Your knowledge base is the heart of Didask AI. It determines the quality and relevance of the answers provided by the learning assistant and coach. Without a well-populated knowledge base, you will not be able to activate Didask's AI features.
🚀 The three sources of data ingestion
1️⃣ Building on existing studio projects
Automated integration in real-time:
All your training content created in the Studio is automatically added to the knowledge base
Updates are made in real time: every change to a training course is immediately reflected in the knowledge base
No action required on your part
2️⃣ Importing existing documentation manually
Import directly from your platform interface:
Access: “Documents” tab in the Knowledge Base section
Import methods: Drag and drop or file selection (single or multiple)
Supported formats: PDF, DOC/DOCX, XLS/XLSX, CSV, MD, PPT/PPTX, PNG/JPG/JPEG
Size limit: 30 MB maximum per file
This option is ideal for your business documents: processes, guides, internal policies, technical documentation, etc.
3️⃣ Creating automatic connections
Synchronization with your tools:
Automated ingestion from your collaborative tools (e.g., SharePoint)
Daily content updates
Setup: This feature is available upon request—contact the Didask team to learn more about options and pricing
📑 Managing document access rights
For manually imported documents, you can set specific access rights:
Restriction levels available
Public:
All users with access to the AI can view the document
Restricted
By training space: Limit access to members of certain training spaces
By group: Restrain to specific user groups
Defining access rights per document
When importing the document, define the access restrictions;
Rights apply individually to each document;
A user asking a question to the AI will only be able to access information in documents to which they have rights.
Practical example: An HR document can be restricted to managers, while a product guide remains accessible to all sales representatives.
💬 Impact on AI features
Once your knowledge base is populated:
The learning assistant can answer learners' questions based on the content of your training courses;
The coach can support your employees using all of your business knowledge;
Answers are always contextualized and based primarily on your content.
🤖 Optimize according to actual usage
Once your knowledge base is up and running, continue to enrich it based on usage data.
The “most used knowledge” ranking shows you the topics that are most important to your teams: prioritize enriching these topics with more details, examples, and practical cases. At the same time, analyze negative feedback to identify gaps or inaccuracies. Documents that generate a lot of “off-topic” questions or “incomplete answer” responses probably need to be restructured or supplemented with additional information.
This data-driven approach allows you to optimize your efforts where they will have the greatest impact on user satisfaction.
To go further:
✨ Optimization tips
💡 Start simple: If you're just getting started, your existing training courses are already an excellent foundation. Gradually add your most frequently consulted business documents.
💡 Organize your permissions: Define a clear access permissions policy from the start to avoid confusion during rollout.
💡 Quality over quantity: Focus on having fewer up-to-date, well-structured documents rather than a large volume of outdated content.
Keywords: AI Didask, coach, learning assistant, knowledge base, data ingestion
Any additional questions? Feel free to contact us at [email protected]. Our team is available to help and support you with your projects! 💬