CSV import rules: what gets rejected and why
Last verified 2026-07-01
Bad data quietly ruins outreach: bounced sends, broken personalization, embarrassing messages. Every imported row passes a validation gate before it can persist — rows that fail are reported with the reason, never silently dropped.
What a row needs
- A way to reach the person: a valid email address or a valid LinkedIn URL.
- Email must be a real address shape; web addresses must start with http or https.
- Phone numbers must include the country code, e.g. +1 415 555 0123.
What gets rejected
- Placeholder junk: literal "undefined", "null", "n/a", or whitespace-only fields.
- Machine-generated filler text that was never written by a human.
- Styled lookalike characters (𝐀𝐜𝐦𝐞-type unicode) that sneak past naive checks.
Duplicates are folded automatically: a contact is fingerprinted by email or LinkedIn profile, so the same person imported twice stays one record.