Research question and scope

This protocol measures how often a client contact's delegated mailbox access remains after the contact leaves. The unit is one mailbox-permission path tied to a named person who is no longer authorized. A path may be full access, send-as, send-on-behalf, calendar editing, or a forwarding rule. The measure describes the control around a departure, not the intent of the person who holds the access.

Set the scope to one client, one mail platform, and a departure window. Record the date the contact left, the mailboxes involved, and the administrator who can read the permission list. If the platform or the client's mail policy changes during the period, split the sample at that date. A different permission model creates a different denominator.

Keep the measure separate from a security incident count. A residual permission is a control failure even when no message was read. The purpose is to find where the departure process leaks. A neutral framing helps the client share the admin record without treating the audit as an accusation.

Population and unit of analysis

Define the population as every mailbox that the departed contact could access or act for on the day before departure. Include personal mailboxes, shared inboxes, role addresses, and distribution groups. The unit is the permission path, not the mailbox. One mailbox may carry four paths, and each should be judged separately.

Include paths that were intentionally retained for a defined business reason. Report them as approved retention with the owner and the end date. Exclude a path with no recorded approval from that group. The two categories describe different levels of risk.

Exclude mailboxes that never had a delegated permission for the contact. Report the total number of mailboxes reviewed so the reader can see the base. A rate built only from the mailboxes that had access will overstate the problem across the client estate. State the exclusion count beside the denominator.

Definitions that make the measure reproducible

Define a departure as the last authorized working day recorded by the client. Define a residual path as one that still resolves to the departed person after that day. A path that was reassigned to a new named person is not residual. Publish one example of each outcome before the sample begins.

Define an approved retention as a written decision by the mailbox owner that names the reason and an end date. A verbal request is not enough. A role mailbox that no longer belongs to any individual is reviewed separately. The classification should be reproducible by two reviewers.

Decide how to treat a forwarding rule that points to an alias rather than a personal address. Trace the alias to its current members before classifying it. An alias that still includes the departed person is residual. Record the trace method with the result.

Sampling and data collection

Draw the sample from the mail administration export for the full period rather than from a list of known removals. Record the export date, the filters, and the row count. Include every mailbox type that the client authorized for review. A sample built from removed entries cannot measure what remains.

Capture the mailbox, the path type, the grant date, the last authorized day, the current holder, the approval reference, and the review date. Add the admin console screenshot reference where the platform supports it. Store the export with the report. The permission list and the departure date are the core evidence.

Check a small overlap sample with a second reviewer. If the reviewers disagree about whether a path is residual, the definition of reassignment is too vague. Rewrite the rule and retest before the full sample. Agreement matters more than sample size for a measure that will be reused.

Record the time taken to collect the admin export. A report that requires a full day of manual work will not be repeated. Note the platform fields that were missing. The collection cost belongs beside the result.

Measuring residual delegation

Report the share of permission paths that remained after departure, the share retained with written approval, and the share removed before the first review. Show the counts behind each share. The three groups describe different controls and should not be collapsed into one rate. A single residual figure hides the difference between an approved exception and a missed removal.

Break the result down by path type and mailbox type. Send-as may be removed faster than calendar delegation because it is more visible. A role mailbox may show a different pattern from a personal inbox. Compare only groups with similar volume and review practice.

Report the age of each residual path in days after departure. A path found at day three is different from one found at day ninety. Include the review date and the client's stated review cadence. The delay distribution shows whether the problem sits in the departure event or in a later audit.

Limitations and next action

The measure cannot show whether a residual path was used to read or send mail. It also cannot show a path that was removed and later restored without a record. Treat the result as a measure of permission state, not of data exposure. Note the platforms and mailbox types that were not included.

Give the result to the client owner and the service manager with one change to test. The change might be a same-day removal step, a departure checklist item, or a weekly permission export. Record the effective date and repeat the measure with the same definitions. A controlled change is easier to judge.

Review the boundary cases before widening the rule. A few role mailboxes with approved retention may explain a large share of residual paths. Fix the classification first if that is the pattern. The goal is a clean permission state, not a lower number by definition.