Overview
The Gumloop Analytics agent connects to your organization’s data in BigQuery and answers questions in natural language. Ask about credit usage trends, top workflows, active users, or any other organizational metric, and the agent returns results as tables, charts, or CSV exports.
Credit Tracking
Monitor credit consumption across users, workflows, and agents over any time period
Usage Insights
Understand which workflows and agents are most active and who is using them
User Activity
See which team members are running workflows, chatting with agents, and consuming credits
Visual Reports
Generate charts and download CSV exports for stakeholder reporting
How to Access
In Chat
Organization Analytics is available in the Gumloop Chat interface. When you start a new conversation, select the Gumloop Analytics agent to begin querying your organization’s data.In Slack
You can also use the Gumloop Analytics agent directly in Slack:Add the Gumloop bot to your channel
Invite the Gumloop bot to the Slack channel where you want to use analytics.
Enable the analytics agent
Type 
/gummie add analytics in the channel to activate the Gumloop Analytics agent.
What You Can Ask
The analytics agent has access to the following data about your organization:| Data | What It Covers |
|---|---|
| Workflow Runs | Run history, credit costs, execution counts, completion timestamps |
| Agent Chats | Chat sessions with agents, credit costs per chat, chat volume over time |
| Agents | Agent names, descriptions, models used, tools configured, creator info |
| Workflows | Workflow names, descriptions, creator info |
| Users | User emails and activity across your organization |
Example Questions
Credit usage:Data Access and Permissions
Organization Analytics enforces role-based access to ensure data security:| Role | Data Scope |
|---|---|
| Admin | Full access to all organization-wide data across all users |
| Manager | Full access to all organization-wide data across all users |
| Member | Personal data only — can only see their own workflow runs, agent chats, and credit usage |
Security
Organization Analytics is built with multiple layers of data protection:- Organization isolation: Every query is automatically scoped to your organization. The agent cannot access data from other organizations, even if prompted to do so.
- Parameterized queries: All queries use parameterized SQL — user input is never interpolated into query strings, preventing SQL injection.
- Schema validation: The agent can only query pre-defined tables and columns. It cannot run arbitrary SQL or access tables outside the analytics schema.
- Role-based scoping: Non-admin users are automatically filtered to their own data at the query level, not just at the display level.
- Prompt injection protection: The agent is designed to refuse attempts to bypass data access restrictions through prompt injection, role-play scenarios, or other techniques.
Credit Usage
Queries made through the analytics agent consume credits based on the amount of data scanned in BigQuery. The agent is optimized to minimize data scanning by:- Using aggregation queries instead of raw row dumps
- Applying automatic partition filters (defaulting to the last 90 days for time-series tables)
- Limiting result sets to only the data needed to answer your question