Clean only what you can review
This cleaner is intentionally rule-based. It can trim surrounding whitespace, remove empty rows or columns, normalize headers and deduplicate records. It does not invent missing values, guess data types or rewrite dates.
Deduplicate by the fields that define a record
Exact duplicate rows are only one kind of duplicate. Sometimes the same customer appears twice with a different timestamp or note. Select the columns that define identity—such as ID or email—and choose whether the first or last matching row should be kept.
Empty rows and empty columns are different
An empty row contains no useful values across its fields. An empty column contains no useful values down the entire dataset. They can be removed independently so you do not lose structure you intended to keep.
Malformed rows are warnings, not silent fixes
If a data row contains a different number of fields than the header defines, TaskThimble shows a format warning. Review that issue before assuming a clean export is semantically correct.