Find duplicate time records before merging files
Distinguish duplicate identity from merely similar work.
Identify duplicates using a stable source key and version context. Similar descriptions or equal durations are not enough to delete a record.
Define what makes one record unique
Prefer the source record ID plus its source system and version where those are available. If a report aggregates by person and date, that combination may identify a summary rather than an underlying entry. Write the key for the actual file. A spreadsheet row number changes when the file is sorted and is not a reliable source identity.
Classify duplicates before acting
Use these fictional cases to distinguish repeated delivery from separate work.
Scroll horizontally to compare every column.
| Observed pair | Interpretation to test | Safe first action |
|---|---|---|
| W-010 appears twice with identical values | Same record delivered twice | Compare batch/source identity |
| W-010 has 30 and 45 minutes | Two versions of one record | Find the accepted correction version |
| W-010 and W-011 both say meeting, 30 minutes | Potentially separate work | Ask for context; do not auto-delete |
| Daily total plus its detailed entries | Mixed granularities | Choose one aggregation basis |
Keep repeated imports from creating new totals
Maintain a handoff register of batch ID, period, file version, recipient and acceptance status. Before retrying an import, establish whether the first attempt applied any records. A timeout or missing success message does not prove nothing happened. Test the receiving system’s duplicate behavior using fictional data rather than assuming it safely ignores repeats.
Resolve the record without erasing its history
Once a duplicate is confirmed, follow the authorized correction process and retain the reason, affected IDs and resulting total. Preserve the original file so the decision can be explained. If identity is ambiguous, isolate the records for review instead of selecting one arbitrarily. Do not use a generic “remove duplicates” command across descriptions or names as a substitute for understanding the key.
Continue this work-record check
Use the linked guide for the next decision in this workflow. Keep your original records separate from experiments and record any unresolved requirement before changing a live process.
Sources and boundaries
Primary documentation supports the dated product facts. Proposed checks and fictional examples are our editorial method, not observed product results.
- Clockify — detail · checked September 23, 2026
- Microsoft — import · checked September 23, 2026