Why KnowStack? The Complete Workflow How Features Connect Use Case Examples Getting the Most Value
Creating Your Account First Login & Dashboard The General Flow: From Data to Knowledge Understanding Navigation
Website Crawling Email Integration Document Upload Meeting Transcripts Telegram Integration Slack Integration
Generating a Knowledge Base Iterative Generation
Viewing & Navigating AI Search Editing Content Version Management Exporting Access Control
Inviting Team Members Roles & Permissions
Profile Settings Company Settings Notifications
Plans & Pricing AI Usage & Billing
Data Protection Account Security Secure Integrations
Integration Overview Managing Integrations
Tracking Activity
Getting Help Common Issues
Common Questions

Iterative Generation

Iterative generation lets you update and expand an existing Knowledge Base with new data instead of starting from scratch. This is how you keep your KB current as your business evolves.

How Iterative Generation Works

When you select an existing KB as a source in Step 1, the wizard enters 'iterative mode'. In this mode, the AI merges new data with your existing KB content, preserving your previous structure and any manual edits you have made.

  1. Go to Content Generation.
  2. In Step 1 (Choose), you will see your existing KBs listed as a source alongside your data sources.
  3. Select the KB you want to update as a source.
  4. Also select any new data sources you want to incorporate (new emails, new documents, etc.).
  5. Proceed through Review and Generate as usual.
  6. The AI will create a new version of your KB that merges old and new content.

Conflict Resolution

When new data contradicts existing KB content, KnowStack's conflict resolution system helps you decide what to keep. After an iterative generation, if conflicts are detected, a Conflict Wizard will appear when you view the new KB.

The Conflict Wizard shows you each conflict side by side -- the old content and the new content -- and lets you choose which version to keep, or merge them manually.

Iterative generation creates a new KB version. Your original KB is never modified. You can always switch back to a previous version using the version selector on the Knowledge Base page.

Best Practices for Iterative Updates

  • Run iterative updates regularly (weekly or monthly) to keep your KB current
  • Include only genuinely new data -- re-including old data wastes processing tokens
  • Review the conflict wizard carefully to ensure the best content is preserved
  • Edit the merged KB after generation to fine-tune any AI decisions
  • Use the output language from your original KB (it is automatically preserved in iterative mode)