Content gap
A content gap is a question people actually asked that you have never answered anywhere.
Why it matters
Content plans are usually built from keyword tools and opinion, neither of which observes what your visitors wanted. Gaps identified from refusals are evidence rather than estimate.
- Articles written about topics nobody asked about
- Recurring support questions with no documentation behind them
- No feedback loop between what support hears and what content covers
How gaps are identified
Every refusal is logged with the question that caused it. Similar questions are grouped, and the groups are ranked by frequency, so the most-demanded missing answer surfaces at the top rather than being lost among one-offs.
Why the phrasing is valuable
The report preserves how people asked. That wording is usually a better heading than the one you would write, because it is how the question is genuinely posed rather than how you name the feature internally.
Closing the loop
Write the page, wait for the next crawl, and confirm the question is now answered and the frequency falls. A gap that keeps climbing after you wrote a page means the page did not answer what was actually being asked.
Common questions
- Are all gaps worth filling?
- No. Some questions should stay refused — individually negotiated pricing, advice you are not qualified to give, questions about other companies. A permanent entry in the list is fine when refusal is correct.
- Is this useful for SEO?
- Unusually so. These are questions real people asked, in their own phrasing, ranked by observed demand — which is a stronger signal than search-volume estimates for a term you guessed at.
From the blog
All posts- EvaluatingAI chatbots for documentation sites: six compared on citations, gaps and price (2026)Matter Chat, DocsBot, kapa.ai, Biel.ai, Docsie and Inkeep for docs sites: section-level citations, refusal behaviour, content-gap reports, docs-generator installs and whether you can buy without a sales call. Checked on the vendors' own pages.Read
- EvaluatingThe demo is a bad predictor of AI support qualityA polished demo tells you the product can answer questions it was prepared for. The useful evaluation happens at the edges — outside the script, on your content.Read
- EvaluatingWhen “I don’t know” is the correct product behaviourAssistants that always answer look helpful in demos and dangerous in production. When refusal is the right behaviour — and how to evaluate it.Read
Answer honestly. Capture the rest.
Point Matter Chat at your site and see what it can — and can't — answer. It's honest about both.
No credit card. 2 minute setup.
Every answer cites the source it came from. When there isn't one, it says so — and hands the visitor to you.
Installs on the tools you already run.