share?
- Yes. Every tier includes a monthly one-page report you can share with a link: your score and trend, share of voice against rivals, what changed, and what was fixed. It is written to be forwarded, to a boss, a board or a client, without you having to translate anything first.
- Do you use the engines' official APIs?
- Yes: scans run through official APIs plus the real search surfaces people see, and every question is sampled repeatedly so one odd answer cannot swing a score. Answers are kept in full, which means every number in your dashboard can be traced back to the actual text an engine produced. No screenshots of cherry-picked chats, no unverifiable claims.
What's your AEO score?
Send your domain and a named expert will run it through all six engines, score the results and email your report within 1 to 2 working days. Free, no card, no signup.
- ChatGPT
- Claude
- Gemini
- Perplexity
- Google AI Overviews
- Google AI Mode