Email bounce rate is the percentage of sent messages a receiving server refused, and the line that matters is 2% — above it, providers treat your sending as evidence that you do not know who is on your list. Under 1% is healthy. Between 1% and 2% something has changed and is worth finding.
The rate itself is rarely the problem. Bounce rate is a lagging indicator of how a list was built, which is why fixing it by deleting bounces works for exactly one send and then the number climbs again.
This guide covers how to classify a bounce correctly, what each cause actually means, and the order to fix them in.
The thresholds
These are the lines providers themselves draw. They apply regardless of industry or list size.
| Bounce rate | Reading |
|---|---|
| Under 1% | Healthy |
| 1–2% | Warning. Something changed recently |
| Over 2% | Emergency. Providers read this as poor list control |
| Over 5% | Expect throttling or rejection, and expect it quickly |
A single send above 2% is not a crisis on its own. A pattern is. One bad number after an import is a diagnosis; the same number three sends running is a reputation problem forming.
The comparison that matters is against your own history, not an industry benchmark. A send that jumps from 0.4% to 1.9% is a serious event even though 1.9% sits inside the warning band. Benchmark ranges for other metrics are in email marketing benchmarks, and the caution there applies here: published figures are averaged across senders whose list practices you know nothing about.
Hard and soft bounces are different failures
They get reported in one number and they mean opposite things.
| Hard bounce | Soft bounce | |
|---|---|---|
| What it means | The address does not exist | The address exists, delivery failed anyway |
| Typical cause | Typo, closed account, invented address, dead domain | Full mailbox, server down, message too large, temporary throttling |
| Will it resolve? | Never | Usually, on retry |
| Action | Suppress immediately and permanently | Retry, and suppress after repeated failure |
| Reputation impact | Direct and significant | Minimal unless persistent |
Hard bounces are the number that damages you. They tell a provider that you are mailing addresses you have never verified, which is a statement about your acquisition rather than your infrastructure.
Soft bounces are mostly noise — until the same address soft-bounces on several consecutive sends, at which point it has stopped being temporary. A dormant mailbox that filled up and was never emptied is functionally dead, and it is also the raw material for a recycled spam trap.
A practical soft-bounce rule: suppress after three consecutive soft bounces, or after 30 days of continuous soft bouncing, whichever comes first.
Your platform's classification is not authoritative
Sending platforms classify bounces inconsistently. The same rejection can be recorded as hard by one provider and soft by another, and some record a blocklist rejection — which is neither — as a hard bounce, quietly deleting subscribers who were perfectly valid.
Read the bounce text, not just the category. The receiving server states its reason in the message, and the reason is the diagnosis.
| Bounce text | What it actually is |
|---|---|
550 5.1.1 user unknown / no such user |
A genuine hard bounce. Suppress |
550 naming a list — "listed", "blocked using" |
A blocklist rejection, not a bad address. See blocklist removal |
550 5.7.515 / 550 5.7.26 |
Authentication failure. Your setup, not their mailbox |
452 / 552 mailbox full / over quota |
Soft. Retry, then sunset |
451 4.7.1 / "greylisted" |
Temporary defer. No action needed |
Two of those five would be recorded as bounces and are not address problems at all. A domain rejecting your mail for authentication failure produces a bounce on every address at that domain, which looks exactly like a bad list and is not one. That misdiagnosis — deleting valid subscribers while the actual fault sits in DNS — is the most expensive mistake on this page, and the Email Header Analyzer settles it in ninety seconds.
What a high bounce rate is telling you
Bounce rate is downstream of acquisition, so a number outside the healthy band names a source rather than a send.
| Pattern | Cause |
|---|---|
| Sudden jump after an import | The imported file. Almost always the answer |
| Sudden jump with no import | Authentication or a blocklist, not addresses — check before deleting anyone |
| Slow creep over many months | Natural list decay, and no sunset policy |
| High only on one segment | That segment's acquisition source |
| High from day one on a new list | Purchased, scraped, or inherited |
| B2B list decaying fast | Job changes — B2B addresses die faster than consumer ones |
On an inherited list, verification routinely removes 5–20% of addresses. Above 30% means the list was purchased or is very old, and at that point the honest question is whether it should be mailed at all rather than how to clean it.
The B2B case deserves its own note. Corporate addresses die when people change jobs, and that happens at a rate consumer addresses do not approach. A B2B list left untouched for a year will bounce heavily through no fault of the sender — which makes cadence and re-verification a standing cost of B2B email rather than a one-off cleanup.
How to bring a high bounce rate down
Six steps. The first two exist because the most common response to a high bounce rate — deleting the bounces — is sometimes the wrong action entirely.
Step 1 — Read ten bounce messages before changing anything
The text, not the rate. Two minutes. If they name a blocklist or an authentication failure, stop here — you have a different problem and deleting subscribers will not touch it.
Step 2 — Rule out authentication
If bounces cluster at one provider, or arrived suddenly with no import, check your records with the Email Domain Scanner and count your SPF lookups with the SPF Lookup Checker. A second SPF record or a crossed lookup limit produces mass rejection that reads as a bad list. The full diagnostic order is in why your emails suddenly went to spam.
Step 3 — Suppress every hard bounce, permanently
Not delete — suppress, so nobody re-imports them in six months while solving a different problem. Most platforms do this automatically; confirm yours actually is, because the ones that do not fail silently.
Step 4 — Isolate the source
Segment by signup source and date and find the batch. One import usually accounts for most of the damage, and the remedy is to stop mailing that batch rather than to clean the whole list.
Step 5 — Verify before the next send, if the list is inherited
This is the one situation where paid verification clearly earns its cost — an inherited or long-dormant list, where the alternative is discovering the bounce rate by sending. What verification can and cannot do is in email verification tools.
Step 6 — Set the policy that stops it recurring
Automatic hard-bounce suppression, soft-bounce suppression after three consecutive failures, and a sunset schedule. Without the policy the number climbs back, because the cause was never the addresses you deleted.
Then rebuild the sending pattern. After a heavy clean, send to your most engaged segment first and widen gradually rather than returning to full-list sends immediately.
Where this stops being a metric and becomes a recovery. If your bounce rate crossed 2% and placement has already dropped, cleaning the list is step one of a longer sequence. The Email Deliverability Playbook has the day-by-day recovery protocol, the re-permission templates for a list you inherited, and the 100-point audit scorecard that tells you whether the list is worth saving before you spend a month saving it. Get the Playbook →
Frequently asked questions
What is a good email bounce rate?
Under 1% is healthy, 1–2% is a warning that something changed recently, and over 2% is treated by providers as evidence of poor list control. Compare against your own history rather than an industry average — a jump from 0.4% to 1.9% is a serious event even though it sits inside the warning band.
What is the difference between a hard and a soft bounce?
A hard bounce means the address does not exist and will never accept mail, so it should be suppressed immediately and permanently. A soft bounce means the address exists but delivery failed for a temporary reason such as a full mailbox, and it usually resolves on retry. Hard bounces are the ones that damage sender reputation.
Why did my bounce rate suddenly increase?
Usually a recent import — that is the answer most of the time. If nothing was imported, suspect authentication or a blocklist rather than addresses, because both produce mass rejections that look identical to a bad list. Read ten bounce messages before deleting anyone.
Should I delete bounced email addresses?
Suppress rather than delete. Suppression keeps a record so the address cannot be re-imported later by somebody solving a different problem, which is how deleted bounces reappear months afterwards. Suppress hard bounces immediately and soft bounces after three consecutive failures.
How many soft bounces before I remove an address?
Three consecutive soft bounces, or 30 days of continuous soft bouncing, whichever comes first. Beyond that the mailbox is functionally dead, and an abandoned mailbox is the raw material providers convert into recycled spam traps.
Does a high bounce rate affect deliverability?
Yes, directly. A high hard-bounce rate tells providers you are mailing addresses you have never verified, which is a statement about acquisition rather than infrastructure. Sustained rates above 2% attract throttling, and above 5% rejection follows quickly.
Can authentication problems look like a high bounce rate?
Yes, and this is the most expensive misdiagnosis in email. A domain rejecting your mail for SPF, DKIM or DMARC failure bounces every address at that domain at once, which looks exactly like a bad list. Codes such as 550 5.7.515 and 550 5.7.26 are authentication failures, not bad addresses.
How much of an inherited list is usually invalid?
Verification routinely removes 5–20% of an inherited list. Above 30% means the list was purchased or is very old, and at that point the question is whether it should be mailed at all rather than how best to clean it.
What to do next
Open your last campaign report and read ten bounce messages — the text, not the rate. That one habit separates "delete these subscribers" from "fix this DNS record", and it takes two minutes.
If the bounces name a provider rather than a mailbox, run the Email Domain Scanner before touching your list.
Free: The 60-Minute Email Authentication Fix — worth running first, because a meaningful share of bounce spikes are authentication failures wearing a bounce message.
Related guides
- Email list building that doesn't wreck deliverability — where a high bounce rate actually comes from
- Email spam traps: how to find and remove them — what ignored bounces turn into
- Email verification tools: which are worth paying for — when paid cleaning is justified
- Why your emails suddenly went to spam — the diagnostic order for a sudden change
- Email blocklists: how to check and get removed — when the bounce names a list
- How to read Google Postmaster Tools — the delivery errors panel
Free tools used in this guide
The short version
- Read ten bounce messages before changing anythingThe text, not the rate. If they name a blocklist or an authentication failure, you have a different problem and deleting subscribers will not touch it.
- Rule out authenticationIf bounces cluster at one provider or arrived with no import, check your DNS records and count your SPF lookups. A second SPF record or a crossed lookup limit produces mass rejection that reads as a bad list.
- Suppress every hard bounce permanentlySuppress rather than delete, so nobody re-imports them later. Confirm your platform is actually doing this, because the ones that do not fail silently.
- Isolate the sourceSegment by signup source and date to find the batch. One import usually accounts for most of the damage, and the remedy is to stop mailing that batch.
- Verify before the next send if the list is inheritedAn inherited or long-dormant list is the one case where paid verification clearly earns its cost, because the alternative is discovering the bounce rate by sending.
- Set the policy that stops it recurringAutomatic hard-bounce suppression, soft-bounce suppression after three consecutive failures, and a sunset schedule. Without the policy the number climbs back.
Free: The 60-Minute Email Authentication Fix
A no-fluff checklist to set up SPF, DKIM & DMARC correctly and pass Gmail & Yahoo's sender requirements.
Muhammad Basim has worked in digital marketing since 2013, focused on email deliverability and AI-assisted content production. He is the author of The Email Deliverability Playbook and The Email Copywriting Playbook.
Related Articles
Abandoned Cart Emails: Why Timing Beats Discounting
An abandoned cart email sequence is three automated emails sent over 48 hours to someone who added an item to a cart without buying, with the first sent one to four hours after abandonment. Speed has a larger effect on recovery than any copy decision in the flow. The discount belongs in email three, and […]
Email Blocklists: How to Check, and How to Get Removed
An email blocklist is a published list of IP addresses or domains that receiving servers consult before accepting mail, and only a handful of them meaningfully affect delivery. Checking tools flag dozens. Most of those listings change nothing about where your email lands. Knowing which listings matter is the difference between a ten-minute fix and […]
Why Your Emails Suddenly Went to Spam: The Diagnostic Order
Email that went to spam suddenly has one of five causes: a DNS or authentication change, a list or sending change, a linked-domain problem, a blocklist listing, or a complaint spike. Each one leaves a different trace, and each can be confirmed or ruled out in minutes. The order matters more than the effort. Working […]