On-demand review summarization uses a Large Language Model (LLM) to distill key themes and sentiment from your customer feedback instantly. Instead of manually exporting data for analysis, you can generate a concise overview of specific review subsets directly within the platform.
The system analyzes the most recent 100 reviews in your current view. Because the summary respects all active filters—such as entity, label, or date range—you can use this feature to identify emerging local trends or troubleshoot specific operational issues on the fly.
Generate a Review Summary
To generate an AI summary of your current review set:
- Click Reviews in the navigation bar, then click Monitoring.
- (Optional) Apply filters to narrow the list of reviews. You can filter by Entity, Folder, Label, Date Range, and more.
- At the top of the reviews list, click Summarize.
- A section appears containing the AI-generated summary, highlighting primary themes and overall sentiment for the reviews in your view.
Summary Constraints and Behavior
When using on-demand summarization, keep the following technical details in mind:
- Review Limit: The feature analyzes the most recent 100 reviews based on your active filters and sort order. If your view contains fewer than 100 reviews, it will summarize the entire set.
- Filter Persistence: The summary is context-specific. If you change your filters, you must click Summarize again to generate a new analysis reflecting the updated results.
- Granular Intelligence: Unlike account-level insights, on-demand summaries provide intelligence specific to the exact entities or categories you are currently analyzing.