Document Batch Reconciliation: Free Output Check CSV
Document batch reconciliation means matching each expected document to its actual output file. A folder containing 100 files does not establish that the correct 100 records were processed. One duplicated record and one missing record can leave the total unchanged.
Download the free output reconciliation log (CSV). It contains three invented examples and a blank row. This is a manual review worksheet, not a file scanner or an automatic feature of Docs from Excel. No account or email is needed to download it.
Write the expected list before generating files
Start from the approved input export for this run. Assign a batch ID and a document ID that identifies what one output represents. For one certificate per attendee per course, a person's name alone may be insufficient: the same attendee can have two valid certificates for different courses.
Keep the source reference stable. A row number can help locate a record in a frozen export, but it can change after sorting or deleting rows. Record the export version and the business identifier as well. Save that export so that a later investigation refers to the same input you actually used.
How to fill the worksheet
- Batch ID and document ID: identify the run and the expected document.
- Source reference: identify the original export and record.
- Expected filename: record the naming plan before generation.
- Actual filename: copy the name of the file you found.
- Identity check: record whether the document contents match the intended record.
- Result: mark Match, Missing, Duplicate or Needs review.
- Reviewer and action: record who checked the output and how to resolve exceptions.
The example rows show a matching document, a missing output and an output whose identity still needs checking. Delete the examples before using the log for real records. Avoid adding personal information you do not need; a document ID can often identify the task without copying a full address into another file.
Check identity as well as the filename
Open an output and compare its identifier and important fields with the source. A correct filename can still contain the wrong recipient or an old amount. Conversely, an unexpected filename can hold the correct document. Record these as separate observations instead of assuming that one proves the other.
Check unexpected files too. If a generated file has no place in the expected list, investigate whether it came from another batch, an extra input record or a rerun. Keep the new outputs in a separate folder during review so that old files do not make a missing result appear present.
Resolve exceptions without losing the audit trail
For a missing document, first confirm that the source record was selected and included in the actual export. For a duplicate, check the intended rule for one document before removing anything. The duplicate-recipient guide explains why matching names do not always mean duplicate documents.
If you rerun a subset, give the rerun a distinct reference and record which output replaces which earlier file. Avoid distributing both versions. Preserve the first review result and add the resolution so the log explains why a replacement exists.
Keep identifiers intact when opening CSV files
Identifiers with leading zeros or many digits should be treated as text. Excel can convert numerical text when importing it; Microsoft documents how to select text data types in its guide to keeping leading zeros and large numbers. Changing the display after information is lost does not recreate the original identifier.
Use this after a small test batch
Start with the mail merge test checklist before running a full batch. Then use this log to account for the resulting files. If the naming plan needs work, read custom filenames for merged documents.
You can try invented records from the certificate starter kit, or open Docs from Excel with your prepared input. Check the current download terms in the tool before purchasing. Completing this worksheet records your checks; it does not automatically validate every field in every output.