CSV Lead List Format for Cold Email

The basics

CSV looks simple and breaks in predictable ways. The file is plain text: a header row, then values separated by commas. Any value containing a comma, a quote or a line break must be wrapped in double quotes, and a quote inside a value is doubled.

That is the whole format. Most import failures come from a tool exporting something that only looks like CSV: a tab-separated file with a .csv name, a semicolon delimiter from a European locale, or a spreadsheet with a title row above the header.

Columns that import cleanly

Keep the header names simple and consistent: email, first_name, last_name, company, role, country, research_note. Lowercase with underscores avoids most mapping problems. If a tool expects a specific name, follow its documentation and map the fields once.

Only the email column is truly required. Everything else supports personalization and segmentation. A file with fewer, correct columns is better than a wide file full of empty cells.

Encoding and line endings

Save as UTF-8. Accented characters and non-Latin scripts break under other encodings, and the resulting garbled text appears in the email. Windows line endings are usually fine, but a file with mixed endings can confuse a parser.

If the file opens with strange symbols at the start, it may contain a byte-order mark. Re-save without a BOM, or import through a tool that handles it.

Quoting and escaping

A company called "Smith, Jones & Co" must be quoted or the comma splits it into two columns. A note with a line break must be quoted. A quote inside a value must be doubled.

Most spreadsheet exports handle this automatically. Problems appear when someone edits the file in a text editor, concatenates fields by hand or uses a script that joins with commas without quoting.

A checklist before upload

1. Open the file in a plain text editor and check the first rows.
2. Confirm one header row and no title rows.
3. Confirm the delimiter is a comma.
4. Check that every row has an email.
5. Search for duplicate addresses.
6. Check accented characters render correctly.
7. Confirm no field contains an unquoted comma.
8. Verify the addresses before sending.

Run the same checks on a sample of 20 rows if the file is too large to inspect fully. Errors usually repeat, so a small sample plus a verification report catches the pattern without reviewing every row. Keep the clean file in version control outside the campaign tool so a later import can be traced.

What not to put in a CSV

Never include passwords, API keys, payment details or full personal profiles. A lead file should contain the fields the campaign needs and nothing sensitive. If the source includes extra personal data, remove it before importing.

Also remove internal test addresses and role addresses that belong to your own company. Those rows waste sends and can distort reporting.

A worked example

A team exports 5,000 rows from a CRM. The file uses a semicolon delimiter and includes a title row with the export date. The import produces zero contacts. They re-export with commas, delete the title row, quote the company names that contain commas and verify the email column. The second import succeeds with 4,600 records after duplicates and invalid addresses are removed.

The data was the same. Only the file format changed, and that was enough to break the campaign. Test an import with a small file before uploading the full list; a formatting mistake is much cheaper to find on twenty rows than on five thousand.

Keep a canonical copy

Store one clean CSV per segment, with the date and source in the filename, and treat it as the input rather than the working list. When the campaign changes, export again from the canonical file with the current suppression list applied.

That habit makes imports repeatable and keeps the audit trail clear. For the wider process, see how to clean a cold email list and Google Sheets import.