NYTM GUIDE · SEPTEMBER 2026
Clean a recurring CSV without losing useful rows
Handle quoted fields, choose the right duplicate rule, and export a checked file for the next step.
Open CSV Cleanup Workbench →1. Keep a source copy
Start from the original UTF-8 CSV export. NYTM works on a copy in memory. If your data contains names with accents and they appear corrupted, re-export with UTF-8 encoding instead of cleaning an already damaged file.
2. Match the delimiter and header
Choose comma, semicolon, or tab to match the export. Mark First row is header only if the first row contains column names. Quoted commas and line breaks inside fields are supported. An inconsistent column-count error usually means the delimiter or source needs checking.
3. Choose what a duplicate means
Whole-row deduplication removes exact matching rows after the selected trimming step. A column rule instead keeps the first occurrence of each value in that column. For example, using a customer ID can discard later updates for the same customer. Review whether that is appropriate before exporting.
4. Inspect the preview and columns
Check the row count, keep the columns needed downstream, and rename headers for your destination system. The preview shows a sample; the download includes the complete cleaned output. Preserve leading zeros in identifiers by importing those columns as text in your spreadsheet.
5. Export for the destination
Spreadsheet-safe export prefixes values that may be interpreted as formulas with an apostrophe. This can also turn negative numbers into text. Keep it on for unfamiliar spreadsheet data; consider disabling it only when you understand the data and the destination requires the original values.