Answer page: answer the buyer's question in the first sentence
One page, one question, answered in the buyer's own words before anything else.
Flemming Rubak · November 18, 2026 · 5 min read
Early in the decision, buyers ask AI plain questions about what you do. In UK HR software they ask “Are there HR platforms designed specifically for UK compliance requirements?”1 Behind the scenes, AI turns each question into searches of its own: 2.4 on average.2
How well the title matches those searches is an important factor in which page ChatGPT cites.3 An answer buried in a services page is hard for the model to find and lift.
An Answer page gives one clear answer to one question, in the first sentence. Here is what one looks like.
Example: Fernway IT Support (a fictional IT support company)
Title
Do you support companies whose staff work from home?
Why it’s here
It is the buyer's question, word for word. That is what AI searches for.
Opening line
Yes. Fernway supports companies of 10 to 200 people wherever their staff work: remote fixes start within 30 minutes, and a replacement laptop is posted the next working day.
Why it’s here
The answer comes first, with the facts that make it specific. AI lifts the opening as the answer.
Updated date
Updated 6 November 2026
Why it’s here
Answers change when the offer changes. Three in four pages AI cites were updated within the last year.4
The details
| Service | At the office | At home |
|---|---|---|
| Remote fixes | Within 30 minutes | Within 30 minutes |
| On-site visits | Next working day | Not included |
| Replacement laptop | Same day | Posted the next working day |
| Security setup | Included | Included, on company laptops |
Why it’s here
It answers the next question before it is asked, in a form the model can read line by line.
The limit
“We don’t support personal devices. Staff need a company laptop.”
Why it’s here
A clear limit makes the yes believable, and stops AI from recommending you to the wrong buyer.
Follow-up questions
Do you work outside office hours? Yes, from 7:00 to 20:00 on weekdays. Urgent cases are handled at weekends too.
How do new employees get set up? We post a ready laptop before their first day. They log in and work.
What does it cost per employee? From €35 a month per person, all of the above included.
Why it’s here
Each one is another search the page can answer. FAQ sections marked up with schema go with more AI citations across engines.5
Next step
See what Fernway’s support costs
Why it’s here
It moves the buyer from the answer to the offer.
Build yours
- Collect: the facts that answer the question: what is included, numbers, limits and price. The product or service owner holds them. Price questions get their own Offer page.
- Outline: the short answer, the details, the follow-up questions buyers ask next.
- Four versions:
- Single question: one question, answered in the first sentence. The example above.
- Buyer situation: the same answer for one kind of buyer, such as "for a 50-person company that…".
- How-to: a procedure, step by step, for buyers who ask how something is done.
- Definition: what a term means in your category, for buyers who ask "what is".
- The page mechanics: how to mark up the page, structure the headings and show the update date are in the techniques: schema, headings and update dates.
Put it where AI looks
- Your help centre or FAQ, with the question as a heading and a link to the Answer page.
- The same answer where buyers ask: a short LinkedIn post or a forum answer by a named employee, linking the page.
- Check after eight weeks: ask the buyer’s question again and see whether your answer now appears.
Questions buyers ask
What heading structures work best for AI-readable content?
Headings that say what follows, in the words people search: the page title as the question, one H2 per follow-up question, and the answer in the first sentence under each. AI reads a page in parts, so each part must answer on its own. More in heading hierarchy for AI.
What kind of data does AI look at when deciding which brands to include in an answer?
Pages it can find through ordinary search: 88 in 100 of ChatGPT's citations come from the general search index.3 On those pages, it looks for facts it can check: numbers, dates, names and limits.
How do AI models decide which brands to mention in their answers?
They match the buyer's question to pages that answer it, judge the brands on criteria such as cost, proof and fit, and name the ones whose evidence they find. More in how AI models decide which brands to recommend.
How Seedli does this for you
Seedli finds the plain questions buyers ask AI in your market, and the ones nobody answers well. It recommends the content type that fits each gap and writes a brief to match: the facts to collect and who holds them, the outline, and what helps AI find the page: the title, the meta description, the page address, answers to the follow-up searches AI runs behind the question, and the links and link text to use from your other pages. Eight weeks after placing, Seedli checks whether the answers changed.
Good to know
- One question per page. When buyers ask many, give the important ones their own page, and keep a short questions section on each.
- Update the page when the answer changes.
Sources
- The Sunday Shortlist, UK HR software, Recommendation Design, October 2026.
- Josh Blyskal, Profound, I analyzed 250M AI responses, 2025.
- Ahrefs, Why ChatGPT cites one page over another, 2026.
- Seer Interactive, How content recency impacts AI citations, 2026.
- HubSpot, Content formats that earn AI citations, 2026.