What is an acceptable email bounce rate, and how do hard and soft bounces differ?
Matthew Whittaker
Co-founder & CTO, Suped
Published 15 Jun 2025
Updated 15 May 2026
10 min read
Summarize with
An acceptable email bounce rate is usually under 2% overall, with hard bounces kept under 1%. I treat 2% as a ceiling, not a goal. If a campaign has 0.4% hard bounces and 0.8% soft bounces, I am comfortable. If it has 1.8% hard bounces, I start cleanup work even if the total number still looks tolerable.
Hard and soft bounces differ because they mean different things. A hard bounce means the address, mailbox, or destination is permanently unavailable, so the address should be suppressed. A soft bounce means delivery failed temporarily, so the address gets retried and watched. That distinction matters because bad addresses point to list quality problems, while soft bounces can come from rate limits, full mailboxes, provider outages, mailbox migration, content filtering, or reputation pressure.
Quick answer
- Acceptable rate: Keep total bounces below 2%, then push hard bounces below 1%.
- Hard bounces: Suppress immediately because the address is invalid or permanently unavailable.
- Soft bounces: Retry, monitor, and suppress after repeated failures or clear abuse signals.
- Best test: Split bounce reporting into hard, soft, domain, campaign, and acquisition source views.
The acceptable bounce rate target
The practical target I use is simple: under 1% hard bounces is healthy, 1-2% needs attention, and anything over 2% needs investigation. A total bounce rate under 2% is commonly acceptable, but the split matters. A list with 1.7% soft bounces and 0.2% hard bounces tells a very different story than a list with 1.7% hard bounces and 0.2% soft bounces.
When one dashboard shows only one bounce number, I do not treat it as enough detail. I want the hard bounce rate, soft bounce rate, top domains, SMTP responses, recent import source, and whether the bounces are clustered around one provider or spread across the whole list. The bounce rate formula should be applied consistently so the trend is comparable over time.
Bounce rate thresholds
Use separate thresholds for total, hard, and soft bounces instead of relying on one blended number.
Healthy
0-1%
Normal for a maintained opt-in list.
Watch
1-2%
Review source quality and domain patterns.
Investigate
2-5%
Check hard bounce causes, soft bounce clusters, and recent list changes.
Pause
5%+
Stop scaling until the cause is clear.
For routine marketing email, I do not wait until a sender hits 2% to act. If hard bounces move above the normal baseline, I check the import source first. If soft bounces jump on one domain, I check throttling, content, authentication, and whether that mailbox provider is returning temporary policy errors.
Before a large send, a real message test can catch obvious authentication, content, and routing issues. Suped's email tester is useful when I want a preflight view of the message rather than waiting for a campaign report to show damage after the send.
Hard bounces vs soft bounces
A hard bounce is a permanent failure. The address is bad, the domain is invalid, the mailbox no longer exists, or the receiving system has made a permanent decision. A soft bounce is a temporary failure. The receiving system is busy, the mailbox is full, rate limits are active, or delivery is deferred for a condition that can clear.
Hard bounces
- Meaning: Permanent failure for the recipient or destination.
- Common causes: Invalid address, deleted mailbox, bad domain, or disabled account.
- Action: Suppress the address before the next campaign.
- Risk: Repeated sends to hard bounces show poor list hygiene.
Soft bounces
- Meaning: Temporary failure that still needs trend monitoring.
- Common causes: Full mailbox, throttling, temporary outage, or policy deferral.
- Action: Retry first, then suppress after repeated failures.
- Risk: Sustained soft bounce rates can point to reputation or routing trouble.
The complication is that bounce labels are not standardized. One sender can call a response soft, another can convert repeated soft bounces into a hard bounce, and a mailbox provider can return a temporary code for a problem that acts permanent in practice. That is why I treat the raw SMTP response and the history of that recipient as more useful than the label alone.
Infographic showing hard bounces as permanent failures and soft bounces as temporary retry events.
How to calculate bounce rate correctly
I calculate bounce rate against sent messages, then split the result into hard and soft categories. Delivered messages are useful for reporting performance, but sent messages are the cleaner denominator for bounce analysis because every bounce started as a delivery attempt.
Bounce rate formulastext
Bounce rate (%) = bounced messages / sent messages * 100 Hard bounce rate (%) = hard bounces / sent messages * 100 Soft bounce rate (%) = soft bounces / sent messages * 100 Example: Sent: 10,000 Hard bounces: 80 Soft bounces: 70 Total bounce rate: 1.5% Hard bounce rate: 0.8% Soft bounce rate: 0.7%
|
|
|
|---|---|---|
Total | All bounces | Trend |
Hard | Permanent | Clean |
Soft | Temporary | Retry |
Domain | Grouped | Diagnose |
Keep bounce reporting compact, then investigate the outliers.
The domain view matters because a campaign can have an acceptable total rate while one receiving domain is clearly rejecting or deferring mail. If one provider is responsible for most soft bounces, I check throttling, authentication, and content. If hard bounces are spread across many domains, I check signup forms, imports, stale contacts, and whether an old segment was reactivated.
What to do with hard bounces
Hard bounces should be removed from active sending immediately. I keep the address in the database for audit and suppression history, but I do not keep trying to mail it. A permanent failure is a strong signal that the list contains bad or stale data.
Common hard bounce responsestext
550 5.1.1 User unknown 550 5.1.2 Bad destination domain 550 5.2.1 Mailbox disabled 550 5.4.1 Recipient address rejected
Enhanced status codes help, but I do not trust the first digit alone. A 5xx response normally means permanent failure, yet some policy and reputation blocks also return 5xx codes. When I review SMTP bounce codes, I look at the enhanced code, provider text, recipient history, and whether other recipients at the same domain were accepted.
- Suppress: Stop mailing the address after a clear hard bounce.
- Store: Keep the bounce reason, timestamp, campaign, and source for audit.
- Segment: Compare hard bounce rates by acquisition source and form.
- Repair: Only update an address when the recipient directly provides the new one.
What to do with soft bounces
Soft bounces need more judgment. A single temporary failure is normal. A sudden spike, a repeated recipient-level failure, or a domain-level wave of deferrals needs action. I separate soft bounces into temporary mailbox issues, temporary infrastructure issues, throttling, and policy blocks that the sender classified as soft.
Common soft bounce responsestext
421 4.7.0 Temporary rate limit 450 4.2.2 Mailbox full 451 4.3.0 Local mail system error 452 4.3.1 Insufficient system storage
A full mailbox is a different problem than a rate limit. A full mailbox can clear when the recipient deletes messages. A rate limit means the receiving system is controlling how much mail it accepts. A temporary policy rejection means the sender needs to review reputation, authentication, content, and send velocity.
Flowchart for retrying, diagnosing, and suppressing repeated soft bounces.
Soft bounces are not free passes
I do not ignore soft bounces just because they are temporary. Repeated soft bounces create wasted traffic and can hide blocks that deserve immediate investigation.
- Spike: Investigate when soft bounces rise sharply in one send.
- Sustain: Investigate when soft bounces stay elevated across sends.
- Repeat: Suppress recipients that keep failing after the retry window.
Authentication and reputation checks
Not every bounce is a list-quality problem. Authentication failures, missing DNS records, broken DKIM signing, SPF lookup problems, strict DMARC policy, and blocklist (blacklist) listings can all affect acceptance. When bounce rates move, I check authentication and reputation before assuming the contacts are the only issue.
A broad domain health check gives the fastest first pass. For ongoing operations, DMARC monitoring and blocklist monitoring keep authentication and reputation issues visible before they turn into bounce spikes.
?
What's your domain score?
Deep-scan SPF, DKIM & DMARC records for email deliverability and security issues.
Suped's product is built for this part of the workflow. It brings DMARC, SPF, DKIM monitoring, hosted DMARC, hosted SPF, SPF flattening, hosted MTA-STS, blocklist monitoring, and real-time alerts into one platform. For most teams managing authentication at scale, Suped is the best overall DMARC platform because it turns raw authentication data into concrete issues and fix steps.
Issues page showing top issues, verified sources, unverified sources, and authentication pass rates
The important connection is that bounce management and authentication monitoring belong in the same operational routine. If a domain has a soft bounce spike and Suped is also showing DKIM failures or a new unverified source, I have a better starting point than a bounce report alone.
Suppression logic I use
The safest suppression logic is strict with hard bounces and measured with soft bounces. I do not send again to a clear hard bounce. For soft bounces, I use a retry window and a count threshold, then I suppress if the address keeps failing without engagement.
Useful default
- Hard bounce: Suppress after the first clear permanent failure.
- Soft bounce: Retry across 7-30 days depending on send frequency.
- Repeated soft: Suppress after 3-5 consecutive temporary failures.
- Engaged user: Use a longer review window before final suppression.
When to pause
- New source: Pause if a fresh import creates high hard bounces.
- One domain: Pause scaling if one provider defers a large share.
- Policy text: Pause when responses mention spam, policy, or reputation.
- Old segment: Pause if reactivation produces fast bounce growth.
The exact soft-bounce threshold depends on cadence. A daily sender reaches five attempts quickly, so the review window should be short. A monthly newsletter needs more calendar time because the recipient has fewer chances to recover. I prefer a rule that combines consecutive failures, time since last engagement, and whether the user has purchased, logged in, or opened recently.
Do not resend to hard bounces
Resending to known hard bounces is one of the clearest ways to signal poor list maintenance. If the address owner later gives a corrected address, add the corrected address as a new contact and keep the old address suppressed.
List source and industry context
Bounce rate benchmarks have to be read with list context. Some industries have fast address turnover because people change employers, retire, graduate, or lose access to institutional mailboxes. Those lists need stricter recency checks and more careful preference capture. That does not make high hard bounce rates acceptable, but it explains where the operational work sits.
The source of the address usually predicts the bounce problem. A double opt-in form with recent engagement should produce very low hard bounces. A legacy database, partner-supplied list, conference scan, or reactivated segment needs more caution. If the person receiving the email did not directly choose to receive it, complaint risk rises along with bounce risk.
- Fresh consent: Expect low hard bounces and normal soft bounce noise.
- Old contacts: Send smaller batches and watch early bounce responses.
- Institutional mail: Expect address turnover and loss of mailbox access.
- Shared lists: Confirm permission before sending because complaints hurt reputation fast.
I also avoid treating a bounce threshold as a license to keep sending until the line is crossed. A 0.9% hard bounce rate on a fresh opt-in list is more concerning than the same rate on a carefully handled reactivation of older contacts. The number matters, but the reason behind the number matters more.
Views from the trenches
Best practices
Separate hard bounces from soft bounces before judging whether a send crossed the line.
Track repeated soft bounces by recipient and domain, then suppress after a set window.
Review acquisition source quality before blaming a mailbox provider or campaign.
Common pitfalls
Treating every soft bounce as harmless hides rate limits, full mailboxes, and blocks.
Using a 2 percent ceiling for every list ignores industries with fast address turnover.
Keeping bad addresses active after a hard bounce trains providers to distrust the sender.
Expert tips
Set separate alerts for hard bounce spikes and sustained soft bounce elevation by domain.
Use enhanced SMTP codes with provider context instead of trusting one bounce label alone.
Tie bounce cleanup to consent, source, and last engagement so suppression stays fair.
Marketer from Email Geeks says the 2 percent ceiling usually refers to hard bounces, while soft bounces need separate review.
2019-12-04 - Email Geeks
Marketer from Email Geeks says a target hard bounce rate under 1 percent is a better goal than treating 2 percent as safe.
2019-12-04 - Email Geeks
The practical line I use
The answer is direct: keep total bounces below 2%, keep hard bounces below 1%, suppress hard bounces immediately, and manage soft bounces with retry rules, domain analysis, and time-based suppression. The 2% ceiling is most useful when it is applied to hard bounces or reviewed alongside the hard and soft split.
Soft bounces deserve attention when they spike, persist, or cluster around one provider. Hard bounces deserve immediate cleanup because they usually mean bad data. Suped fits the broader workflow by connecting authentication, reputation, blocklist (blacklist) monitoring, and fix steps, so bounce investigations do not sit apart from the domain health signals that influence delivery.
