Our first-time customers often have a lot to ask about our "unknown" result code. While every situation varies, we break down the commonly asked questions about this result group, below.
What are unknown results?
“Unknown” is the result code that is used for anything but a conclusive result (i.e. anything other than “verified”, “undeliverable”, “unreachable”, “illegitimate” or “disposable”). An email address that falls under this category has passed all of our domain and formatting checks, and our rule-based checks (e.g. not a known, disposable domain, not a known trap, etc.), but we were unable to definitively determine whether or not the mailbox, itself, is deliverable.
Why do unknown results occur?
The most common reason this happens is that the domain is configured to be “accept-all," meaning that its mail servers do not reject invalid addresses when we query them, but rather, accept all addresses that we try. Normally, when we query a mail server with an invalid email address, we’ll say, “Hi, mail server, does this address exist?” and it will say, “Nope, no such mailbox exists”. These “accept-all” domains will instead say, “Yes, please proceed” — even when the email address does not exist. Senders only receive a non-delivery report after actually sending an email message. So without sending a message (which we never do), it is impossible to validate addresses from these domains.
How are unknown results identified?
We identify these domains by not accepting a mail server’s reply at face value unless we’ve seen it recently reject an address that is statistically certainly invalid (e.g. email@example.com). If it accepts the address, we know the domain is accept-all, and we therefore don’t trust subsequent replies to be meaningful. If this is the case, rather than saying “verified” merely because the mail server accepted the address, we try and avoid giving you a false sense of confidence, and instead say, “unknown.”
Besides accept-all domains, what other records can return unknown results?
Another common reason unknown results occur, is because we cannot get a conclusive result within a certain period of time. This occurs through API integrations when a timeout is set. On our website we set a timeout of a few seconds, and it’s possible that with at least one of your test emails, we defaulted to unknown because we didn’t get a reply quickly enough.
What should be the expected percentage of unknowns within a file?
There are a few areas where "unknown" results are more common. For B2B lists, these sorts of domains are relatively common, often yielding "unknown" rates of anywhere from 10% to 60%; 20~30% is most typical, however. In the US, we have slightly higher domain/ISP coverage, so "unknown" results may be more common for international data.
What best practices can you suggest when dealing with unknown results?
We suggest uploading your verified results, along with a small percentage of your unknowns, then gradually adding in the remaining unknowns over time. Depending on your verified-to-unknown results ratio, we suggest splitting up the unknowns into one or more batches and upload over a period of time. It is important to keep in mind that, if your total verified results are, for example, 36% of your total, we can typically expect 36% of your unknowns to be valid, deliverable email addresses. Also understand that you will see bounces from the unknowns.
What should I take away from this?
The best interpretation of “unknown” results is neither good nor bad. It’s not a warning sign, but neither is it a vote of confidence. If you give us a segment of addresses, and we say “unknown” for them, we essentially haven’t told you anything. As a result, we don’t charge you for “unknown” results. You shouldn’t reject these addresses, as they—like most of your addresses—contain valid email addresses.