When multiple team members report lost edits after a restore, collect the reports before attempting another rollback. Similar symptoms can have different causes: one file may have reverted, another may have moved, and a third may contain work that was never uploaded.
Turn reports into a reconciliation list
Ask each contributor to supply the shared file URL, the exact missing change, when they last saw it and whether they still have a copy. Record the time zone. Deduplicate reports by location, not filename alone, so that two departments’ “Plan.docx” files are not treated as one item.
Use statuses such as “candidate found,” “content approved” and “still missing.” Do not mark a file recovered just because it is present or its modified date is recent. Assign one person to apply changes and another appropriate owner to validate important documents.
Check the first restore’s activity boundary
Compare the selected unwanted activity with the edits now reported missing. The OneDrive restore workflow undoes selected activity and later activity; it is not a merge of everyone’s preferred revisions. If useful work happened later, identify how to recover it separately.
Do not invent a uniform waiting period for all contributors or infer that a missing date means a twelve-hour snapshot failed. Record the actual browser state and error message. If the operation affected a shared library, involve its site administrator rather than repeating a personal OneDrive restore.
Resolve differences without overwriting other contributions
- Preserve the current shared file and any contributor copies independently.
- Inspect available versions for the reported content. Compare documents, not only their timestamps.
- For a deleted file, use the owning location’s deleted-item recovery path.
- Where no single candidate contains all accepted changes, prepare a new reconciled copy and ask the relevant contributors to review it.
For example, if sales recovered a revised price table while operations retained newer delivery notes, neither copy should silently replace the other. Record the accepted source for each change and the person who approved the combined result.
Escalate remaining gaps with useful evidence
Give the administrator the original URL, expected revision, checks already performed and any restore errors. Ask about actual history limits and approved backup coverage. Extending a retention policy does not recreate versions already lost or extend the built-in bulk rollback to 93 days.
Keep independent recovery copies until the team confirms the authoritative shared documents. Announce where work should resume and record exceptions that still need attention. A large-team recovery is finished when the missing contributions are accounted for, not when all users have been told to click Restore.