Email bounce rate
What it means, what counts as good, and how to bring a high one down.
What a bounce rate actually measures
Your bounce rate is the share of a send that never reached a mailbox and came back undelivered. Send 10,000 messages, have 300 rejected, and your bounce rate is 3%.
It is worth saying plainly, because it causes real confusion: this has nothing to do with the bounce rate in web analytics. That one measures visitors who land on a page and leave without interacting. Same words, unrelated metric, and most of what you find searching the bare phrase is about the other one.
Hard bounces and soft bounces are not the same problem
Email platforms separate them because they mean different things and call for different responses.
| Hard bounce | Soft bounce | |
|---|---|---|
| What happened | Permanent failure. The address does not exist, or the domain cannot receive mail at all. | Temporary failure. Mailbox full, server down, message too large. |
| Will it work later? | No. | Possibly. Most platforms retry. |
| What to do | Remove it. Most platforms suppress it automatically after one. | Leave it, but watch it. Repeated soft bounces to the same address usually mean it is dead. |
Hard bounces are the ones that damage you. They are the measurable evidence that you are sending to addresses you never checked, and mailbox providers read them exactly that way.
The reply codes that tell them apart, and the rule for when a repeated soft bounce should be treated as a hard one, are in hard bounce vs soft bounce.
What counts as a good bounce rate
Start with the part almost every article on this subject gets wrong. No mailbox provider publishes a bounce rate limit.
Google's sender guidelines and Yahoo's sender best practices each set exactly one numeric threshold, and it is not about bounces. It is the spam complaint rate, which must stay below 0.3% — Google adds that you should aim below 0.10% and never reach 0.30%.
On bounces, both are deliberately qualitative. Google says to reduce your sending volume if messages start bouncing or being deferred. Yahoo says to monitor hard and soft bounces along with inactive recipients and remove them promptly, and recommends confirmed opt-in to keep invalid addresses out in the first place. Neither names a number.
So when you read that "the acceptable bounce rate is 2%", that figure is an industry aggregate — email platforms reporting on their own customers — not a rule written by anyone with the power to block your mail. It is still useful as orientation, which is why it is here:
- Under 1% — healthy. A well-kept list of people who opted in.
- 1% to 2% — normal, and where most senders sit.
- 2% to 5% — a list-hygiene problem that is costing you inbox placement.
- Above 5% — the range where email platforms start warning, throttling, or suspending accounts.
Those ranges are commonly reported across ESP benchmark publications rather than measured by us — we send no campaigns and have no data of our own to quote. Each platform also counts bounces slightly differently and samples only its own customers, so treat them as the rough shape of the thing. Your own trend over time is the number that actually matters, and it is the only one measured on your list rather than someone else's.
Why the number matters more than it looks
A 4% bounce rate sounds survivable. Ninety-six percent got through. The reason it is not survivable is that the bounces are not the damage — they are the symptom.
Mailbox providers use bounce rates to judge whether you know who you are sending to. A sender who repeatedly mails addresses that do not exist looks like a sender who acquired the list rather than earned it, and that judgement applies to your whole domain, not to the failed messages. The practical result is that your mail to valid addresses starts landing in spam folders.
So the cost of a high bounce rate is not the small fraction that bounced. It is the deliverability of everything else.
This is also why the absence of a published bounce limit is not permission. The rule that does exist — spam complaints under 0.3% — sits directly downstream of the thing that is not ruled. Bounces tell a provider your list is unverified, that judgement pushes your mail toward spam folders, and mail that lands in spam folders gets marked as spam. The number nobody polices feeds the number everybody does.
What actually causes a high bounce rate
- Typos at signup.
gmial.com,hotmial.co.uk, a missing letter in a surname. These enter the list looking perfectly valid. - Natural decay. People change jobs and abandon addresses. A B2B list loses a meaningful share of its addresses every year without anyone doing anything wrong.
- Bought or rented lists. The single fastest way to a bounce rate that gets an account suspended, and the reason most platforms prohibit it outright.
- Long gaps between sends. A list mailed once a year has a year of decay in it, arriving all at once.
- Spam traps. Abandoned addresses reactivated by providers to catch senders who never clean. These do not always bounce — sometimes they just quietly destroy your reputation.
How to bring it down
In roughly the order that gives you the most improvement for the least work:
- Check the list before the send, not after. Bounces are counted against you the moment they happen. Verifying beforehand is the only step that prevents the damage rather than recording it.
- Use confirmed opt-in. A subscriber who has to click a link in an email cannot enter your list with a typo, because a typo never receives the link.
- Remove hard bounces immediately. Most platforms do this for you. If yours does not, do it manually — mailing a known-dead address twice is worse than once.
- Re-engage or remove the long-inactive. Someone who has not opened anything in two years is a deliverability liability whether or not they bounce.
- Send more regularly. Frequent sends spread decay thinly instead of letting it accumulate into one damaging send.
A limit worth knowing about
No verification service can settle every address. Some mail servers accept everything sent to them and decide afterwards whether the mailbox exists — a configuration called catch-all — and from outside there is no way to tell a real mailbox from an imaginary one on such a domain.
Any service claiming 100% accuracy is either not counting these or not telling you about them. The honest handling is to mark them as undetermined and say so, which is what ListCheckup does: an address we cannot settle comes back labelled as such, with the credit returned rather than charged as a guess.
How the detection works, and what to do with those addresses rather than delete them, is in catch-all domains.
Where this leaves you
If your bounce rate is under 1%, leave it alone and keep sending. If it is climbing past 2%, the list is aging faster than you are cleaning it, and the fix is a check before the next send rather than after it.
ListCheckup does exactly that job: it checks a list before you send, shows the price before anything is spent, never bills for duplicates or malformed rows, and gives every result a plain-English reason. It opens in the coming weeks. Leave your address and we will tell you the day it does.
Related: hard bounce vs soft bounce, catch-all domains, and why emails go to spam.