Email List Quality Data: What 597,834 Verified Addresses Show

In the 373 bulk lists that BounceCheck users uploaded, 33.8% of the 597,834 addresses came back invalid, 7.0% came back unknown, and 59.2% were valid. The pooled figure hides a wide spread: the median list was 34.5% invalid, and half of all lists fell between 18.0% and 60.6%.

This page is the source for the numbers we quote elsewhere on the blog about catch-all addresses, role-based addresses and greylisting. Every figure comes from BounceCheck's own verification results, pulled on 2026-11-04, and none of it is estimated from other companies' reports.

What share of a typical list is bad?

A typical list was about a third invalid, but lists differ a lot. Counting each list on its own, instead of pooling every address, gives this spread:

Percentile of lists	Invalid share
10th (cleanest lists)	8.5%
25th	18.0%
Median	34.5%
75th	60.6%
90th (worst lists)	80.0%

One in ten lists was at least 80% invalid. Size does not predict quality in a simple way:

List size	Lists	Addresses	Invalid share
50 to 999	282	55,933	37.2%
1,000 to 9,999	72	186,480	42.6%
10,000 or more	19	355,421	28.7%

The 19 largest lists hold most of the addresses and are the cleanest group, which pulls the pooled figure down.

What this data is, and what it is not

Read these numbers as a picture of lists people chose to verify, not of email in general.

• Who is in the sample. Completed bulk jobs of 50 or more addresses, from 62 accounts. Single checks and unfinished jobs are left out.
• The sample leans toward trouble. People tend to verify a list when they already suspect a problem, so the invalid share is probably higher than on a random list.
• A few accounts weigh a lot. The largest account holds 21.3% of all addresses and the three largest hold 50.3%. Without the largest account, the invalid share is 37.1% instead of 33.8%. This is why we lead with the per-list median.
• Invalid is our verdict, not a bounce. We did not follow these addresses into real sends, so we cannot say how many would have bounced. An invalid share is not a bounce rate, because invalid addresses should be removed before you send.
• Privacy. We publish only aggregated, anonymized statistics, as our privacy policy describes. Any group under 1,000 addresses or from fewer than 5 accounts is folded into "other". No address, domain, account or file is shown.

Results by mailbox provider

Invalid rates differ sharply by where the mailbox lives. We read the provider from the MX host when we stored one, and otherwise from the domain of a free mailbox such as gmail.com or outlook.com. Business domains with no stored MX host stay in the last row, so this table says little about companies that use Microsoft 365 or Google Workspace on their own domain.

Provider	Addresses	Invalid	Unknown	Catch-all*
Google (Gmail)	179,374	14.7%	0.1%	0%
Microsoft (Outlook, Hotmail)	26,473	48.7%	0.3%	0.2%
Yahoo and AOL	13,764	13.1%	46.2%	0.2%
iCloud	5,784	58.6%	0.4%	0%
Custom domains, provider not identified	372,439	42.4%	9.5%	5.5%

*The catch-all column covers results from 2026-09-01 onward, the period for which reason codes are recorded. The other columns cover all results.

Two things stand out. Yahoo and AOL are where verification most often ends without an answer: 46.2% of those addresses came back unknown, against under 0.5% for Google, Microsoft and iCloud. And custom business domains are the hardest to keep clean, with 42.4% invalid.

Why some addresses come back unknown

An unknown result means the verifier could not reach a verdict, and it was 7.0% of all addresses (41,948). Since 2026-09-01 the reasons behind those results, with the share of all 316,579 addresses that carry a reason code, were mainly:

• Probe blocked (4.4%): the mail server dropped the connection or refused the check.
• Catch-all domain (4.4%): the domain accepts every address, so a real answer is not possible.
• Provider does not allow verification (4.2%): the provider does not permit mailbox checks.

For what each status means in practice, see email verification status meanings.

How common are catch-all domains?

Catch-all domains were a small minority. Of the 84,457 distinct domains checked since 2026-09-01, 7,269 (8.6%) came back as catch-all and could not be verified. Counted by address instead of by domain, catch-all results were 13,834 of 316,579 (4.4%).

They are concentrated in custom domains: 5.5% of addresses on those domains were catch-all, against 0.2% or less on the large free providers. A domain that is catch-all but returned a valid verdict is not counted here, so the true share may be somewhat higher. For what to do with these addresses, read our guide to catch-all emails.

Are role-based addresses worse than personal ones?

In this data they were not. Role-based addresses such as info@ and support@ were 0.63% of addresses (3,738). 23.5% of them were invalid, against 33.9% for all other addresses.

We have not tested why. One possible reason is that role addresses are usually taken from a company's public contact page, where the mailbox is kept alive on purpose. Invalid is the wrong test for them in any case: the problem with role-based addresses is consent and complaints, not whether the mailbox exists. See role-based email addresses for how platforms treat them. Disposable addresses were 0.88% of the sample (5,252).

How much of this is greylisting?

Very little. Greylisting is one of five reason codes that each stayed under the 1,000-address floor and are grouped as "other", together 1,596 addresses or 0.5% of results since 2026-09-01. The other four are invalid syntax, mailbox full, service unavailable and unrecognized answers. So greylisting is at most 0.5% of the sample. More on this in what greylisting is.

Every result reason, since 2026-09-01

Reason	Addresses	Share
Mailbox exists	186,453	58.9%
Mailbox not found	41,237	13.0%
Undeliverable (negative verdict, no cause given)	28,677	9.1%
Probe blocked	14,082	4.4%
Catch-all domain	13,834	4.4%
Provider does not allow verification	13,271	4.2%
No MX records	8,495	2.7%
Mailbox temporarily unavailable	5,721	1.8%
Disposable address	3,213	1.0%
Other (five smaller reasons)	1,596	0.5%

What to do with these numbers

• Do not use the pooled figure as your benchmark. Your list could be at the 10th or the 90th percentile. The only way to know is to check it.
• Treat unknown as its own group. It is not the same as invalid, and the right handling differs by provider. Yahoo and AOL addresses are the ones to expect it from.
• Check before you send, not after. Run a small sample through the email checker, then follow the steps in how to clean up your email list.

Frequently asked questions

What percentage of an email list is typically invalid?

In our data the median list was 34.5% invalid, and half of the lists were between 18.0% and 60.6%. This sample leans toward lists people already worried about, so a list kept clean from the start will usually be lower.

Is a 34% invalid rate the same as a 34% bounce rate?

No. Invalid is the result of a check before sending. If those addresses are removed, they never bounce. A bounce rate is measured from real sends. See why verified emails still bounce for what bounces remain after verification.

How many email domains are catch-all?

8.6% of the domains we checked between 2026-09-01 and 2026-11-04 came back as catch-all and unverifiable. The share is higher on custom business domains than on large free providers.

About this data

Data pulled on 2026-11-04 by the BounceCheck team from BounceCheck's own verification database. It covers all completed bulk jobs of 50 or more addresses up to that date, except the reason-code figures, which start on 2026-09-01 when reason codes were standardized. We will recompute the figures each quarter and note the date at the top of the page when we do.