We’re trying to import 5,000 contacts using Data Import Wizard but getting two major issues. First, about 300 records fail with ‘INVALID_EMAIL_ADDRESS’ even though the emails look valid (format: name@company.co.uk, name+tag@domain.com). We need proper RFC 5322 email validation but the wizard seems overly strict.
Second issue is duplicate detection - we have duplicate rules configured but they’re not catching obvious duplicates during import. Same email addresses are creating multiple contact records. Our CSV has been cleaned but we’re still seeing:
Error on row 234: INVALID_EMAIL_ADDRESS
Field: Email, Value: john.doe+sales@techcorp.io
Error on row 456: Duplicate detected but record created anyway
The CSV preprocessing should handle this but something’s wrong with our field mapping or duplicate rule configuration. This is blocking our Q2 lead onboarding and we need to import these contacts by end of week. Any guidance on proper email validation setup and why duplicate rules aren’t firing during import?
For large imports with complex validation requirements, I always recommend using Apex Data Loader with a custom CSV preprocessor. You can validate emails against a proper RFC 5322 regex before import and handle duplicates programmatically. The wizard is great for simple imports but lacks the flexibility you need here.
The plus sign in email addresses is a known issue with Data Import Wizard’s validation. Salesforce uses a stricter subset of RFC 5322 that doesn’t always accept special characters like + or uncommon TLDs. For the duplicate rules, they need to be activated AND set to block on insert. Check Admin Setup → Duplicate Rules → your contact rule and verify the action is set to ‘Block’ not just ‘Alert’.
I’ve seen this exact scenario. Your duplicate rules might be using fuzzy matching which doesn’t work well during bulk imports. Switch to exact matching on Email field for imports. Also, Data Import Wizard has a 50,000 record limit but performance degrades above 2,000 records. For 5,000 contacts, consider using Data Loader instead - it handles email validation better and gives you more control over duplicate detection. You can also pre-validate emails using a regex pattern in Excel before importing to catch format issues early.
Check your duplicate rule matching criteria carefully. If you’re matching on Email field, make sure it’s set as the primary matching field with 100% match threshold. Also verify the rule is active for the profile you’re using to import. I had a similar issue where the duplicate rule was only active for standard users but not for my admin profile doing the import. The CSV preprocessing is important too - make sure you’re using UTF-8 encoding and removing any hidden characters or trailing spaces from the email column.
The RFC 5322 compliance issue is tricky. Salesforce accepts most standard formats but rejects some valid ones. For your specific errors, emails with plus signs should work in newer releases. Are you on Summer '24? There was a fix in Spring '24 for extended character support. If you’re on an older version, you might need to strip the plus signs during preprocessing. For the .co.uk domains, those should definitely work - might be a whitespace issue in your CSV. Try trimming all email values before import.
Have you checked if your duplicate rules are configured for ‘on create’ actions? Sometimes they’re only set for ‘on edit’ which wouldn’t catch import duplicates.