Protocol premise

The study would test whether the aging report is an evidence-backed list or a summary that depends on records outside the report.

Protocol status and question

This is a proposed research protocol, not a report of completed findings. It asks what share of aging report lines can be traced to a source document that supports the balance and status shown. The answer would tell a finance team how much of its collections view is independently verifiable. Keep the protocol and its approval date with the data file. Record who approved the question and when.

The protocol does not test whether a customer will pay. It tests whether the report line rests on evidence that a reviewer can inspect. The sponsor should name the operational decision the result could change, such as a cleanup project or a reporting control. Record who approved the question and when. A later reviewer should be able to see what the study was allowed to conclude.

Population and observation unit

Define the population as every open line in a frozen aging export for a stated date. The unit is one line, identified by customer and invoice, not one customer or one file. Lines with a zero balance or a closed status should be excluded and counted. The frozen export should be saved before any cleanup begins. Check the frame against a second source before collection.

Decide how to treat a customer with several invoices. Each invoice remains a separate line because each can have a different evidence status. A customer-level summary can be reported later, but it should not replace the line-level observation. Name the system that produced the frame and the export time. A count that cannot be rebuilt is not a population.

Definitions and measures

Define traceability as the presence of a source record that supports the invoice amount, the open balance, and the current status. The source might be the invoice, a signed contract, a remittance, or a dated dispute record. A note that says customer will pay is not a source document for the balance. Keep the definition stable for the whole window. Separate missing, unknown, and not applicable values.

Measure traceability separately for amount, status, and ownership. A line can have a valid invoice but no named owner, or a named owner but no evidence for the status. Report each dimension and the combination. The combination shows where a cleanup effort would have the most effect. Separate missing, unknown, and not applicable values.

Evidence standard

Every traceable line should link to a document that a second reviewer can open without asking the original analyst. Use a stable file reference or system link and record the date the link was checked. A screenshot of an email is weaker than the message itself, but it may be the only available evidence if the system does not retain the record. Link each classification to a document a second reviewer can open. Keep confidential records behind a keyed reference.

Mark missing evidence as missing rather than inferring it from another line. If a customer-level statement supports the total but not the individual invoice, record that limitation. The study is about the report's traceability, so the quality of the underlying document matters as much as its existence. Keep confidential records behind a keyed reference. Mark reconstructed evidence as an estimate.

Sampling and collection

A census is preferable for a small ledger. For a larger ledger, stratify by aging bucket, customer size, and owner so the sample reflects the report's composition. Select the sample before reviewing evidence and keep a log of any replacements. Do not replace a line just because its file is hard to find. Keep the sampling rule with the data file.

Have two reviewers classify an overlap sample and compare their results. Discuss the cases where one counted a document as sufficient and the other did not. Update the definitions only if the change applies to the whole sample, and record the change date. The disagreement rate is itself a useful finding. State the final sample size and how it was reached.

Analysis and reporting

Report traceability by aging bucket and by owner, with counts and rates. A high overall rate can hide a weak bucket or a single owner with many unsupported lines. Show the missing-document categories separately so the reader can see whether the gap is a filing problem or a record-creation problem. Show the distribution, not only the average. Report counts with rates and keep the denominator visible.

Present the results as a description of the report at a point in time. Avoid language that implies the customers are unwilling to pay or the team is careless. The finding is about evidence, and the recommendation should focus on the control that would make the evidence easier to produce. Avoid ranking people unless the design supports it. Name the counterexamples as well as the main pattern.

Limits and sources of error

The study may overstate traceability if reviewers accept documents that do not actually support the balance. It may understate it if a valid document exists but is not linked to the line. The date of the export matters because later payments can change the status. Describe the missing-data rate and its likely direction. Name the records the study could not see.

Confidentiality limits may prevent linking some documents in the research file. Use a reference key and keep the crosswalk in a restricted location. The report should say what was excluded for privacy and how that affects the result. Name the records the study could not see. Say what a second study would need to test.

Review cadence

Repeat the review after a ledger cleanup or a reporting change. Use the same definitions and the same aging buckets so the comparison is meaningful. A second review can also test whether the cleanup reduced the missing-evidence rate. Store the data and the protocol together for the next pass. Set the next review date before the current one ends.

Assign the follow-up to the owner of the reporting control and set a date. The protocol should produce a specific decision about which evidence to require at the next close, not a general instruction to improve documentation. Set the next review date before the current one ends. Compare the result with the prior period and explain the change. Record the decision the review is expected to inform.

Sources

  1. U.S. Government Accountability Office, Standards for Internal Control
  2. IRS Publication 583, Starting a Business and Keeping Records
  3. Financial Accounting Standards Board
How complete is the promise-to-pay record at the month-end review?How complete is the cash application record, for the month-end review?How often do credit holds lack recorded evidence, for the month-end review?

Put the protocol into operational context.

Discuss the queue, evidence, ownership, and safeguards that apply to your workflow.

Discuss your workflow