Real CSVs are rarely clean. Headers have typos, rows have wrong column counts, values are duplicated, dates are inconsistent. The tools below find the problems, then fix what's fixable — without leaving the browser.
The validator runs nine detectors — missing values, type mismatches, duplicate rows, empty columns, duplicate headers, broken rows, invalid quotes, empty rows, and bad delimiters. Each issue is ranked by severity.
Each of these takes a CSV in and gives a CSV out — chain them together in the pipeline if you have multiple steps.
The pipeline mode lets you stack validate → dedupe → filter → sort → SQL → split into a single pass that re-runs every time you tweak a step.
Every cleaning tool has a 'send to dashboard' link that hands the cleaned CSV off to the dashboard builder for full visualization.
Missing values, type mismatches in numeric/date columns, duplicate rows, empty columns, duplicate or blank headers, broken (uneven-column-count) rows, invalid quote characters, empty rows, and likely bad delimiters (e.g. semicolon-separated mislabeled as comma).
Yes — the dedupe tool lets you tick which columns count for matching. Leave all unchecked to dedupe on entire-row identity.
No. The original file stays where you dropped it from. The tools output a new CSV that you choose to download or send to the dashboard.