Pick exactly the columns you need from a wide CSV
Select columns to keep or specify columns to drop from a CSV, with optional reordering.
🛡️ Your rows never leave your browser.
Why Onfyt: Supports both allowlist (keep these) and denylist (drop these) modes in the same tool, with drag-to-reorder in the column picker UI.
Interactive Sample Data
| id | name | internal_id | audit_log | |
|---|---|---|---|---|
| 1 | Grace Kerubo | grace@corp.com | INT-0001 | created_by:admin |
| 2 | Peter Njoroge | peter@corp.com | INT-0002 | created_by:admin |
How this tool works
Wide CSV exports from CRM and ERP systems often contain 50+ columns, but downstream systems typically need only 5–10 of them. Passing all columns through an import pipeline wastes storage, slows processing, and may expose sensitive internal fields (audit logs, internal identifiers) to systems that shouldn't see them.
DataFix lets you specify columns to keep (allowlist) or columns to remove (denylist) and optionally reorder the surviving columns—all in one pass.
Exporting raw database tables often includes sensitive, deprecated, or irrelevant columns (such as password hashes, internal tracking UUIDs, or obsolete timestamps) that should not be shared with external stakeholders or ingested into downstream tools. DataFix provides an intuitive column selector that lets you toggle individual columns on or off, reorder retained fields, or bulk-remove unnecessary dimensions, yielding a streamlined, compliant dataset in seconds.
Data minimization compliance (such as GDPR and HIPAA) mandates stripping unnecessary personal identifiers or internal database tokens prior to sharing files with vendors or analytical teams. Toggling column inclusion extracts clean, compliant subsets in seconds.
Data privacy compliance often mandates stripping personally identifiable information (PII) before sharing datasets with external consultants. Toggling columns on or off allows data teams to generate sanitized, compliant exports in seconds without risking unauthorized data disclosure.
Things that usually go wrong
- Removing a column that is a join key used downstream will break subsequent merges.
- Column names with leading/trailing spaces may not match — use Trim Whitespace on headers first.
Edge cases we tested
🧪 Keep columns that don't exist in the CSV
Warning issued for non-existent column names; existing matches proceed.
🧪 Column name with spaces ('First Name')
Matched correctly; output header preserves spaces.
🧪 Duplicate column names in source CSV
Both columns shown in picker; user selects which instance(s) to keep.
Frequently Asked Questions
Can I reorder columns at the same time?
Yes — in 'Select' mode, drag the selected columns into the desired order before clicking Apply.
What happens if I select a column that doesn't exist?
Non-existent column names are silently ignored.
Related tools
Reshape
Transpose CSV
Flip a CSV matrix by turning rows into columns and columns into rows.
Open Tool →
Reshape
Pivot a CSV
Summarise long-format CSV data into pivot tables grouped by row and column keys.
Open Tool →
Reshape
Group By and Count
Aggregate CSV rows by one or more group-by columns with count, sum, mean, min, or max.
Open Tool →