Email list cleaning, and what it can actually remove
How the list got this way
- Every list ages. People leave jobs, close accounts and abandon addresses, and none of that generates a notification you receive.
- Sign-up forms collect typos at a steady rate, and a confirmation email to a mistyped address never arrives to correct it.
- Imported files arrive from events, partners and old systems with no record of how consent was obtained or when.
- Suppression lists get lost in migrations, so addresses you already removed come back in the next import.
Cleaning is a protocol conversation, not a filter
A cleaner looks up where the domain receives mail, opens a connection, and asks the receiving server whether it will accept a given recipient. The server answers with a code. A 550 means no such user, and that address is removed. A 250 means accepted.
The connection then closes, before the stage where a message body would be transmitted. Nothing is delivered and nobody receives anything, which is the question most people ask first.
The complication is that acceptance is not proof. Plenty of domains answer 250 to every recipient, real or invented, because they accept all mail and sort it later. Against those domains the question has no answer left to give.
What a check removes, and what it leaves
Both columns matter. A verifier that only advertises the left one is describing half the job.
| Removed by checking | Still there afterwards |
|---|---|
| Addresses the receiving server rejects outright, which are your hard bounces waiting to happen. | Confirm a mailbox on a catch-all domain. The server accepts every recipient by design, so no product can resolve it. |
| Domains with no mail server at all, including dead company domains and typo domains nobody registered. | Tell you whether a person reads the mailbox. A valid address abandoned three years ago still checks out as valid. |
| Syntax that cannot be a working address, and disposable domains built for throwaway signups. | Remove pristine spam traps. They look identical to working mailboxes at the protocol level, so anyone claiming to strip them is selling probability. |
| Duplicates, which most lists carry more of than their owners expect. | Fix a consent problem. An address can be perfectly deliverable and still not be yours to mail. |
What to do instead
Count what came back unresolved before you do anything
That number tells you what kind of list you have. A consumer file resolves almost completely. A business file with a quarter of its rows unresolved is normal and not a sign the check failed.
Remove the rejections, segment the rest
Hard rejections go. Catch-all and unknown rows move to their own segment and get a smaller, closely watched send rather than a deletion.
Re-import your suppression list first
Unsubscribes and complaints that went missing in a migration will outrank every other kind of damage on this list, and no verifier detects them.
Fix the intake so this does not recur
A list that needed cleaning will need it again in a year unless the form checks addresses as they are typed. Batch cleaning treats the symptom.
Questions people ask
Does cleaning an email list send anyone an email?
No. The check asks the receiving server about the recipient and then closes the connection, before the point where a message body would be sent. Nothing arrives in anyone's inbox.
How much of a list usually comes back unresolved?
It depends entirely on the mix. Consumer lists on the large free providers resolve well. Business lists carry far more catch-all domains, and around 28% of one 10,000-address B2B benchmark sat on domains that accept every recipient.
How often should a list be cleaned?
Before any large send, after any import, and on a schedule matched to how fast the list decays. A list you mail weekly needs less attention than one that sat untouched for two years.
Is a free email list cleaner good enough?
For a sample, yes, and running one is a reasonable way to see the shape of your list before paying for anything. Our free tier is 100 checks a month with no card, which is enough to measure the unresolved share.
Related list problems
Measure your own list before you clean it
100 free checks a month, no card. The rows we could not get an answer on come back labeled rather than guessed at, and we never bill for them.