Skip to main content

Analytics

The Analytics dashboard visualizes tool usage trends, performance metrics, and user activity across your MCP integrations. Use it for aggregate trends over a time window; for individual events, use Logs. Open it from Monitor > Analytics.

To build your own queries and boards beyond this prebuilt dashboard, use Explore and Dashboards. Those pages query the store set by LOGS_READ_STORE. On PostgreSQL (the on-prem default) they offer only the Gateway activity (audit logs) dataset. To query the other datasets on-prem, bring your own ClickHouse and set LOGS_READ_STORE=clickhouse.

Analytics Dashboard with Start Date and End Date pickers, an Advanced Filters toggle, four stat cards (Tool Calls, Active Users, Success Rate, Avg Calls/User), a Usage Over Time chart, and MCP Servers and AI Agent donut charts

Date range and filters​

  • Start Date and End Date set the reporting window. Every metric and chart on the page reflects this range.
  • Advanced Filters narrows the dashboard by user and AI agent. Overview also supports MCP server and grouping filters; Skills & Commands adds a specific skill or command filter.
  • The refresh button at the top right reloads the data.

Each stat card shows the change versus the previous period of the same length.

Enable more data sources​

Willow gateway activity appears automatically. Open Data sources to connect agent OpenTelemetry, scanner discovery, and Guard Hooks.

See Analytics Data Sources for the source cards and configuration steps.

Summary metrics​

Four cards summarize the selected range:

MetricDescription
Tool CallsTotal number of tool calls made through the gateway.
Active UsersDistinct users who made at least one call.
Success RateShare of tool calls that completed successfully.
Avg Calls/UserAverage number of tool calls per active user.

Usage Over Time​

The Usage Over Time chart plots activity across the selected range on a dual axis: Tool Calls as bars (left axis) and Active Users as a line (right axis), so you can see volume and reach together. Use it to spot spikes, drops, and adoption trends.

Below it, two donut charts break the same activity down by MCP Servers and by AI Agent, showing each one's share of total usage (for example, Everything MCP 83% and Context7 17%, or Cursor 100%).

Breakdown tables​

The lower half of the dashboard holds detailed tables and charts. Each table has its own Export CSV action.

Lower Analytics sections showing Top MCP Servers and Top AI Agents tables, an Average Response Time chart, and Top Errors and Top Users tables, each with an Export CSV button
SectionTypeColumns
Top MCP ServersTable · Export CSVIntegration, # Tool Calls, # Users, Success Rate, Avg Time
Top AI AgentsTable · Export CSVClient, # Tool Calls, # Users, Success Rate
Average Response TimeLine chartAverage tool response time (ms) plotted across the range
Top ErrorsTable · Export CSVIntegration, Tool, # Errors, Error Rate, and a View action to inspect each error. Shows "No data available" when there are no errors
Top UsersTable · Export CSVUser, # Tool Calls, Success Rate

Guard breakdowns​

Below the usage breakdowns, Runtime Guards and Build-time Guards each have their own section and Export CSV action. These sections show No data available when there are no results for the selected range.

Adoption tab​

The Adoption tab shows how many people in the company use Willow, relative to your user directory. A person has adopted Willow when they made at least one tool call through the gateway in the last 90 days, from any AI agent. Adoption, Not Using, Connected to Willow and Onboarded always look back 90 days, whatever date range you pick. Call counts, Activity Over Time, and the Top Users and Least Active Users lists follow the selected range. The AI agent filter narrows the numbers to one agent. The user filter doesn't apply on this tab.

SectionWhat it shows
Company AdoptionThe share of users who adopted Willow, with the count of those users out of all users.
Not UsingUsers with no tool calls in the last 90 days.
Connected to WillowUsers who opened a session with Willow in the last 90 days.
OnboardedUsers who made at least one tool call in the last 90 days.
Adoption Over TimeCurrent employees who have ever made a tool call, by day or by week. Because it counts everyone up to each date, a date shows the same value whatever range you pick, and the line only goes up. Hover to see the share of headcount.
Activity Over TimeCurrent employees who made a tool call in each day or week of the selected range. The count can fall as well as rise.
Most Adopted Groups and Least Adopted GroupsFor each group, the members who adopted Willow out of all members, calls per member, and the adoption rate. Export all groups downloads every group as CSV.
Top Users and Least Active UsersTool calls in the range and each user's share of the company's calls.
All UsersEvery user with their groups, Connected and Onboarded flags, and calls in the range. Search by name, email, or group, filter by No calls in range, Not connected, Connected, no calls, or Onboarded, and select Export to download the filtered list as CSV.

If the tab says adoption analytics needs the analytics query engine, that engine isn't enabled for your organization.

MCP Server tab​

The MCP Server tab shows how one MCP server is doing. Pick a server, or open the tab from View analytics on the server's Overview tab or from its Auth errors (7d) number in the MCP server list. The date range and the user and AI agent filters apply.

SectionWhat it shows
UsageTool calls, active users, calls per user, and average response time, with a Usage over time chart.
ReliabilitySuccess rate, failed calls, and calls that failed for auth, with Errors over time, Errors by type, and Failing tools.
ToolsEach tool of the server with its number of calls and users.
Users & agentsThe top users and the AI agents that call the server.
Connector authSign-ins to the server's connector, token refreshes, and calls that failed for lack of a working credential. All connectors opens the connector auth view for every server.

If Connector auth says it needs the logs explorer store, this deployment doesn't have one.

Export data​

Select Export CSV on any table to download that breakdown for offline analysis or reporting.

What to do next​

  • Explore: build ad-hoc analytics queries with the guided builder or WQL
  • Dashboards: assemble your own visuals into a shared board
  • Logs: drill from an aggregate trend into the individual events behind it
  • Investigate with Logs: diagnose a spike or a drop in success rate
  • Radar: review flagged security and operations gaps
  • Dashboard: the at-a-glance overview of your organization