Agentic Insight Quality Overview

Agentic Insight Quality is an AI-powered collaborator built directly into the Medical Insight workflow. It acts as a real-time coach, helping users transform raw observations into structured business intelligence. By providing immediate feedback against organizational standards, the agent ensures that every insight submitted is a valuable asset for the company.

To ensure high-quality data capture, the system provides a two-step process: proactive guidance and AI-powered evaluation. Before a user begins typing, a set of Guidelines displays within the interface. These guidelines serve as a roadmap, helping users understand the necessary components of a high-quality insight before they even begin drafting their text. After the user enters their text, they can run a quality check. Agentic Insight Quality runs a real-time analysis, evaluating the draft against specific instructions to ensure the insight is actionable and compliant.

Agentic Insight Quality checks insights against the following Veeva-provided rules by default:

  • Identification
  • Observation
  • Rationale
  • Business Relevance
  • Observation Context
  • Writing Style

Admins can activate or deactivate any of these default rules. However, the Veeva-provided default rules cannot be edited or deleted.

To provide customized guidance, Business Admins can create their own rules for use with Agentic Insight Quality. For information on creating custom rules, see Custom Insight Quality Rules.

Business admins only need to create the rules in English. By default, all rules are translated into the user's language, and responses from the AI in the quality tab are displayed in the language in which the insight was entered.

Not all features are available on all platforms. For a list of available platforms for each feature, see Platform Functionality.

Accessing Agentic Insight Quality

To access Agentic Insight Quality, select a supported text field. The following entry point is supported:

  • Medical Insight - Description field