§01 · The engine

Every bad address is a letter returned  sort them with Purelist.

Purelist is the most accurate email verification API and bulk email cleaner behind lists that actually reach the inbox. We run real SMTP mailbox checks and a decision engine that returns a clear send/skip verdict, confidence score and bounce risk — without ever sending a test email.

Free, no signup · 3 checks / day

Live registry
Sorted today
Deliverable
412,902
Risky
38,118
Invalid
22,190
Engine online · 12 ms
/verification/logLast 4 events
  • 0421
    alex.chen@enterprise.com
    SMTP handshake · MX ok · catch-all: false
    VALID
  • 0420
    marketing@unknown-server.io
    MX records not found · domain unresolvable
    INVALID
  • 0419
    hello@disposablemail.com
    Disposable provider · low engagement risk
    RISKY
  • 0418
    info@acme.io
    Role-based address · catch-all: true
    ROLE
Featured in
Times Of India DailyBusiness Line PrintScroll India NewsIndia Today UpdatesIndia Global LiveThe India Watch
7-stage
verification pipeline
Real SMTP
mailbox checks
Send / Skip
clear verdict
Catch-all
detection
Self-improving
learns from outcomes
No test sends
zero spam-trap risk

What is Email Verification?

Email verification is the technical process of verifying that an email address exists, is active, and can receive messages without bouncing. It involves checking syntax, verifying domain MX records, and performing a deep SMTP probe to communicate with the mail server directly—without sending an actual email. Purelist’s API automates this entire pipeline in milliseconds.

How does a Bulk Email Cleaner work?

A bulk email cleaner processes thousands of email addresses simultaneously to filter out hard bounces, disposable emails, and risky catch-all addresses. By uploading your CSV to a bulk email verification service like Purelist, the engine tests each row using high-speed parallel probes. You receive a clean list with a definitive send or skip recommendation, protecting your sender reputation.

§03 · Decision layer

A verdict you can act on
not just valid / invalid.

Every result carries a send/skip action, a confidence score, reason codes and the method used to verify it — so a person or an automation knows exactly what to do.

Traditional services
{
  "catch_all": true,
  "risk_score": 68,
  "smtp_check": "unknown"
}

Cryptic codes. You still don't know whether to send.

Purelist · decision

Risky but usable.

Company domain on a catch-all setup. We can't confirm john.doe specifically exists, but the domain is real and active.

Recommendation · Safe for password resets and order confirmations. Use cautiously for cold email.

  1. /01 · Honest verdicts

    Catch-all, not faked

    On a catch-all domain we say so — flagged as risky/unconfirmed — instead of returning a fake “valid”. You always know what we actually confirmed.

  2. /02 · Reason codes

    Why, not just what

    Every result ships with machine-readable reason codes and the exact verification method used (SMTP mailbox, domain-only, catch-all, history) — no cryptic flags.

  3. /03 · Self-improving

    Learns from your sends

    Report your real delivered/bounced outcomes and the score sharpens — especially on the catch-all and unknown domains that break other tools.

§02 · Index of capabilities

Seven layers between
your list and a bounce.

Purelist doesn’t just check syntax — it runs a multi-stage pipeline that probes the real mail server, then turns the result into a clear decision. Every address is checked from several angles so risky ones never reach your CRM or campaigns.

Instead of raw technical flags, every result carries a send/skip action, a confidence score and reason codes — so a person or an automation knows exactly what to do.
  1. /01

    Bulk Verification

    Process up to 1 million emails at once

    Upload a CSV and our high-speed parallel engine cleans it in minutes. Get a clean deliverable list plus a full audit report.
    ≈ 50,000 rows in 3 minutes
  2. /02

    Real-time API

    Machine-readable decisions

    A drop-in REST API for signup forms, CRMs, and automation. Every response carries a send/skip action, confidence score and reason codes — ready for AI agents to act on. Block bad emails before they enter your database.
    REST · Webhooks · SDKs
  3. /03

    Syntax & MX Checks

    Standards-compliant first defense

    Validates against RFC 5321/5322 and checks live mail server availability — not just cached lists.
    Foundation of the 7-layer pipeline
  4. /04

    SMTP Deep Probe

    Real mailbox confirmation without sending

    Opens a safe conversation with the mail server to verify the mailbox exists — without delivering any email or risking your sender reputation.
    Zero-send protocol
  5. /05

    Disposable & Temporary Detection

    Maintained blocklist

    Automatically flags 10-minute mail, temp inboxes, and burner services from a maintained list refreshable on demand from the upstream community dataset.
    Keeps your metrics clean
  6. /06

    Role & Catch-all Intelligence

    Smarter handling, not blind rejection

    Identifies role accounts (info@, support@, sales@) and catch-all domains so you can route them intentionally instead of guessing.
    Classified, not dropped
  7. /07

    Decision engine

    Send / skip, scored

    Every signal above flows into a decision layer that returns a verdict you can act on — not just a flag.

    • A calibrated confidence score (0–100) and bounce risk per address
    • A clear recommended action: send, send low-priority, risky, or skip
    • Honest verdicts: mailbox confirmed, catch-all, or domain-only — never a fake “valid”
    • Machine-readable reason codes your CRM or an AI agent can act on
    Decision-ready, not just data
§04 · Self-improving

It gets smarter
with every send.

Purelist isn’t a static rulebook. When you report real delivery outcomes — or wire up your email provider’s bounce webhook — the engine updates its per-domain reputation and remembers individual addresses, so the hard cases get more accurate over time.

/01 · Outcomes

Real delivery results

You report delivered/bounced results (or pipe in your ESP’s bounce webhook) and the engine records the outcome per domain and per address.

/02 · Learning loop

Continuous improvement

Each outcome refines future verdicts. A domain’s real bounce rate is blended into its score, so shifting catch-all behaviours are caught faster.

/03 · Resolution

Beats catch-all

An address known to bounce is flagged invalid on the next check — even when the server accepts everything. Per-address memory is the only way to resolve a catch-all domain.

Others vs Purelist
Static rules & public lists
Learns from your real delivery outcomes
Fixed risk scores
Scores that sharpen as you report outcomes
Technical jargon
Reason codes + the method behind every verdict
Same accuracy for years
Improves as email behaviours change
Generic flags
A clear send / skip action per address
Result you can trust

Fewer surprises, better deliverability, decisions you can actually trust.

Because the engine learns from the real outcomes you report, it handles the edge cases that break static systems — catch-all domains, brand-new domains, and temporary providers that haven’t hit public blacklists yet.

Real outcomes
Feedback signal
Per-address
Outcome memory
§04 · The sorting line

From upload to clean export
in under three minutes.

Three simple steps that replace hours of manual list cleaning. Upload, inspect, and export a delivery-ready audience — with full transparency at every stage.

Bulk job · acme_leads.csv
Processing
48,210 / 50,000 verified96%
Deliverable
41,902
Risky
4,118
Invalid
2,190
sara.lopez@stripe.comVALID
ops@northwind.coRISKY
team@quanta.aiVALID
Upload
Stage 01
Inspect
Stage 02
Export
Stage 03
Step 1

Connect your source

Sync your CRM or drop a CSV. We prepare the data for inspection.

Step 2

Multi-layer inspection

Seven stages — syntax, MX, SMTP, disposable, role, catch-all, deliverability score.

Step 3

Export & automate

Download a clean list or trigger workflows for only the deliverable contacts.

Read the engineering brief
§05 · Tariff

Priced by what you actually verify.

Credit-based, no seat tax, no overage trap. Credits never expire and roll between bulk and API usage interchangeably.

§06 · Field reports

The teams who stopped fighting their inbox.

Portrait of Marta R., Head of Lifecycle, NorthwindNorthwind
8.4% → 0.6%bounce rate
We cut our bounce rate from 8.4% to 0.6% in a week. The reputation curve we'd been fighting for six months turned around the same afternoon.
Marta R.
Head of Lifecycle, Northwind
Portrait of Dev O., Staff Engineer, QuantaQuanta
Send / Skipper-address decision
Other tools just gave us valid/invalid flags. Purelist returns a send/skip action with a confidence score and reason codes — our automations finally know what to do with catch-all and risky addresses.
Dev O.
Staff Engineer, Quanta
§07 · Last entry

Stop sending letters
to addresses that don't exist.

Start with 25 free verifications. No credit card. Bring your CSV or your API key — we'll sort the rest.

No credit card1,000 free creditsNo test sends