How to find unanswered questions and knowledge gaps
Turn real visitor questions into a prioritized, testable knowledge-base roadmap.
The Questions tool records questions and classifies whether supporting information was found. “No Information” entries are direct evidence of missing or unreachable knowledge.
Frequency helps prioritize, but the source conversation explains intent. Add the smallest authoritative fact that resolves the repeated need.
What you will have at the end
- A prioritized list of knowledge gaps
- One new text or document source
- A verified answer and documented source owner
Before you start
- Questions tool enabled
- Enough real conversations to reveal patterns
- Permission to add or update sources
Observe, understand, fill, verify
Classification identifies likely gaps; frequency ranks impact; conversation context prevents solving the wrong interpretation. A new source is not complete until indexing and retesting succeed.
Repeated variants may describe one intent, so consolidate them before creating duplicate content.
01–05
Set it up step by step
Filter “No Information” questions
Separate likely gaps from answerable questions.
Open Dashboard → Questions, select the chatbot and filter classification to No Information. Use language and search filters only after the base list is visible.
Prioritize by frequency and impact
Group variants of the same intent.
Sort or group repeated questions and estimate business impact. Fix a frequent purchase-blocking gap before a rare curiosity, while still escalating legal or safety-critical gaps immediately.
Inspect the source conversation
Understand what the visitor really needed.
Open the linked conversation and read the messages before and after the question. Record the intended answer, constraints and wording variants; do not create content from an isolated sentence alone.
Add authoritative text or a document
Use the smallest maintainable source.
Choose Add Information and enter concise text for a focused fact or upload a maintained document for a broader policy. Add a descriptive name and category so the source remains findable.
Retest and monitor variants
Prove the gap is closed without harming related answers.
Wait for indexing, reset the test chat and ask the original plus two natural variants. Confirm source evidence, then monitor whether new No Information entries for the same intent decline.
Example & result
See the practical test and its result
Every tutorial includes a fixed input, the expected outcome and a transparent record of what was actually verified locally.
Practical example: How to find unanswered questions and knowledge gaps
This exact scenario was completed with the temporary tutorial account.
Exact test input
Ask: “What is the 2028 warranty policy for Northstar Services?”
Expected result
The bot abstains safely and the question becomes visible as a knowledge gap.
What was actually verified
The live chatbot said it had no information. Questions then stored the exact question, answer and context with the classification “No Information”.
Tips & tricks
Make the setup reliable
Test with realistic examples, record your baseline and change one setting at a time. That makes real improvements visible.
Fix the source of truth
If the official website or handbook lacks the answer, update it there and re-index instead of creating a hidden duplicate.
Keep a gap backlog owner
Every unresolved high-impact gap needs an owner and due date, not just a dashboard row.
When something does not work
Troubleshooting
Check status, permissions and test data systematically before changing the model or prompt.
The expected option is missing
Confirm the account plan, feature permissions and selected chatbot. Paid or beta features can be hidden when prerequisites are not met.
The test result is inconsistent
Reset the test conversation, keep the input identical and change one setting at a time so the cause remains measurable.
Ready for a production-style test
Review No Information weekly, publish a top-gap report and add closed questions to the regression set used for model, prompt and source changes.