Sensitive data: markings and what happened to them
Insights > Sensitive data: the four headline numbers, detections over time, outcomes, per platform and data type, departments and the privacy threshold.
Monitor sensitive-data detections across your team
Insights > Sensitive data shows the markings of sensitive data in your organisation: how many, where, which types, and what people did with them. Everything is a count; the typed text appears nowhere, and a detection carries no user id.
Sensitive data in the dashboardHeadline numbers, outcomes and breakdowns per platform and per data type, for the period you pick.Period and filters
Period sets the window for the whole page: a preset from Last 7 days to Last year, or your own start and end date in the calendar. Days older than the 24-month retention window cannot be picked; the calendar says "Data from before (date) is not kept." With Department insights on, you pick a department next to it.
The Filters button has three fields that apply to every card at once:
| Filter | Options |
|---|---|
| Platform | All platforms, or one app |
| Entity type | All entity types, one type, or the custom terms (per category or All custom terms) |
| Level | All levels, Standard or Critical |
Active filters show as badges next to the button; Clear all resets them.
The four headline numbers
Each card compares with the period before.
- Detections: the total for the period.
- Critical detections: the Critical share, with its percentage of the total.
- Adjusted before sending: the share of detections the user masked or removed before sending; old replacements count too. Masking on send counts as masked here as well.
- Critical sent despite warning: critical detections that were not adjusted and still went out, with their share of the critical detections, the total sent and how many messages that covered. This is the number to manage.
Detections over time
Detections per day, split into Standard and Critical, as bars or a line. When a detection profile, a policy or a notice changes something, you see it here: detections drop, or the adjusted share rises.
Outcomes
The Outcomes card spreads every detection in the period across one bar:
| Segment | Meaning, as the screen puts it |
|---|---|
| Masked | Turned into a label like [NAME1] |
| Removed | Deleted from the message |
| Sent despite warning | Detected but sent without changes |
| No action recorded | Left in place or predates outcome tracking; not known to have been sent |
Masked and removed together form the headline "adjusted before sending". A Replaced segment appears only while the period still holds replacements from before version 5.7.1; the legend says that choice is no longer offered. The grey segment carries no judgement: it also holds detections from before outcomes were tracked.
Per platform and per data type
Detections by platform shows per app the detection count, how many were critical and what share was adjusted before sending. An app that exists in both the browser and the desktop app appears with the label Web or Desktop. Click a row and the whole page filters on that platform.
Detected entity types does the same per data type, critical types first. Custom terms appear as "(custom term)" with their category. Here too a click filters the whole page. Financial data and identification numbers are often low in volume and high in risk, so do not read only the tallest bar.
Departments
With Department insights on (Organisation > Settings), the Departments card shows per department the critical detections per person, alongside how many of those were in non-allowed AI tools. The figures are normalised per person, so a large department does not look worse than a small one by default. Click a department to filter the whole page on it.
From 5 active users
Usage figures appear from 5 active users: members who have activated the extension or the desktop app. Invited members who never started do not count, and the count does not depend on the chosen period. Below that, this screen shows the card Figures from 5 active users: "Below that, the figures describe too few people. Your organisation has 3 right now." From that card you invite colleagues; during a trial you can also switch on demo data there, or the sandbox while you are the only active user.
The threshold looks at the number of people, not at how finely you filter. What BeeSensible does make firm: no user is recorded with a detection, and there is no per-employee view. See Roles and responsibility.
The counting is on or off per organisation, with the switch Team insights under Organisation > Settings. With it off, the clients send no detection events and this screen stays empty. When there are no detections yet, the screen says "No detections yet" with the buttons Set up detection profiles and Install the extension.
