BeeSensible Spell-check for privacy
Documentation Changelog
Analytics

Reading the analytics dashboard

How to interpret detection volume, handled rates, and trends in the admin dashboard.

Analytics

Detection activity across your organisation.

Total Detections
12.438
Critical Detections
159
1.3% of total
Most Active Platform
ChatGPT
8.124 detections
Detections over time
Standard
Critical
Analytics shows which apps and categories generate the most detections.

The analytics dashboard shows how sensitive data is moving through your covered apps. The dashboard does not show statistics for any group smaller than 10 active users. That threshold is hardcoded and applies at team level and at organization level. Individual users are non-identifiable in reports by design.

Detection metadata is kept for a maximum of 24 months, rolling. The text of any individual detection is not stored.

Top of the dashboard

Four summary cards show totals for the selected time period:

  • Total detections: every sensitive value detected across all users and apps.
  • Handled rate: the share of detections where the user took an action (Remove, Replace, or Mask).
  • Active users: users who had at least one detection in the period.
  • Top app: the app with the most detections.

The date range selector in the top right controls the period. The default is the last 30 days.

Detections by app

A bar chart shows where detections are happening. Use it to identify which tools carry the most risk for your organization.

If one app dominates, decide whether the detection profile for that app is too strict, or whether the team using it would benefit from extra training.

Detections by category

A breakdown of which categories trigger most often.

  • Identity-and-contact data is usually the highest-volume category.
  • Financial and credential detections are lower in volume but higher in individual risk.
  • A spike in credential detections is worth a closer look.

Handled rate

The handled rate tells you how often people act when they see a detection. A handled rate that sits well below your expectation can mean:

  • Detection fatigue from too many low-relevance detections. Consider lowering severity for the noisy categories.
  • Confusion about what the extension is asking for. Consider a short refresher.
  • A category that should be at Warning but is at Critical, or that should be at Off entirely.

Teams tab

The Teams tab shows detection and handled rates broken down by team. The same 10-user threshold applies: a team with fewer than 10 active users does not appear in the breakdown. Use this to identify which teams have the highest exposure or the lowest handled rates, and to target training or profile adjustments where they are most needed.

Exports

Click Export in the top right to download a CSV of the current view. Exports respect the active date range and filters. Schedule monthly exports under Settings → Reports.