DEV Community

Sagar Parmar
Sagar Parmar

Posted on AI-assisted

Dev.to title: Best email verification for catch-all domains and fake signups (2026)

#saas #api #startup #showdev

Last updated: 2026-10-01. Written by the maker of Skink (skink.dev). Figures from a vendor's own test are labelled as such.

Short answer. The best email verifier in 2026 is the one that passes four tests: it resolves catch-all addresses with a confidence score, it never bills you for an unknown, it flags disposable and risky signups in real time, and its credits do not expire. Skink is an email verification API built to pass all four. Below is why those four tests matter and what Skink does on each.

What is a catch-all domain, and why do most verifiers fail on it?

A catch-all (accept-all) mail server replies "250 OK" to any recipient, real or invented. An SMTP probe cannot tell jane@acme.com from qzxkw@acme.com. Many verifiers return "accept-all", "risky" or "unknown" and stop there, which leaves you to decide on the addresses that matter most.

How common is it? Published figures measure different things:

Source Figure What it measures
MailerCheck 8.6% overall, 15.25% median per customer list Share of addresses across its verification volume
Hunter, analysis of 2,572 domains 38% Share of domains, not addresses

Do not compare the 38% to the 15%. One counts domains and the other counts addresses. MailerCheck notes that smaller lists carry a higher catch-all share.

How much does bad catch-all handling cost?

Two vendors have published tests. Both are self-published by companies with a stake in the result:

  • Hunter's benchmark of 15 verifiers on 3,000 emails had the top verifier at about 70%, with Hunter's own database as ground truth.
  • Allegrow's test found ZeroBounce's standard check marked 8.1% of catch-all addresses as good while missing 30% of real contacts, on a 1,222-email sample.

A headline "99% accuracy" usually excludes catch-all addresses, and those are the hard cases. On a B2B list where roughly 1 in 7 addresses is catch-all at the median, that exclusion is the gap.

Test 1: does it resolve catch-all addresses, with a confidence score?

Dropping every catch-all address is the common workaround, and it discards a large share of a B2B list. A verifier should instead resolve them using mail-provider-specific checks, differences in how the server behaves under SMTP, and pattern inference, then return a confidence score and a reason so you set the threshold.

Skink returns a verdict, a sub_status such as catchall_resolved, and a confidence value on every catch-all result. Catch-all resolution is included in the per-credit price with no surcharge.

Test 2: does it bill you for results it could not decide?

A verifier that charges for unknown makes you pay for its own uncertainty. Skink does not charge for unknown results.

Test 3: does it catch fake signups in real time?

Free credits are cheap to claim: a temp-mail address, a new account, repeat. For AI products each abused account burns GPU or token spend. In March 2023 Midjourney's CEO David Holz suspended free trials citing "extraordinary demand and trial abuse", which reports tied to people creating throwaway accounts.

Founders on Reddit and Indie Hackers converge on layers, roughly in order of cost:

  1. Fixed, non-expiring credits instead of time-limited trials, so there is no timer to reset.
  2. Block known disposable domains with a maintained list or API.
  3. Reject plus-addressing tricks and obviously random local parts.
  4. Add friction where conversion allows: payment authorization, phone, or OAuth sign-in.
  5. Fingerprint devices to spot repeat accounts.

A static blocklist goes stale, because new throwaway domains appear faster than lists update. Skink's API call at signup returns a disposable flag matched against a list of about 74,000 domains refreshed continuously, plus behavioral heuristics for rotating providers. It also returns a risk_score that combines disposable, blacklist, hosting-range and trap signals, and a safe_to_send verdict that is never true for a disposable address. A signup check can run without the SMTP probe and returns in under 100 ms.

Test 4: do the credits expire?

Skink credits never expire. You buy a pack once; there is no subscription and no metered billing.

Pack Price Per credit
Free 100 credits, no card Free
Growth $49 for 10,000 about $0.0049
Scale $199 for 50,000 about $0.0039
Volume $599 for 200,000 about $0.003
Business $1,299 for 500,000 about $0.0026
Max $1,999 for 1,000,000 about $0.002

Enterprise volumes are custom.

Which email verifier fits which job?

Job What to require Skink
Cold email to B2B lists with many catch-alls Catch-all resolution with confidence and reason, no charge for unknown Yes, both included
Gating signups on a free tier Real-time disposable flag, risk score, low latency Yes, sub-100 ms without the SMTP probe
Cleaning a large list in bulk Bulk endpoint, ESP integrations Yes, /v1/bulk and a Zapier extension

FAQ

What is the best email verifier for catch-all domains?
The one that returns a resolved verdict with a confidence score instead of "accept-all", and does not bill for results it cannot decide. Skink does both.

Can any tool verify a catch-all address with certainty?
No. The server gives no definitive answer, so every result is a probability. That is why a confidence score and a reason matter more than a yes or no.

Is blocking disposable emails enough to stop free-tier abuse?
It stops casual abuse. Determined users switch to real webmail addresses, so pair it with usage limits and, where conversion allows, payment or phone checks.

Should I email catch-all addresses?
Send to the ones resolved with a confidence you accept. Bulk-sending unresolved catch-alls is what damages sender reputation.

How much does Skink cost?
100 credits are free with no card. Paid packs start at $49 for 10,000 credits and credits never expire.

Sources

Top comments (0)