Questions & Knowledge Gaps
The Questions page tracks all questions your chatbot receives and identifies which ones it cannot answer. This Knowledge Gap Detection feature is one of WebChatAgent's most powerful tools for continuously improving your chatbot.
Filtering & Search
| Filter | Description |
|---|---|
| Chatbot | Filter questions by the chatbot that received them. |
| Language | Filter by the language of the question. |
| Classification | Filter by question type: Answerable or No Information (unanswered). |
| Search | Full-text search across question content. |
Question Classification
Every tracked question is automatically classified:
| Classification | Description | Action Needed |
|---|---|---|
| Answerable | The chatbot found relevant information in its data sources and could provide an answer. | No action needed — your content covers this topic. |
| No Information | The chatbot could not find relevant information to answer this question. | Add the missing information to your data sources. |
Filling Knowledge Gaps
When you find an unanswered question (classified as "No Information"), you can add the missing information directly from the Questions dashboard:
- Find the unanswered question in the list
- Click to expand the question detail
- Choose one of two methods:
Method 1: Enter Text
- Type or paste the answer into the "Add Information" text field
- The text is added as a new data source for your chatbot
- No character limit — add as much context as needed
Method 2: Upload a File
- Click "Upload a File" and select a document
- Supported formats: TXT, PDF, DOCX, MD, XLSX
- The file content is indexed and added to your chatbot's knowledge base
Both methods immediately make the new information available to your chatbot for future questions on the same topic.
Knowledge Gap Workflow
The recommended workflow for continuous improvement:
- Weekly review — Check the Questions dashboard for new unanswered questions
- Prioritize by frequency — Address commonly asked unanswered questions first
- Add missing content — Use the built-in tools to add answers directly
- Verify improvement — Check the Analytics dashboard to see if the Unanswered metric decreases
- Repeat — Make this a regular part of your chatbot maintenance routine
Best Practices
- Focus on "No Information" questions — These are your biggest improvement opportunities
- Check weekly — Regular reviews prevent knowledge gaps from accumulating
- Be thorough in your answers — When adding information, include enough context for the AI to generate comprehensive responses
- Use the Questions data as a content roadmap — It tells you exactly what your visitors need to know
- Combine with Analytics — Cross-reference unanswered questions with session data to understand how many visitors are affected
- Consider patterns — If many questions relate to the same topic, create a comprehensive data source rather than answering each question individually