research

Measuring the time to revoke access after a departure

A method for timing the gap between a departure and the removal of every account, session, and token, using identity and ticket records the company already keeps.

Short answer

Use this benchmark to size repeatable IT work, set the review cadence, and decide what stays with the technical owner before assigning the workflow to an IT virtual assistant.

Research playbook

MeasureVolume and handling time
OwnerTechnical manager validates
Risk ruleName sensitive access
RefreshQuarterly benchmark review
On this pageJump to section

Key stats

Sample floor20 departuresStudy protocol
Timing milestones2 per pathStudy protocol
Reporting groupsAccess typeStudy protocol

Key takeaways

A departure creates a race between the exit process and the accounts the person still holds. The manager knows the last day, the identity team disables the main account, and a handful of applications keep working because nobody linked them to the leaver record. This study times the gap between the end of work and the removal of each access path. The result shows which systems close cleanly and which depend on someone remembering.

Question and scope

This study asks how long each access path remains usable after a person's last working day, and which paths account for the longest gaps. The unit is one access path attached to one departure, such as a directory account, a mail delegation, a virtual private network profile, or an application role. The population is every voluntary and involuntary departure in a declared quarter. The result describes that environment and that period, not the performance of an individual technician.

Freeze the departure list and the quarter before exporting data. Include contractors and temporary staff whose engagements ended, because they often carry the same accounts as employees. Exclude a person who moved to a new role inside the company, since that is a transfer rather than a departure. Record the exclusion and the reason so a reviewer can rebuild the population.

Define the revocation event

Define the start as the end of the person's last scheduled working hour, recorded from the HR or contract system. Define the end as the first moment the access path is confirmed unusable, not the moment someone clicked disable. The distinction matters because a disabled account can retain active sessions and tokens. A revocation is complete only when the access attempt fails.

Record the two timestamps separately for every path, because the disable action and the effective removal can be hours apart. Note the person who performed the action and the ticket or change that records it. If the access path has no reliable timestamp, mark it as unmeasured rather than estimating from memory. Unmeasured paths belong in the report, not in the average.

Build the identity frame

Export the directory accounts, group memberships, application roles, mail delegations, and token grants associated with each leaver. Add the local accounts on managed devices and any shared credentials that the person could reach. Include the accounts held in a vendor portal if the person used one on the company's behalf. The frame should list the access paths, not only the applications.

Link each path to an owner who can confirm whether it was removed. The identity team owns the directory, but the finance system owner is the only person who can verify a finance role. Record the owner beside the path before sampling. A finding sent to the wrong owner will be argued about instead of fixed.

Sample departures

Select a random sample of departures across departments, employment types, and months, with at least twenty cases for a first estimate. Oversample involuntary departures if the company has few of them, because those carry a higher risk. Record the sampling rule and the full population count. A sample drawn only from the most recent month may miss a seasonal account that nobody reviews.

For each sampled departure, pull the HR end date, the manager's exit ticket, and the identity change log. Compare the three dates before timing anything. A mismatch between the HR date and the manager's ticket is itself a finding. Resolve the official date with the HR owner rather than choosing the one that makes the result look better.

Separate disable from removal

Disabling a directory account does not remove the sessions, tokens, mail delegation, or application roles that were granted separately. Time the disable action as one milestone and the effective removal as another. Report both numbers so a fast disable does not mask a slow cleanup. The gap between them is often where the real exposure sits.

Check whether the identity provider revokes refresh tokens when an account is disabled, and whether the application honours that revocation. Test with a sample session where it is safe to do so. Record the test method and the result. A policy statement is not evidence that the session actually ends.

Measure the residual access window

Calculate the hours between the end of work and the confirmed removal for each path. Report the median, the longest gap, and the share of paths that exceed the company's target. Group the results by access type so a pattern is visible. A single average hides a directory that closes in minutes and a vendor portal that closes in weeks.

Check whether the gap includes waiting time, such as an approval from an application owner. Separate the technical removal time from the waiting time, because the fixes are different. A slow approval needs a process change, while a slow removal needs automation or a checklist. Report both phases for each path.

Compare with the documented target

Find the company's written revocation target, including any rule that applies to privileged or financial access. Compare the measured gaps with that target rather than with an industry claim. If no target exists, propose one based on the risk of the access path and record the proposal. A measurement without a standard cannot decide whether the process is working.

Note where the policy is silent, such as shared mailboxes or vendor portals. Silence is a finding because the owner has no rule to follow. Recommend a target and an owner for each gap in the policy. Keep the recommendation separate from the measurement so the two are not confused.

Report to the owner

Lead with the median and longest revocation gaps, then show the paths that exceeded the target. Add one example of a departure where a path stayed open long enough to matter. Keep the individual names out of the summary and use case numbers. The report should be readable by an operations manager who does not know the identity console.

Ask for a specific decision, such as an automated leaver workflow for the directory or a quarterly review of vendor portal accounts. Name the owner, the action, and the review date. If the decision is to accept the current gap for a low-risk path, record that choice with the evidence. A documented decision prevents the same finding from being rediscovered next quarter.

Keep the underlying case list with the report so the owner can check any figure that looks surprising. Show the paths that were unmeasured beside the measured ones, because a gap in the evidence is part of the result. Give the owner a short explanation of how the timing was derived. A report that can be reproduced is easier to defend in a later review.

Keep the daily log and the matching notes with the report so a reader can rebuild the count. Show the days with incomplete logging rather than averaging them away. If a channel was used by only one team member, say so, because the pattern may not apply to the whole group. The report should make the limits of the evidence visible beside the headline figure.

Keep the source documents linked from the report rather than attached as loose files. A reviewer should be able to open the contract or invoice behind each figure. Note the products where the evidence is incomplete and the owner who can supply it. The report is a working record, not a one-time presentation.

Limits and next use

The study measures access paths that appear in the exported records and can be confirmed. It does not capture a credential that was shared verbally, a personal account used for company work, or a session that was never logged. It also cannot prove that no one used the access during the gap. State those limits so the result is read as a lower bound on exposure.

Repeat the measurement after the leaver workflow changes and compare the same path types. If the directory gap falls but the application gap stays flat, the next fix belongs with the application owners. Record the change date so the two samples are not mixed. The method becomes a control when it runs on a schedule rather than once.

Sources and next step

Use the NIST Cybersecurity Framework 2.0 and the CISA Cross-Sector Cybersecurity Performance Goals as primary references for the identity, platform support, request intake, data export, and licence inventory controls discussed here. The NIST SP 800-53 Rev. 5 supports the access and audit record checks, and the GAO data reliability guidance supports the sampling and verification approach used in these methods. Apply your own policies, contracts, and system documentation when adapting the method.

Compare this study with the access revocation measurement and the software licence entitlement audit before combining them into one review cycle. If the measurement work is recurring and your technical owner can define the scope, review the service areas and contact ITVirtualAssistant to discuss a limited pilot.

Benchmark brief

What this research page must produce

Working number

A practical estimate for volume, review time, escalation rate, and assistant capacity.

Operating boundary

A clear split between routine support, preparation work, and technical ownership.

What the measuring the time to revoke access after a departure data shows

Treat this as a planning benchmark, not a universal number. Compare the benchmark against your ticket volume, SaaS stack, documentation backlog, and support risk before assigning recurring work.

The useful output is a decision about capacity, not a static statistic. If the workflow is high volume and low judgment, an IT virtual assistant can absorb coordination and upkeep. If the workflow is low volume but high risk, keep it with the technical owner and use the assistant only for preparation, reminders, and documentation.

Workflow

Recommended operating workflow

1

Collect a baseline

Pull the last 30 to 90 days of examples related to measuring the time to revoke access after a departure, including completed work and unresolved exceptions.

2

Classify the work

Tag each item by routine admin, manager approval, technical decision, security risk, or vendor dependency.

3

Set the operating number

Use the median weekly volume and review time to decide how many assistant hours the workflow deserves.

4

Refresh the benchmark

Recheck the numbers quarterly so tool growth, new systems, and security requirements do not silently change the scope.

Decision rules

MetricUse it to decideManager action
Weekly volumeWhether the workflow is worth assigning as recurring assistant work.Approve a weekly capacity target and backlog threshold.
Access sensitivityWhether the assistant can work directly or only prepare review notes.Set least-privilege permissions and removal dates.
Escalation rateWhether the workflow is stable enough to delegate.Rewrite the SOP when exceptions exceed the agreed threshold.

Consolidated statistics

StatisticFigureSource
Sample floor20 departuresStudy protocol
Timing milestones2 per pathStudy protocol
Reporting groupsAccess typeStudy protocol

Sources

  1. NIST SP 800-53 Rev. 5Account management, access revocation, and audit record controls.
  2. NIST SP 800-63B, Digital Identity GuidelinesAuthenticator and session management guidance for account lifecycle events.
  3. CISA Cross-Sector Cybersecurity Performance GoalsIdentity and access management goals for small and mid-sized organizations.
  4. NIST Cybersecurity Framework 2.0Identity management and access control outcomes for the workforce.
  5. GAO Assessing Data ReliabilitySampling and record reconciliation checks for the departure frame.

Measurement checklist

FieldWhat to captureOwner
VolumeWeekly request count, backlog age, and repeat issue patternsAssistant prepares, manager reviews
RiskAccess level, customer impact, security sensitivity, and approval needsTechnical owner
CadenceDaily, weekly, monthly, or quarterly review rhythmManager
EvidenceSample tickets, logs, screenshots, and before-after examplesAssistant collects, owner validates
EscalationTriggers, approval path, response time, and stop-work rulesTechnical owner

How to read the result

A good research page should leave the manager with a working number and a clear boundary: what the assistant can do every week, what the assistant can prepare for review, and what must never move without the accountable technical owner.

Source and refresh note

This planning page is dated for 2026 and should be refreshed quarterly as tool stacks, ticket patterns, and security expectations change.

How should teams use this benchmark?

Use it to define task volume, access limits, review cadence, and escalation rules before assigning work.

Get free benchmark review