View Analytics and Reporting
Understand how to access, interpret, and export usage analytics in Harvey, and when to use the dashboard versus APIs for reporting.
Last updated: Apr 15, 2026
Overview
Harvey provides workspace-level analytics to you understand how your organization is using the platform.
With the Analytics dashboard and reporting tools, you can:
- Track product adoption across users and teams
- Analyze usage by feature, workflow agent, or matter
- Identify engagement trends over time
- Export detailed data for reporting or analysis
- Access usage data programmatically via Application Programming Interface (API)
Note: For details on what data Harvey collects and how it is handled, refer to What Usage Data Does Harvey Collect?
Before you Start
- Workspace Admins have access to Analytics by default unless opted out of at workspace configuration
- Admins can grant roles or users the permission,
View analytics dashboardto enable access
What Reports are Available?
Our Analytics dashboard includes reports for Assistant queries over time, Active Users over time, and Users by activity. Watch the video below for a visual overview of the dashboard, and continue reading for a breakdown of each report.
Assistant Queries Over Time
This bar chart displays Assistant activity over the selected time period. Each bar breaks down queries into two categories:
- With knowledge sources — queries grounded in your connected vaults and knowledge bases
- No knowledge sources — queries without added context
Use the Daily / Weekly dropdown to switch the view granularity. The headline numbers above the chart show cumulative totals for the selected period.
This chart helps you understand how thoroughly your team is leveraging Harvey's knowledge source integrations — a key indicator of whether users are getting the most from the platform.
Active Users Over Time
Review active user counts across different windows, with the total number of active users in the period.
A line chart plots active users over time, making it easy to spot growth trends or usage dips.
Users by Activity
This table reflects each user in your workspace and a summary of their activity. By default, users are sorted by how many days they've been active. You can click any column to re-sort, or use the search bar to find a specific user.
Column | What it Counts |
|---|---|
Active days | Number of days the user logged activity in the selected period |
Assistant threads | Total conversation threads started in the Assistant |
Follow-ups asked | Follow-up questions within those threads, indicating depth of engagement |
Workflow runs | Number of custom workflow executions |
Playbook runs | Number of playbook executions |
Review tables | Review tables created or used |
Word Add-in | Queries made via the Microsoft Word Add-in |
How to View Analytics
1. Open the Analytics Dashboard
In Harvey’s web application, select Settings then select Analytics to review workspace data.
2. Review the Overview
The default view shows data across all product areas.
Use this view to:
- Understand overall usage trends
- Identify active users
- Compare engagement across product areas
3. Apply Filters
For targeted analysis, use the available filters. For example, filter by:
- Past month (default)
- Past week
- Today
- Custom — opens a calendar picker to set any date range
4. Export Analytics Data
Export reports when you need more detailed or offline analysis.
- From the dashboard, select Export
- Adjust settings as needed (such as date range)
- Select Export to download the CSV file.
Note: for definitions of each column refer to the “child attributes” in our developer documentation on Usage History.
The download will include:
- Event-level data
- Detailed logs of user activity
- Clear distinctions between:
- Creating vs. running workflow agents and playbooks
- Starting new threads vs. submitting follow-ups
Note: Dashboard views and exports use the same underlying data to ensure consistency.
Access Analytics via API
If you need to integrate analytics into external systems or automate reporting, Harvey provides APIs.
History Export API
- Includes event-level metadata
- Does not include prompts or responses
- Best for tracking adoption and usage trends at scale
- Refer to our History Exports developer guide to get started
Usage History API
- Includes usage history and metadata
- Refer to our Get Usage History developer guide to get started
Query History API
- Includes query inputs to and outputs from the AI system
- Refer to our Get Query History developer guide to get started
Important: Resource names from these APIs may include sensitive information from across your workspace. This information isn’t limited to specific clients or teams. Only access it when needed, and be careful about who you share API access with.
Tips for Success
- Identify champions: Sort the Users by activity table by active days or assistant threads to find your heaviest users — they're good candidates to coach colleagues.
- Spot under-adoption: Users with low active days or low activity in a product area, may benefit from additional training
- Track trends over time: Use the custom date range to compare month-over-month or quarter-over-quarter active user counts ahead of QBRs.
- Maximize grounded queries: Referencing Assistant queries over time, a high share of "with knowledge sources" queries suggests users are leveraging your connected vaults — aim to grow this percentage over time.