On this page Jump to section
Published November 4, 2026
Question and scope
When a journal entry is reversed, do the supporting attachments remain linked to the original entry and readable by an independent reviewer? A reversal can clean up the ledger while leaving the evidence behind in a place nobody can find. The result is an entry that is technically corrected but practically unexplainable. This test asks whether the attachment record survives in a usable form. It does not decide whether the reversal was correct or whether the attachment was sufficient. The narrow purpose is to compare the pre-reversal and post-reversal evidence paths.
Population and unit
Start with a dated list of reversed journal entries for the review period. Record the matter or period, the original entry, the reversal entry, the preparer, the approver, and the attachment location. Keep the list stable and document any later addition. The unit is one original journal entry with at least one attachment. An entry with several attachments has several evidence objects but one unit. If the reversal created a new entry with copied attachments, keep the copy linked to the original rather than treating it as a separate source. Exclude entries with no attachments and entries that were deleted rather than reversed. Record those exclusions because a deletion is a different evidence problem.
Definitions and measures
An attachment survives when the original entry or its reversal still provides a readable link to the same document, and the document can be opened by a reviewer who did not upload it. An attachment is orphaned when the link remains but the file cannot be opened. An attachment is lost when no link remains. For each entry, record the attachment type, the upload date, the uploader, the pre-reversal link, the post-reversal link, and whether access was tested with a second account. Report exact survival, orphaned, lost, and unverified separately. Also record whether the reversal changed the attachment description or the period reference. A description change can be legitimate, but it should be documented so the reviewer knows which document belongs to which entry.
Test design
Use a census when the number of reversed entries is small. For a larger set, select a documented sample that includes entries with multiple attachments, entries reversed after a period close, and entries uploaded by a different preparer. Freeze the sample before review. Have one reviewer compare the records and a second reviewer check a subsample. Record disagreements before resolving them, because a difference in what counts as accessible can change the result. Keep the original links and the comparison sheet separate. Where the file cannot be opened, capture the error and the account used. Do not assume the document is fine because the link appears in the list.
Evidence standard
Each entry needs the original reference, the reversal reference, the attachment link, the upload date, the uploader, and the access test result. If the attachment was moved to another system, record the new location and the person who approved the move. Preserve the original attachment and any replacement. Do not overwrite the file to fix a broken link. If a replacement is necessary, record why the original could not be recovered and who accepted the substitute. Keep sensitive documents in the approved system. The analysis file should reference the document rather than copying its contents into a spreadsheet.
Interpretation limits
A surviving attachment does not prove that the reversal was correct. It shows that the evidence path remained usable. The accounting decision still requires the source document and the approver's review. A lost attachment does not prove that the entry was unsupported. The file may exist in another system, or the upload may have failed. Report the observed state and the evidence available. Avoid using the result to rank preparers. The same reversal can behave differently depending on the attachment system and the period. The useful question is whether the reversal checklist includes an evidence check.
Review cadence
Run the test after any reversal that affects a closed period, and summarize the results at the next close review. A short check catches the immediate break, while a periodic review catches patterns across several reversals. End each review with a decision record: add an attachment check to the reversal procedure, require a second reviewer, change the storage location, or leave the process unchanged. Name the owner and the recheck date. Archive the sample and the notes so the test can be repeated.