Apollo Lead Verifier

Apollo says 91% email accuracy. Real users report 65-80%. This Actor closes that gap.

Upload any Apollo/LinkedIn/ZoomInfo CSV export. Get back a clean CRM-ready CSV with per-lead email deliverability scoring, role-account detection, and bounce risk flags.

Why this exists

Across 9,000+ G2 reviews of Apollo, "Inaccurate Data" is the #1 complaint (638 mentions). Real-world bounce rates sit at 15-35% per Apollo-sourced lists. Teams running cold outbound off Apollo end up:

• Burning sender reputation (bounces tank deliverability)
• Wasting rep time on dead emails
• Hitting credit limits on data they can't use

What it does

1. Parses any Apollo/LinkedIn/ZoomInfo CSV (auto-detects email column)
2. Normalizes email addresses (strips whitespace, fixes common typos)
3. Verifies each email:
   • RFC format check
   • DNS MX record lookup
   • SMTP RCPT TO probe
   • Optional catch-all detection (strict mode)
4. Detects role-accounts (info@, contact@, noreply@...) that hit shared inboxes
5. Detects disposable email providers (Mailinator, Yopmail, ...)
6. Flags free providers (Gmail, Yahoo, ...) separately from business email
7. Scores each lead 0-100 for deliverability
8. Sorts output high-score first — what you want in your CRM is at the top
9. Writes a clean CSV back to the Key-Value Store, ready for direct download

Input

Field	Type	Required	Default	Description
csvUrl	string	one of	-	Public URL to a CSV (Google Sheets "Publish to web" CSV, Dropbox, S3)
csvContent	string	one of	-	Raw CSV text (max 5MB)
emailColumn	string	no	auto-detect	Header name of the email column
firstNameColumn	string	no	auto-detect	Header name of first name
lastNameColumn	string	no	auto-detect	Header name of last name
verificationLevel	enum	no	standard	fast (format+MX), standard (+SMTP), strict (+catch-all detection)
maxLeads	integer	no	0 (unlimited)	Cap on rows processed

Output

For each row, the Actor adds:

Column	Description
verified_email	The normalized email
verification_status	valid / risky / invalid / role_account / disposable
deliverability_score	0-100 (higher = safer to send)
verification_reason	Human-readable reason for the status
format_ok	Boolean: RFC format check passed
mx_found	Boolean: domain has MX records
smtp_check	ok / ok_likely / fail / catch_all / catch_all_likely / skipped
is_role_account	Boolean: shared inbox (info@, sales@, ...)
is_free_provider	Boolean: Gmail/Yahoo/etc.
is_disposable	Boolean: Mailinator/Yopmail/etc.
domain	The email's domain

A summary record is also pushed to the dataset with counts per status.

Pricing

Pay-per-event:

• actor-start: $0.05 (once per run)
• lead-verified: $0.04 per lead verified

Example: verify 1,000 leads = $0.05 + (1,000 × $0.04) = $40.05

How to deploy

1. Create a new Actor in Apify Console
2. Choose "Import from GitHub" (after pushing this repo)
3. Or push via CLI: apify push (requires APIFY_TOKEN)
4. Set base image to Python 3.11
5. Build command: pip install -r requirements.txt
6. Start command: python src/main.py

Test locally

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python3 test_e2e.py

The test runs against 10 simulated Apollo leads and prints a verification report.

Limitations

• SMTP probes require outbound port 25. Some cloud providers (AWS, GCP) restrict this by default. Apify's Docker host allows it; if you self-host, you may need to allow port 25.
• Catch-all detection (strict mode) is heuristic and can produce false positives for permissive servers.
• This Actor does NOT enrich data — it only verifies what you upload.

License

MIT