After an AI agent completes its run, the results appear in the trial’s analysis section. This page explains what each part of the output means and how to use it effectively.
Understanding the Results
Agent results include several key components:
Engagement Score
A numerical score reflecting how actively the prospect is engaging with your product during the trial. This is based on session frequency, session duration, and the breadth of features being explored. A higher score indicates stronger engagement.
Sentiment Analysis
AI-analyzed sentiment from call transcripts and communications. Sentiment is categorized as positive, neutral, or negative, with specific quotes and themes highlighted. This helps you understand how the prospect feels about the product beyond what usage data alone can tell you.
Feature Adoption Metrics
A breakdown of which features the prospect is using, how often, and how deeply. This section also highlights features that are available but haven’t been explored — potential opportunities for a guided walkthrough or demo.
Risk Indicators
Flags that suggest the trial may be at risk. Common risk indicators include:
- Declining session frequency over the past week
- Key contacts becoming less responsive
- Negative sentiment trends in recent calls
- Important features not being adopted
Recommended Actions
Based on the full analysis, the AI suggests specific next steps. These might include scheduling a check-in call, sending a feature-specific resource, involving a technical specialist, or adjusting the trial timeline.
Acting on Results
The value of AI analysis is in the actions it drives. When reviewing results:
- Start with the risk indicators — address any urgent concerns first
- Review recommended actions and decide which to prioritize
- Check engagement and sentiment trends to understand the overall trajectory
- Use feature adoption data to tailor your next conversation with the prospect
Compare results across multiple agent runs to track trends over time. A single snapshot is useful, but the real power comes from seeing how engagement, sentiment, and adoption change throughout the trial.
Where Results Appear
Agent outputs feed into multiple parts of Superhawk AI:
- Trial detail page — The full analysis report for each trial
- Key Outcomes — The most important, actionable insights surfaced across all trials
- Hawkeye View — Health scores on the dashboard are derived from agent analysis
For a broader view of how your trials are performing, visit Trials or Reports & Analytics. 
