SieveSieve vs MillionVerifier
We built this comparison, so assume bias. That is exactly why the rows where MillionVerifier wins are still in it. Where they are genuinely better, we say so.
MillionVerifier competes on price, and it wins that fight: its published rates run from about $0.59 to $3.90 per 1,000, the lowest floor of any vendor we compare against. It also does not charge you for addresses it returns as unknown, which is a genuinely fair touch. What it does not give you is evidence: results come back as a status without the MX host or SMTP reply code behind them, catch-all domains are returned as unknown rather than scored, and there is no pre-send risk verdict tying in your own domain's authentication health.
Price per 1,000 verifications at ~10k volume
Published list prices, June 2026. Check vendor sites for current rates. Sieve pack math: $5/1k = $5.00, $19/5k = $3.80, $69/25k = $2.76, $169/100k = $1.69, $29 Pro / 15,000 = $1.93 per 1,000.
Sieve vs MillionVerifier, capability by capability
| Capability | Sieve | MillionVerifier |
|---|---|---|
Live SMTP mailbox verification Table stakes: both do it. | ||
Verdict receipt on every address The MX host that answered and the SMTP reply code, in the API response. | ||
Catch-all confidence score (0 to 100) Built from real probe evidence. | Returns unknown, but doesn't charge for unknowns | |
Detects full and disabled mailboxes From the live SMTP reply code, shown on the receipt. | ||
Pre-send risk verdict Fuses the recipient verdict with YOUR sending domain's SPF, DKIM and DMARC health into one send, caution or suppress call. | ||
Single-tenant, self-hosted privacy Your lists never leave our EU-jurisdiction box, are never resold, and never train a pooled cross-customer dataset. | Multi-tenant cloud | |
Lowest price per 1,000 The cheapest floor of the vendors on our comparison. | No, MillionVerifier wins on price | |
Free tier | Unlimited DNS checks + 100 live SMTP/mo | Trial credits |
Competitor capabilities summarized from their public docs, June 2026. One caveat that applies to both of us: free-mail providers (Gmail, Yahoo, Outlook, iCloud) block mailbox probing. No vendor can truly SMTP-verify those addresses, whatever the marketing says.
The honest tradeoff
MillionVerifier is the cheapest per 1,000 and does not charge for unknowns. Sieve costs a little more but returns a verdict receipt, a catch-all confidence score and a pre-send risk verdict that MillionVerifier does not.
Judge us before you pay anything
Grade your sending domain's deliverability free: MX, SPF, DKIM, DMARC, blacklists. No signup.
Then verify a list with receipts
Unlimited DNS checks plus 100 live SMTP verifications a month, free. Every paid verdict ships with its evidence.
Frequently asked questions
Is Sieve a good MillionVerifier alternative?
Yes, if you want the evidence behind each result and extra signal on risky addresses. Sieve returns the MX host and SMTP reply code on every verdict, scores catch-all domains 0 to 100 instead of returning a bare unknown, and adds a pre-send risk verdict. If the lowest possible price per 1,000 is your only priority, MillionVerifier is genuinely cheaper.
Is Sieve cheaper than MillionVerifier?
No. MillionVerifier has the lowest price floor of the vendors we compare, with published rates from roughly $0.59 to $3.90 per 1,000. Sieve costs a little more per 1,000, and the trade you are making is price for evidence: a verdict receipt, catch-all confidence scores and pre-send risk on every address.
What does Sieve do that MillionVerifier does not?
Sieve returns a verdict receipt on every address (the MX host that answered and the raw SMTP reply code), gives catch-all domains a 0 to 100 confidence score instead of a plain unknown, detects full and disabled mailboxes from the live reply code, and produces a pre-send risk verdict from your own domain's SPF, DKIM and DMARC health.
Does MillionVerifier or Sieve charge for unknown results?
MillionVerifier does not charge for addresses it returns as unknown, which is a fair policy and one of its genuine advantages. Sieve takes a different approach to the same problem: instead of returning a bare unknown for catch-all domains, it scores them 0 to 100 from real probe evidence, so you get usable signal on the addresses other tools give up on.