This research is general operational information, not legal, tax, accounting, employment, benefits, privacy, cybersecurity, insurance, or financial advice.

Question and scope
This is a proposed research protocol, not a report of completed findings. It asks whether an approval can be reproduced from the matrix and the record. The result would describe the approval trail. Ask whether a new reviewer could act on the record without a conversation.
The protocol does not measure whether the decision was good. It measures whether the authority and the evidence are visible. Record the exception and the person who approved it. Separate what was observed from what was assumed.
Population and unit of analysis
Define the population as approvals recorded during a stated period. The unit is one approval, with the decision type, the amount, the approver, and the evidence recorded. Review the first few notes together before the process expands. Use the same format for every handover so the next reader can scan it.
Link the approval to the transaction so the same approval is not counted twice. An approval that covers several transactions should be recorded once per transaction. Keep the original note beside the corrected version. Test the rule on one real transaction before applying it to the run.
Definitions and measures
Define reproducibility as the ability of a reviewer to identify the approving role and the matrix band from the record. Define completeness as the presence of the evidence. Keep the original note beside the corrected version. A short note with four fields is easier to use than a long narrative.
Report reproducibility and completeness separately. An approval can be traceable but unsupported. Keep the original note beside the corrected version. Separate what was observed from what was assumed.
Evidence standard
Use the matrix, the approval record, and the transaction. A matrix without the approved version is weak. The reviewer should be able to see which band applied. If the source is unclear, route the question instead of guessing.
Record the date and the reviewer. A later review may need to see what was checked. Keep the boundary visible in the place where the work is assigned. Test the rule on one real transaction before applying it to the run.
Sampling and collection
Use a census of the approvals. If sampling is necessary, stratify by amount band. A sample from one band will not show whether the top band is controlled. Name the owner in the same line as the action.
Pilot the reconstruction on a few approvals and compare the reviewers. Clarify how a delegated approval is treated. Record the date and the source so the next reviewer can repeat the check. Record the exception and the person who approved it.
Analysis and reporting
Report reproducibility and completeness by amount band. Show the approvals that could not be traced. Present the results as a description of the record, not a judgment about the approver. Keep the boundary visible in the place where the work is assigned.
Limits and sources of error
The approval may be recorded in a different system. State the systems included. The matrix may have changed after the approval, which can make the record look wrong. Write the deadline as a date and a local time.
Review cadence
Repeat the review after a matrix change or at the same interval each quarter. Compare the same amount bands. Assign the follow-up to the controller and set a date. If the source is unclear, route the question instead of guessing.