Sorting one column independently can separate names, IDs, dates, and amounts that belong to the same record. Stop editing if the relationships already look wrong; another sort may make recovery harder.
Recover the record relationships first
If the accidental sort just occurred, consider Undo while checking what other changes it will reverse. Otherwise compare version history with the current file and preserve newer work before restoring anything.
Do not reconstruct relationships by sorting each column separately. Equal-looking orders do not prove that a customer’s amount still belongs to that customer, especially when duplicate names or dates exist.
Select the complete record range
Include every column that belongs to each row. Use Data > Sort range > Advanced range sorting options, and identify the header row when present. Choose the primary sort column and any additional tie-breaker.
Keep totals and unrelated side tables outside a record-only sort. If sorting the entire sheet is appropriate, use the sheet-level command deliberately rather than assuming every nearby block shares the same row structure.
Verify identity after sorting
Compare several stable record IDs against their original associated fields. Include duplicate sort keys, blank values, and the first and last rows. Row count alone cannot reveal a detached address or amount.
Review formulas that refer to fixed cell positions, because a positional reference is not an ID-based relationship. For collaborative analysis, a filter view may provide the required order without changing everyone’s working view. Keep a recoverable version before large structural changes, and use one row per complete record so later editors can identify the safe sorting boundary.
References
Reviewed against the linked official product guidance on November 3, 2026. Features, limits, screen labels, and access can vary by account, platform, region, and organization policy.