Email list verification checks every address on a list against the mail server that would receive it and returns one of four results: valid, invalid, catch-all or unknown. Across the 7.93 million addresses BounceZero verified between April and October 2026, non-freemail (business and ISP) lists came back 14.5% invalid and 27% unknown, so a verifier that reports only valid and invalid is hiding a quarter of the answer.
Four results, four actions. The share of each on a typical list depends on where the list came from; the figures here are the corpus averages for business-domain addresses.
| Result | What was confirmed | Share on non-freemail (business and ISP) lists | What to do |
|---|---|---|---|
| Valid | The mail server accepted the exact mailbox during a live check | 32.6% confirmed | Send. Primary sequences, cold or warm |
| Invalid | The server rejected the mailbox, the domain has no mail records, or the syntax is broken | 14.5% | Remove. Each one is a bounce you avoided |
| Catch-all | The domain accepts mail for any address, so the mailbox cannot be confirmed | 2.3% (12.4% on .org domains) | Low-volume warm sequence from a secondary domain, or another channel |
| Unknown | The server did not answer definitively: greylisting, rate limits, corporate gateways that block probes | 27.0% (41% on .de, driven by t-online) | Hold. Re-check in seven days. Credits refunded |
Risky and role-based flags (info@, sales@) are returned alongside the four results. Full breakdown by provider and country: lead list quality benchmarks 2026.
CSV, TXT or XLSX, one address per row, other columns preserved. Up to 1,000,000 rows in a single job. Duplicates are removed before checking and not charged.
Malformed addresses, dead domains and domains with no mail records are marked invalid in seconds without a server probe. Disposable and role-based addresses are flagged here too.
Each remaining address is checked against its mail server, with provider-specific handling for Gmail, Microsoft 365, Yahoo, iCloud and the European ISPs that behave differently from the rest. Catch-all domains are detected by probing a random address alongside the real one.
Addresses that returned a temporary response are retried over the job window. What stays unresolved is reported as unknown and the credit is returned.
Download the full file with a result column, or valid-only, or everything except invalid. Results are also available through the API for CRM and sequencer integrations.
Credits are bought once and never expire. One credit checks one address; unknown results return the credit. Prices in USD.
| Plan | Credits | Price | Per check |
|---|---|---|---|
| Free | 100 a month | $0 | No card required |
| Starter | 2,000 | $6 | $0.0030 |
| Growth | 5,000 | $15 | $0.0030 |
| Professional | 10,000 | $30 | $0.0030 |
| Business | 50,000 | $140 | $0.0028 |
| Scale | 100,000 | $280 | $0.0028 |
| Pro Scale | 250,000 | $650 | $0.0026 |
| Enterprise | 500,000 | $1,200 | $0.0024 |
| Ultimate | 1,000,000 | $2,200 | $0.0022 |
Card payments through Paddle, crypto through Cryptomus. Compare with ZeroBounce pricing and NeverBounce pricing.
No verifier can confirm every address. The difference between vendors is whether they tell you. These are the four situations where a live check cannot reach a definitive answer and how BounceZero reports them.
The server says yes to everything, so a yes means nothing. Reported as catch-all, never as valid. Twelve percent of .org addresses and 24.7% of aol.com addresses in the corpus sit on catch-all configurations.
Mimecast, Proofpoint and some national ISPs refuse verification connections or answer ambiguously. t-online.de returned unknown for 41% of its addresses. These are reported as unknown and refunded, not guessed.
Some servers delay or throttle first contact. The job retries within its window; what remains unresolved is unknown. The share drops on a re-check a week later.
A mailbox nobody reads still accepts mail and verifies as valid. No verifier can detect this from the outside. Engagement data in your own sending tool is the only signal.
A verifier is paid per answer. The temptation is to turn ambiguous responses into a confident label and charge for them. Vendors that report 98% valid on a list of corporate addresses are usually doing exactly that, because the corpus shows 27% of non-freemail (business and ISP) addresses cannot be confirmed by anyone.
BounceZero reports those as unknown and returns the credit. The list you get back is smaller and more honest: everything marked valid was confirmed by the receiving server during the job. The practical effect is a bounce rate under 1% on the valid segment, which is what keeps a sending domain alive.
For lists where a quarter comes back unknown, the next step is a seven-day re-check for the unknown segment, which typically resolves a third of them, and routing the rest to channels that do not depend on mailbox confirmation.
Drag the file in, watch progress, download the result. Best for monthly list cleaning and one-off purchased lists.
One call per address for forms and sign-ups, or submit a file and poll for the result. Documentation at API email validation.
Verified exports drop straight into Instantly, Smartlead, Lemlist and the main CRMs with the result column intact, so sequences filter to valid only.
The product page, limits and formats
When to verify, how often, what to do with each segment
100 checks a month, no card
Why catch-all is its own result
The full hygiene routine
Expected results by provider and country
A 10,000-row list usually completes in 10 to 30 minutes. Large lists of several hundred thousand rows run for a few hours because the retry window for temporary server responses is part of the job. Progress is visible in the dashboard and the API reports status per job.
Every address marked valid was accepted by the receiving mail server during the live check, which keeps bounces on that segment under 1%. The honest limit is the unknown segment: on non-freemail (business and ISP) lists about 27% of addresses sit behind servers that will not confirm, and those are reported as unknown and refunded rather than guessed.
Yes. Every account gets 100 credits a month with no card, which covers a sample check of a purchased list or a small inbound list. Paid credits start at $6 for 2,000 and do not expire.
No. Verification talks to the mail server in the same protocol a sending server would use, but stops before any message is delivered. The recipient sees nothing and no mail is sent.
CSV, TXT and XLSX. The address column is detected automatically and every other column is kept, so the export returns your original data with a result column added. Up to 1,000,000 rows per job.
Upload the file, get valid, invalid, catch-all and unknown back, export with your columns intact. 100 free checks a month, unknowns refunded.
Ayoub built BounceZero's 5-stage validation pipeline, its dedicated BGP-announced IP infrastructure, and the Patroni HA PostgreSQL cluster behind every verification. Previously built high-volume email delivery infrastructure. Trained at 1337 Benguerir (École 42 network, 2019). Open-source: bgp_analyzer.
Where leads come from, how finders work, and what a verified lead is - with 2026 corpus data
Extensions, Sales Navigator exports, accuracy, the legal line
7.93M addresses: invalid, catch-all and unknown rates by provider and country
Email finder for companies
Free email lookup in 2026
Clay data enrichment explained
Apollo lead generation guide 2026
Continue through related topics