Definition
Mean elapsed time between HR termination record timestamp and actual credential deactivation in the access control system, measured in hours.
The Measurement Gap
Measurement Gap and Decision
The Measurement Gap
Most organizations have a termination process that includes credential deactivation as a checklist item. Few measure how long the gap actually is. HR completes the termination record. Security receives a notification, sometimes by email, sometimes by system integration, sometimes by phone call. The badge gets deactivated when someone processes the request. The elapsed time between these steps is CDL, and most organizations have never measured it.
The gap matters because it compounds with other failures. If the access anomaly detection system (measured by AACI) is also generating noisy alerts, security teams may dismiss an after-hours badge read from a terminated employee as another false positive. CDL quantifies the exposure window; AACI determines whether anyone would notice activity during that window.
The Decision It Supports
CDL tells security and HR leadership whether the termination-to-deactivation integration is fast enough to meet organizational risk tolerance. A CDL of 2 hours for involuntary terminations may be acceptable. A CDL of 72 hours is not. The metric also reveals whether the gap varies by termination type: involuntary terminations should have shorter CDL than planned departures, but many organizations treat them identically in their workflows.
Formula
Plain Language
For each termination event, record when HR flagged it and when the badge was actually deactivated. Compute the gap in hours. Report mean CDL and maximum CDL, segmented by termination type.
Formal Notation
$$CDL = \frac{1}{N} \sum_{i=1}^{N} (Deactivation_timestamp_i - Termination_timestamp_i)$$
Variable Definitions
| Variable | Definition | Unit |
|---|---|---|
| Deactivation_timestamp_i | Date and time the credential was deactivated in the access control system | Datetime |
| Termination_timestamp_i | Date and time the termination was recorded in the HR system | Datetime |
| N | Number of termination events in the measurement period | Count |
Units: Hours. Track maximum CDL alongside mean, a single 200-hour lag is a greater security concern than a mean of 4 hours suggests. Also track the "never deactivated" rate separately as a critical companion metric.
Proposed formula, pending validation pilots.
At a Glance
Decision Supported
Is the HR-to-access-control integration fast enough to prevent terminated-employee access?