Research note
How often do internal job changes reach every system by the effective date?
A local method for measuring how often an approved internal job change appears in every affected system by the effective date, with a separate check for old access removal.
Published

Table of contents
Question and scope
The question covers promotions, transfers, restructuring moves, and other changes to a title, reporting line, schedule, or work location. It includes the systems that carry the role, the manager, and the access attached to the job. It excludes new hires and changes that never received written approval. The review covers one employer, one review window, and the roles that commonly change.
A change enters the sample when a written approval exists and an effective date has been set. A verbal plan without a date enters as an unapproved case. Record the source of the approval and the date it was made. If the effective date was later moved, record both dates and the reason for the move.
The stop point is the date each system shows the new role, not the date a request was sent. A system list should be fixed before the review begins. Record the first day the old access is removed as a separate measure. Report each system and the access check rather than a single combined result.
Definitions and evidence standard
Define a complete change before reviewing any case. A change is complete when every listed system shows the new role on or before the effective date and the old access is removed within the written window. Record the system list and the removal window with the case. A single system update does not complete the change.
Use the system audit log, a dated screenshot, and the change record as the primary evidence. A manager's memory can support context but cannot replace a dated entry. Mark a case as unverified when the only evidence is a verbal report. Report the share of unverified cases with the result.
Separate the record change from the access change. A directory can show the new title while the old approval group remains active. Measure both states so a fast title update does not hide a slow removal. The two results answer different questions about the change.
Sampling and measurement
Freeze the list of approved job changes before reviewing any of them. Include more than one manager, more than one system, and more than one effective month. Record the export date and the filter used. Do not remove a case because the record is incomplete; incomplete cases are part of the finding.
If the list is large, take a random sample and record the selection method. Keep the full count beside the sample so a reader can see coverage. A replaced case should be reported as a deviation. The sample should be large enough to compare at least two teams.
Measure the number of changes, the share complete in every system by the effective date, the median and longest late update, and the share with no recorded update. Add the number with old access still active after the window. Report the counts by system and approving manager. A single percentage hides which system drives the delay.
Findings and interpretation
Present the distribution before suggesting a cause. A cluster of late changes under one manager may mean the approval arrived after the date was set. A cluster in one system may mean the request has no named owner. Read the cases behind the largest cluster before naming a reason.
Compare the result with the number of systems involved. A change that touches six systems may take longer than one that touches two. Record the system count and the person who coordinated the updates. The comparison can show whether the problem is volume or handoff.
Check whether late changes were eventually completed and whether the employee was told about the delay. A late change with an explanation is different from a change that quietly stalled. Record the date of any update and the reason given. The quality of the delay matters alongside its length.
Limitations
The review measures system state, not whether the new role works in practice. A perfect update can still leave the employee without the tools or the duties described in the approval. The result describes the administrative clock, not the outcome for the employee. Report it beside a first-month check so a reader does not overread the number.
The written process may change during the window, and an older case may follow a rule that was appropriate at the time. Record the process version rather than applying the current rule retroactively. A case that missed an old standard may not be a current defect. Keep the version note with the case file.
A change that was never written down is invisible to this method. The count may understate the problem if managers move duties without a formal record. Pair the review with a sample of employees who report a recent change in duties. The answer can reveal a gap the system record cannot show.
Practical next step
Use one change record with one effective date and one owner. List every affected system and the person who will confirm each update. Send the employee a short summary before the change starts. A single record makes the change measurable.
Run a pre-change check the week before the date. Confirm the approvals, the directory entry, the access request, and the payroll input. Record each confirmation. The check should catch a missing step while there is still time to fix it.
Test the process for one quarter and compare the on-time share with the baseline. Keep the change if it improves the result without adding a new delay. If one system is always late, give that step a named owner. The goal is a complete change, not a longer checklist.
Sources
References supplied with this research note.
| Source | Publisher | Context |
|---|---|---|
| U.S. Department of Labor, Recordkeeping | dol.gov | Primary reference supplied with this article. |
| U.S. Equal Employment Opportunity Commission, Prohibited Employment Policies and Practices | eeoc.gov | Primary reference supplied with this article. |
| National Institute of Standards and Technology, Privacy Framework | nist.gov | Primary reference supplied with this article. |