How to Batch Redact Excel and CSV Files

Use this guide for payroll sheets, customer lists, supplier records, quotation tables, procurement ledgers or exported business data that need to be shared externally, moved across teams or sent to AI.

Spreadsheet redaction is about columns, rows, sheets and repeated values. Do not rely on a few manual find-and-replace operations.

Step 1: Add spreadsheet files and choose the redaction mode

Import Excel or CSV files on the home screen. You can process multiple files in one batch. Choose detection categories and the redaction mode before starting.

Choose categories based on the file:

• HR or customer tables: names, phone numbers, emails, addresses
• Payroll or reimbursement files: amounts, dates, bank accounts
• Supplier or quotation tables: company names, amounts, invoice numbers
• Operational exports: customer names, project names, contract numbers

Add fixed customer names, project names and internal codes to the custom dictionary before detection.

Recommended redaction modes:

• Names, customers and suppliers: alias replacement.
• Phone numbers, IDs and bank accounts: partial or full masking.
• Amounts and dates: keep or mask based on the analysis goal.

If the spreadsheet will be analyzed by AI, aliases usually preserve more value than replacing everything with asterisks.

Step 2: Do not only check headers

Sensitive values may appear in columns named Notes, Contact, Owner, Address, Description or Remarks. They may also appear in multiple sheets.

After detection, review actual hits instead of assuming the header tells the full story.

Step 3: Review hits by category

Use the hit list to inspect what was detected.

Review in this order:

1. Names, phone numbers, emails and ID numbers.
2. Company names, customers and suppliers.
3. Bank accounts, invoice numbers, amounts and addresses.
4. Free-text columns such as notes and descriptions.

If a repeated value should be kept for analysis, use a stable alias instead of deleting the context.

Step 4: Search for known values

Search for a few known names, customer abbreviations, phone suffixes or project codes.

If search finds unmarked values, add them before exporting.

Step 5: Export and spot-check

After export, check:

1. The first and last rows of each sheet.
2. Notes, descriptions, addresses and invoice fields.
3. Whether filters, formulas and column order are still usable.
4. Whether original names and phone numbers can still be found.

Send only the redacted copy.