Check an email list before you import it
Check a list before you import it, not after the first campaign bounces. Remove addresses that cannot work, domains that cannot receive mail and duplicates, and think twice about role addresses such as info@. Email Digit can run that check on the file during import and tells you plainly that the result is an estimate.
Why the first send to an old list is the risky one
Mailbox providers judge a sender partly by how many of its messages bounce. An established sender with years of clean history can absorb a bad week. A domain you set up last month cannot: it has little history, so a burst of bounces on its first large send is most of what providers know about it.
That is exactly when old lists get imported. People switch tools, export everything they have, including addresses collected years ago, and send a “we have moved” campaign to all of it. Addresses go stale as people change jobs and abandon inboxes, and the new tool has no record of which ones bounced in the old one. So the dirtiest send of the year goes out on the newest domain.
What to look for in a list
- Addresses that cannot be an email at all. Missing @ signs, spaces, typos in the structure. These fail every time.
- Domains with no mail server. If a domain does not exist, or publishes no MX record, mail to it almost never arrives. (The standard lets a sender fall back to the domain’s own address record, but few real mailboxes depend on that.) Expired company domains are a common source.
- Role addresses. info@, sales@, support@ and the like reach a shared box rather than a person. They rarely bounce, but they tend to attract spam complaints, which can hurt a sender more than a bounce does.
- Duplicates. The same person twice inflates your counts and can mean they receive a campaign twice.
Then sort what is left by recency. If you know who clicked, replied or bought recently, send to them first and widen out over a few campaigns, rather than mailing everyone on day one.
How the check works in Email Digit
When you import a CSV, after you map a column to Email, you can run a check on the file before anything is imported. It reports:
| Count | How it is found |
|---|---|
| Invalid addresses | The address does not have the shape of an email |
| No mail server | The domain does not exist, or answers that it has no MX record |
| Role addresses | The part before the @ is a shared name such as info, sales or support |
| Duplicates | The same address appears more than once |
Invalid addresses and dead domains together give a risk level: low at 2% or less, moderate up to 5%, high up to 10%, and severe above that. The result names the domains with no mail server, comes with advice in plain words, and treats role addresses as a complaint risk rather than a bounce risk.
Mail servers are looked up once per domain, not once per address, so a large file with a few hundred domains is checked quickly. If a domain’s DNS cannot be reached, its addresses are counted as unknown rather than bad, because telling you to delete good contacts would be worse than saying nothing.
Why it is labelled an estimate
The check looks at each address and whether its domain accepts mail. It does not contact each mailbox to ask whether the person exists, so a valid-looking address at a working domain can still bounce. A clean result means the file has no obvious problems, not that every remaining address belongs to someone who reads it.
If the check comes back high
- Remove the invalid addresses and the addresses at domains with no mail server. They will almost all bounce.
- Look at the role addresses and keep only the ones you know a person reads.
- Split the rest by how recently each person engaged, if your old tool can tell you, and import the recent group first.
- Consider a paid verification service for the older part of the list. It can check mailboxes in a way this check deliberately does not.
What happens after you import
- Duplicates are merged. With dedupe on, an address already in your contacts is updated rather than added again.
- Your do-not-send list still applies. Anyone on it who appears in the file comes in as unsubscribed and is never made sendable.
- A bad import pauses itself. Email Digit measures bounces for each import separately. If more than 5% of sends to one import bounce, once at least 40 have gone out, further sending to that import is paused while the rest of your contacts keep receiving mail. An Admin can resume it.
- Limits. The importer sends a file larger than 50,000 rows in 50,000-row parts for you, and each part counts as its own import for the bounce pause. Rows without a usable email are skipped, and you can download them as their own CSV.
A clean list is half of a good first send. The other half is a sending domain that authenticates; check your domain before the campaign goes out, and read How to keep email consent records for the consent statement every import asks for.