QM Workflow
The QM (Quality Management) Workflow enables structured, criteria-based evaluations of documents using your dataroom knowledge. It is designed for repeatable quality assessments where documents are reviewed against a defined set of criteria.
Use Cases
Compliance checks against regulatory requirements
Document quality reviews with standardized criteria (e.g., NDA review, spelling checks)
Structured assessments with reproducible results
Multi-criteria evaluations across large document sets
Creating a QM Workflow
Create a QM Workflow
Navigate to the Workflows page from the sidebar navigation.
Click the + Workflow erstellen button in the toolbar.
In the type selection dialog, choose QM Workflow ("Quality Management workflow for systematic quality assurance").
Step 1 – Basic Information:
Workflow Name (required): Enter a clear, descriptive name for the workflow.
Description (required): Describe what this workflow does.
Model Instructions: Define the behavior and role of the AI assistant (e.g., "You are an expert for...").
Tags: Type a tag and press Enter to add it. Tags help organize and filter workflows.
Advanced Settings (expandable):
Default Model: Select a model to use. Toggle User may change model at start to control whether users can override it.
Default Data Sources: Pre-select datarooms or attachments. Toggle User may remove default data sources to control flexibility.
Step 2 – Criteria:
Click + Add Criterion to define evaluation criteria. Each criterion requires:
Criterion Name (required): A descriptive name for the criterion.
Description (required): A detailed description of what should be evaluated.
Check Mode: Select how the criterion is evaluated:
Automatic: The AI determines the best evaluation approach.
Section-based: The document is evaluated section by section.
Complete: The entire document is evaluated as a whole.
You can add multiple criteria. Each criterion is evaluated independently.
Step 3 – Review:
Review the full configuration summary (name, description, model, data sources, criteria) and click Create Workflow to finish.
Running a QM Workflow
There are two ways to start a QM Workflow:
From the Workflows Dashboard
Start via Dashboard
Navigate to the Workflows page.
Find the workflow card and click the Chat button on the card.
A new chat opens with the workflow pre-selected. Enter your prompt (e.g., "Check the NDA in detail against all criteria. Give me a tabular overview of the results.") and send it.
From any Chat
Start via Chat
Open any chat conversation.
Click the + button in the chat input area to open the actions menu.
Select Workflow starten from the menu.
In the "Workflow starten" dialog, search or browse available workflows and select the QM Workflow you want to run. The workflow badge appears in the message input area.
Enter your prompt describing what you want evaluated and send the message.
Configuration Reference
Setting | Description |
|---|---|
Workflow Name | A clear name for identification (shown on dashboard cards). |
Description | Explains the workflow's purpose. |
Model Instructions | System prompt defining the AI's role and behavior during evaluation. |
Tags | Labels for organizing, filtering, and grouping workflows. |
Default Model | The AI model used for evaluation (optional, user-overridable). |
Default Data Sources | Pre-selected datarooms or attachments (optional, user-overridable). |
Criteria | The evaluation criteria, each with name, description, and check mode. |
Sharing and Collaboration
Workflow cards on the dashboard show the sharing status. You can share workflows with other users via the Share button on each card. Shared workflows appear on the dashboards of all users they are shared with.