Email Verification Says Unknown: Should You Send, Recheck, or Suppress?

You upload 5,000 leads to a verifier. 3,800 come back valid, 400 come back invalid, and 800 come back as something fuzzier: "unknown," "risky," "accept-all," "catch-all," or "do not mail." The invalid ones are easy — delete them. The valid ones are easy — send them. The 800 in the middle are 16% of the money you paid for that list, and nobody tells you what to do with them.

Sending all of them can push your bounce rate past the point where Google and Microsoft start filtering you. Suppressing all of them throws away leads that would have replied. The right answer depends on *why* the verifier couldn't decide, and most tools give you enough information to work that out if you know what to look for.

If you're still deciding whether verification is worth the cost at all, start with why you should verify every lead before you send. This post is about the bucket that verification can't resolve.

{"h2":"placeholder"}

Why verifiers return unknown

A verification check is an SMTP conversation. The verifier connects to the recipient's mail server, starts a handshake, and asks whether the address would be accepted. "Unknown" means the server didn't answer that question honestly. There are five common reasons, and they carry very different risk.

{"ul":["Catch-all / accept-all domain. The server accepts mail for every address at that domain, real or not. It isn't lying — it genuinely can't tell you whether john.smith@ exists, because it would also accept asdfgh@. Very common on Microsoft 365 tenants and on small-business domains with aliases.","Security gateway in front of the MX. Mimecast, Proofpoint, Barracuda and similar products terminate the SMTP connection themselves and accept everything, then filter behind the scenes. Looks identical to a catch-all from the outside.","Greylisting or rate limiting. The server returned a 4xx temporary failure because it throttles unknown senders or deliberately defers the first contact. Reruns often resolve these.","Timeout or connection refused. The verifier's probe IP is blocked, the MX was down, or DNS was slow. Pure infrastructure noise, nothing about the address.","Role, disposable, or flagged address. info@, support@, sales@, or a known temporary-mail domain. The verifier knows it's deliverable but marks it risky for a different reason."]},

Most tools lump all five into one label. The better ones (ZeroBounce, NeverBounce, MillionVerifier, Bouncer) expose a sub-status — accept_all, greylisted, timeout, role_based, mailbox_full. Pull that field. It's the difference between a guess and a decision.

{"h2":"What each unknown sub-type actually costs you"}

{"table":{"headers":["Sub-type","Typical hard-bounce rate if you send","Hidden cost","Default action"],"rows":[["Catch-all (real person, confirmed elsewhere)","2-6%","Silent black-hole delivery, no reply signal","Send, separated"],["Catch-all (pattern-guessed address)","Low bounce, high waste","Delivers to nobody, zero engagement","Suppress or confirm first"],["Gateway-protected domain","3-8%","Heavy filtering, scanner clicks inflate opens","Send, low volume"],["Greylisted / 4xx temporary","5-15%","None if you recheck","Recheck in 24-72h"],["Timeout / connection error","10-20%","None if you recheck","Recheck with a second vendor"],["Role address (info@, sales@)","1-3%","Spam complaints, shared inbox, no decision maker","Suppress"],["Disposable / temporary domain","15-40%","Trap risk","Suppress permanently"]]}}

Note what's not in that table: the assumption that unknown means "probably dead." Catch-all addresses often deliver fine. The risk with catch-alls isn't bounces — it's that you have no way to know whether the mailbox is monitored, which means a chunk of your send goes nowhere and drags your reply rate down without ever showing up as an error.

{"h2":"The bounce math that decides how many you can send"}

Your bounce ceiling is the only hard constraint. Keep hard bounces under 2% of total volume per mailbox; Google and Microsoft start applying real pressure above 3-4%, and most cold email platforms will pause a campaign before that. (More on thresholds in what bounce rate is too high for cold email.)

Work backwards from it. Say you have 5,000 leads: 4,000 verified valid, 1,000 unknown.

{"ul":["Verified-valid addresses bounce at roughly 0.5% — that's 20 bounces.","A 2% ceiling across 5,000 sends allows 100 bounces total.","So the unknown bucket can contribute at most 80 bounces.","If your unknowns bounce at 8%, you can send all 1,000.","If they bounce at 15%, you can only send about 530 of them before you blow the budget."]},

You don't know the real rate until you test. So test: pull a random 100-200 unknowns, send them a single first-touch email from one mailbox that isn't carrying your main campaign, and measure the hard-bounce rate after 48 hours. Two hundred addresses is enough to tell 5% apart from 20%. That sample costs you almost nothing and tells you exactly how much of the bucket is usable.

If the sample bounces above 15%, don't try to salvage the segment with volume tricks. Recheck it or drop it.

{"h2":"When to recheck instead of guessing"}

Rechecking is cheap and it resolves more than people expect. Three rules:

{"ul":["Rerun greylisted and timeout results after 24-72 hours. Temporary failures clear. A second pass on the same vendor typically converts 20-40% of these into a definite valid or invalid.","Cross-check stubborn unknowns with a second vendor. Different verifiers probe from different IP ranges with different retry logic. An address that times out for one provider frequently resolves for another. This is worth doing once, not three times — past the second vendor you're paying for noise.","Re-verify any list older than 30 days. B2B contact data decays at roughly 2-2.5% per month from job changes alone. A list verified in March and sent in July is not a verified list."]},

What doesn't work: rechecking a catch-all domain. The server will give the same non-answer forever. No verifier can resolve a true accept-all through SMTP, and any vendor claiming 100% catch-all resolution is pattern-matching, not verifying.

{"h2":"How to send the unknowns safely"}

Once you've rechecked and the remainder is mostly catch-all and gateway domains, treat it as a separate campaign, not as a blended slice of your main one.

{"ul":["Separate campaign, separate mailboxes. Keep unknowns off the mailboxes carrying your clean list. If the bounce rate spikes, the damage is contained to one or two sending addresses instead of your whole pool.","Start at half volume. 15-20 sends a day per mailbox for the first week, so a bad batch shows up as a handful of bounces instead of a hundred.","Confirm the human before the address. For catch-alls on accounts worth real money, check LinkedIn or the company site that the person still works there. If the person is real and the pattern matches other confirmed addresses at that domain, the catch-all is probably fine.","Don't judge catch-alls on opens. Gateway-protected domains generate scanner opens and scanner clicks. Measure replies and verified visits only, or you'll conclude the segment is performing when it's delivering to a filter.","Suppress on first hard bounce, permanently. Not after three. One 550 is enough.","Cap the segment. If unknowns are more than 25-30% of your total daily volume, you're letting a low-confidence list drive your sender reputation."]},

{"h2":"What to suppress without thinking twice"}

Three categories never justify the risk: disposable domains, addresses flagged as spam traps or on known complainer lists, and role accounts at companies where you already have a named contact. Role addresses in particular are a complaint magnet — info@ is read by whoever is least invested in hearing from you, and shared mailboxes report spam at several times the rate of individual ones.

Also suppress anything a verifier flags as "mailbox full" more than once across rechecks. A persistently full mailbox is usually abandoned, and abandoned mailboxes are how recycled spam traps get created.

{"h2":"How Sendvanta handles the risky bucket"}

Sendvanta doesn't sell lead lists or verification credits — you bring your own leads and your own mailboxes. What it does is make the unknown segment survivable: campaigns can be isolated to specific mailboxes so a risky batch never touches your clean senders, bounces move addresses into suppression automatically and permanently, and layered deliverability health scoring watches bounce rate alongside mailbox, domain, DKIM, and IP signals, slowing or pausing a campaign when thresholds are crossed instead of after the damage is done. Engagement classification filters security-scanner opens and clicks, which matters most on exactly the gateway-protected domains that produce unknown results. You can test all of it on the free plan — $0 forever, 1,000 active leads, 3,000 emails a month, up to 10 of your own SMTP mailboxes. See features for the full list.

{"h2":"The short version"}

Pull the sub-status from your verifier. Recheck greylisted and timeout results after a day or two, and cross-check the stubborn ones with a second vendor. Suppress disposables, role accounts, and repeat mailbox-full results. Sample-test the catch-alls with 200 sends before committing. Then send what's left on separate mailboxes at half volume, and keep the whole bucket under a quarter of your daily output. That turns 800 unusable leads into maybe 500 usable ones without putting your domain at risk — which is the actual goal.

{"faq":[{"q":"Is catch-all the same as unknown?","a":"No. Catch-all means the domain accepts mail for every address, so verification can't confirm the specific mailbox — the address often delivers fine. Unknown is a broader label that also covers timeouts, greylisting, and probe failures, which are temporary and usually resolve on a recheck."},{"q":"What percentage of a B2B list is normally unknown?","a":"Typically 10-25%, depending on how the list was built. Lists heavy on mid-market and enterprise domains skew higher because Microsoft 365 tenants and security gateways frequently present as accept-all."},{"q":"Do catch-all addresses hurt deliverability?","a":"Rarely through bounces — they usually accept mail. The real cost is silent: messages delivered to unmonitored aliases produce no replies, which lowers your engagement signals and makes campaign results look worse than your copy deserves."},{"q":"Should I just buy a more expensive verifier?","a":"It helps at the margins. Better vendors expose sub-statuses and resolve more greylisted results, but no verifier can resolve a true catch-all over SMTP. Spend on sub-status detail and a second-vendor cross-check rather than on promises of 100% accuracy."},{"q":"How long is a verification result valid?","a":"About 30 days for cold outreach. B2B contact data decays roughly 2-2.5% a month from job changes, so re-verify anything you're sending more than a month after the original check."}]}