SEO and content
Most content programmes are a spreadsheet of ideas and a freelancer. What works is a repeatable loop: research, write against primary sources, publish, link, and re-check what went stale.
You will recognise this if
How we do it
What you end up with
Questions this answers
Start from the questions buyers ask in the weeks before they buy, then check how well the open web answers each one. The winnable topics are answered thinly, out of date, or only by vendors pitching their own tool. Keyword volume matters after that, not before.
Yes, if the gate is stricter than a human editor would be. The engine we run on a VA loan referral site writes against primary sources, checks the build and schema, has an independent agent re-verify every factual claim, and still puts a human read before anything ships. It has shipped 22 articles that way.
The evidence says not much. Published studies found most llms.txt files receive no requests, and pages that added schema saw little change in AI citations. What does line up with being cited is being mentioned on other sites and publishing original, dated information in plain HTML.
Often because the numbers in them went stale and a fresher page replaced them, in search and in AI answers. A content engine needs a re-check loop that revisits published pieces on a schedule, updates figures from their sources, and changes the visible updated date only when something actually changed.
Updated
Copy this into your agent
It will separate the questions your buyers actually ask from the keywords a tool would hand you.
I want to find the content gaps worth writing into for my product, not just the keywords a tool would give me. Work through this with me one step at a time. 1. Ask me what we sell, who buys it, and what they were doing immediately before they needed us. 2. From that, list the questions a real buyer asks in the weeks before they would be ready to buy. Order them by how close to the purchase each one sits. 3. For each question, tell me whether the honest answer favours my product, is neutral, or favours a competitor. Do not skip the ones that do not favour me. 4. Tell me which of those questions are currently answered badly on the open web, and what makes the existing answers weak: thin, out of date, or written by someone who has not done it. 5. Pick the three I should write first and say why each is winnable for a company my size. 6. For the top one, tell me what primary sources I would need to cite to make it genuinely better than what exists. Where you are guessing about my market, say so instead of guessing.
Fifteen minutes, no deck, no prep. If it isn’t a fit we’ll say so on the call and point you somewhere better.