Being the source an AI answer cites.
Work aimed at being cited when ChatGPT, Perplexity or Google's AI answers a question in your category.
- Baseline
- Set in week one
- Tracked
- Monthly, same question set
- Also improves
- Traditional search
What AI search optimisation is
It is the work that makes an AI system choose your site as a source when it writes an answer. Buyers increasingly ask an assistant to compare options before they open a search engine, and a company absent from that answer is absent from the shortlist.
Google has been explicit that its AI results rest on the same foundations as ordinary search, so most of this is SEO done properly, plus three things: crawler access for the AI bots, answers written before the explanation, and original information other pages cannot repeat.
Nobody can guarantee a citation. What can be done is make your site the easiest credible source to use, then measure whether it is being used.
The mechanics are less mysterious than the category makes them sound. An assistant answers by retrieving sources it can parse and trusts, then summarising them. So the work is making your pages easy to retrieve, unambiguous about what you are, and corroborated somewhere the model already trusts. What changes is which surfaces matter; what does not change is that being the clearest answer wins.
Scope
What is included
AI crawler access
Checking that OAI-SearchBot, PerplexityBot and Google's crawlers can reach your pages, and that your CDN or firewall is not quietly blocking them.
Answer-first restructuring
Rewriting key pages so the heading asks and the next sentence answers. This is the single highest-return change for retrieval and it also helps human readers.
Entity and structured data
Organization, Service, FAQ and author markup, with consistent facts about the company across your site and the places it appears elsewhere.
Original information
Research, benchmarks, frameworks and data an AI system has a reason to cite, because everything generic is already inside the model.
Authority beyond your domain
Mentions, coverage and profiles on the sites that get read in your category. A hundred articles on your own site does not make you an authority.
Visibility tracking
A fixed set of buyer questions run monthly across ChatGPT, Perplexity and Google AI, recording mentions, citations, accuracy and which competitors appear instead.
How it runs
Weeks one and two
Crawler access checked and fixed. A question set of twenty to fifty buyer questions is built and run across the assistants to record where you stand today.
Months one to three
Answer-first restructuring on the pages that matter, entity and structured data work, and the first original piece nobody else can publish.
Month four onwards
The question set is rerun monthly. Work concentrates on the questions where competitors appear and you do not.
What you get
- A crawler access audit, with the fixes applied
- A baseline visibility report across three assistants
- Restructured pages, answer-first
- Structured data and entity consistency work
- A monthly tracking report on the same question set
How it is judged
- Mentions of your company by assistant and by question
- Citations linking to your domain
- Accuracy of how you are described
- Which competitors appear on questions you lose
- Referral traffic arriving from AI assistants
Questions we get asked
Can you guarantee we get cited by ChatGPT?
Is this different from SEO?
How do you measure something this new?
Should we block AI crawlers to protect our content?
How long before this shows anything?
Is this different from SEO?
Can you make ChatGPT recommend us?
What if an AI says something wrong about us?
Next
Find out whether you are being cited today
Send us twenty questions your buyers would ask. We run them across ChatGPT, Perplexity and Google AI and send you back what they say, who they name, and whether the description of you is correct.