Your email bounce rate and complaint rate are two of the clearest signals mailbox providers use to judge you — and letting either climb is one of the fastest ways to wreck your deliverability. Managing them well, by suppressing bad addresses promptly and respecting rate limits, is the day-to-day discipline that protects everything else. This guide explains hard versus soft bounces, what counts as a good email bounce rate and complaint rate, how suppression lists and feedback loops work, and what SMTP throttling codes mean. It is part of our email deliverability hub.
Hard bounce vs soft bounce
A hard bounce is a permanent failure — the address or domain does not exist, or the server permanently rejects it. In SMTP terms it is a 5xx response (RFC 3463 class 5). You must suppress a hard bounce immediately and never send to it again. A soft bounce is a temporary failure — a full mailbox, a server that is down, or a throttle — signalled by a 4xx response (a “persistent transient failure”). You do not suppress a soft bounce on the first occurrence; you retry with backoff, and only suppress after repeated failures across a retry window (sending platforms typically retry for up to 72 hours). Treating a soft bounce like a hard bounce throws away deliverable contacts; treating a hard bounce like a soft bounce inflates your email bounce rate and damages reputation.
What is a good email bounce rate?
There is no single official email bounce rate threshold, but the published enforcement numbers give a clear picture. Amazon SES — which publishes the most concrete figures — puts an account “under review” when its bounce rate reaches 5% and may pause sending at 10%. The widely-cited “keep total bounces under about 2%” is an industry rule of thumb rather than an official limit, but it is a sensible ceiling to aim below. The key is to watch the trend: a sudden jump in your email bounce rate almost always means a list-quality problem — stale addresses, a bad import, or a purchased list. (Amazon SES reputation metrics.)
Email bounce rate vs complaint rate
These are two different numbers and you should never conflate them. Your email bounce rate measures undeliverable addresses; your complaint rate measures recipients who hit “report spam.” Complaints are the more dangerous of the two: Google requires bulk senders to keep the spam complaint rate below 0.3%, and advises a target under 0.1%, measured daily. A complaint rate creeping toward 0.3% will hurt your sender reputation and inbox placement long before it triggers a formal block, so treat 0.1% as your real ceiling for complaints and keep bounces well below a couple of percent.
Suppression lists and feedback loops
A suppression list is the set of addresses you must never send to again — hard bounces, complaints, unsubscribes, and soft bounces that exhausted their retries — and it is checked before every send. Prompt suppression is what protects your reputation: re-sending to a bad address inflates your email bounce rate, and a hard-bounce address today can become a recycled spam trap tomorrow.
Feedback loops (FBLs) are how providers tell you when a recipient reports your mail as spam, so you can suppress that person. Yahoo and AOL offer a DKIM-based complaint feedback loop, Microsoft provides one through JMRP, and many others are available via Validity’s universal FBL. The most-misunderstood case is Gmail: it has no per-complainer feedback loop. Postmaster Tools shows only an aggregate daily spam rate, and Gmail’s separate ESP feedback loop returns anonymised, per-campaign data — never individual addresses. Because you cannot suppress individual Gmail complainers, clean lists and an easy unsubscribe matter more than ever.
Throttling and SMTP rate limits
Mailbox providers throttle senders they do not fully trust, and they communicate this through SMTP reply codes (RFC 5321). The first digit tells you everything: a 4xx code is a temporary deferral where a retry may succeed, while a 5xx code is permanent and must never be retried.
| Code | Meaning | What to do |
|---|---|---|
| 421 | Service not available / rate-limited | Defer and retry with backoff |
| 451 4.7.1 | Greylisting (try again later) | Retry after a delay |
| 452 | Too many recipients / storage | Slow down, retry |
| 550 | Mailbox unavailable (permanent) | Hard bounce — suppress |
| 554 | Transaction failed / blocked as spam | Investigate; do not retry blindly |
Greylisting is a common defence where a receiver deliberately defers the first attempt from an unfamiliar sender; a legitimate mail server that retries is then accepted. The correct response to any 4xx is a queue with exponential backoff and jitter, throttled per receiving domain, capped at 24–72 hours before you give up and record a final bounce. Never retry a 5xx.
How to lower your email bounce rate
- Verify addresses at signup and use confirmed opt-in to keep invalid addresses off the list in the first place.
- Suppress hard bounces immediately and honour the suppression list on every send.
- Clean inactive contacts regularly, since lists decay by roughly a fifth each year.
- Never buy or scrape lists — they are full of invalid addresses and spam traps that spike both bounces and complaints.
- Authenticate and warm up properly — see SPF, DKIM, DMARC and our warm-up guide — so throttling-driven soft bounces fade as trust builds.
Why list hygiene is the real fix
Almost every bounce and complaint problem traces back to list quality, which is why hygiene — not clever handling after the fact — is the durable solution. A high email bounce rate means your list contains addresses that no longer exist, usually because it was imported from an old source, bought, or simply allowed to age without cleaning. A high complaint rate means you are reaching people who do not remember signing up or no longer want your mail. Both are list problems wearing different costumes, and both are prevented the same way: collect addresses through clear, confirmed opt-in; verify them at the point of capture; and remove the ones that bounce or go inactive before they cause damage.
The stakes are higher than the immediate metrics suggest because of recycled spam traps. When a real person abandons an address, the provider eventually reclaims it, and after a dormant period may turn it into a spam trap. An address that gave you a hard bounce or simply stopped engaging a year ago can therefore become a trap that gets you blacklisted if you keep mailing it. Prompt suppression and regular pruning are not housekeeping niceties — they are how you avoid mailing tomorrow’s spam traps. A smaller list of people who want your mail will always out-deliver a bigger list you never clean.
Feedback loops by provider
Setting up the feedback loops that are available to you closes the gap between a complaint happening and you acting on it. Yahoo and AOL offer a complaint feedback loop keyed to your DKIM signature, so enrolling means signing your mail correctly and registering your domain. Microsoft’s JMRP, accessed through SNDS, reports complaints from the Outlook ecosystem at the IP level. Many smaller providers participate through Validity’s universal feedback loop, which consolidates reports from numerous mailbox providers into one feed. Enrol in every loop you can, because each one lets you suppress a complainer before they complain again.
The unavoidable exception is Gmail, which has no per-complainer feedback loop at all — only the aggregate spam rate in Postmaster Tools and an anonymised, per-campaign feed. Since you can never know which individual Gmail user reported you, you cannot suppress them reactively, so the only defence is proactive: keep complaint rates low through permission, relevance, and a one-click unsubscribe that is easier to use than the spam button. In other words, for the world’s biggest mailbox provider, list hygiene and easy opt-out are not just best practice — they are the only practice available. All of this protects the sender reputation that ultimately decides your inbox placement.
Respecting rate limits keeps mail flowing
Throttling is not only a warm-up phenomenon — even established senders hit per-provider rate limits during big campaigns, and how you respond determines whether the mail eventually lands or turns into bounces. The rule is to treat every 4xx deferral as a request to slow down, not a failure: queue the message, back off, and retry at a measured pace rather than re-sending immediately, which only signals aggression and deepens the throttle. Good sending platforms do this automatically with per-domain rate limiting and exponential backoff, but if you run your own mail server you must implement it yourself. Spreading a large send out over time, rather than firing it all at once, keeps you under provider limits and protects the email bounce rate that a flood of giving-up deferrals would otherwise inflate. Patience at the SMTP layer is, quietly, part of good deliverability.
Check it yourself: use our free email tools to check whether your sending IP is on a major blacklist — a common hidden cause of rising bounces.
Related reading
- Email warm-up guide
- Sender reputation guide
- Email blacklist guide
- Inbox placement guide
- Email authentication statistics 2026
Email bounce rate FAQ
What is the difference between a hard bounce and a soft bounce?
A hard bounce is a permanent failure (a 5xx response) — the address does not exist — and you must suppress it immediately. A soft bounce is a temporary failure (a 4xx response) such as a full mailbox or a throttle; you retry it with backoff and only suppress after repeated failures.
What is an acceptable complaint rate?
Keep your spam complaint rate below 0.1%. Google’s hard ceiling for bulk senders is 0.3% measured daily, but you want a comfortable margin beneath that because complaints are the strongest negative reputation signal there is.
Does Gmail have a feedback loop like Yahoo and Microsoft?
Not a per-complainer one. Gmail only exposes an aggregate spam rate in Postmaster Tools and an anonymised, per-campaign ESP feedback loop — never individual complaining addresses. That is why you cannot suppress specific Gmail complainers and why list hygiene and easy unsubscribe are essential.
What does an SMTP 421 response mean?
421 means the receiving server is temporarily unavailable or rate-limiting you — it is the classic throttling code. Treat it as a soft, temporary failure: defer the message and retry with exponential backoff rather than hammering the server, which only makes throttling worse.
What is greylisting?
Greylisting is an anti-spam technique where a receiver deliberately defers the first delivery attempt from an unknown sender with a temporary error. A legitimate mail server automatically retries and is then accepted, while many spam tools never retry. It is normal and resolves itself as long as your server retries correctly.
Why should I suppress bad addresses immediately?
Because re-sending to known-bad addresses inflates your email bounce rate and complaint rate, and abandoned addresses can be recycled into spam traps. Prompt suppression keeps those metrics low, which protects the sender reputation that determines whether your good contacts ever see your mail. Delayed suppression is one of the most common self-inflicted deliverability wounds: every extra send to a confirmed hard bounce or complainer is pure downside with no chance of reaching anyone, so the moment an address bounces hard or reports you, it should never receive another message.
Cite this article
Raj Kapoor. "Email Bounce Rate & Complaint Management (2026)." ToolTrusted, June 24, 2026, https://tooltrusted.com/bounce-rate-guide/.
Raj Kapoor. (2026). Email Bounce Rate & Complaint Management (2026). ToolTrusted. https://tooltrusted.com/bounce-rate-guide/
Notice something outdated or incorrect in this review? Let us know below, and our team will update it within 24 hours.