Throttling is the polite version of being blocked. Instead of rejecting your mail outright, the receiving server says "we will take it, but slowly". You see deferral messages, send times stretch from minutes to hours, and your scheduled campaign is still going out three days later. Here is what causes it and how to recover.
Signs you are being throttled
- SMTP responses in the 4xx range, especially 421 and 451.
- "Temporary failure, retry later" in your ESP's bounce logs.
- Send time per recipient climbing from milliseconds to seconds.
- Particular domains (often Gmail or Outlook) consistently deferring.
Common throttling triggers
Volume spike from a new IP or domain
You went from 1,000 sends per day to 50,000. The receiving server treats this as suspicious until your reputation supports the volume. Solution: warm up. Volume should never more than double day-over-day.
Sudden burst within a day
Sending 100,000 messages in 5 minutes vs spread across 8 hours. Bursts trip rate limits even if your daily volume is normal. Solution: configure your ESP to throttle outbound across the sending window.
High bounce rate signal
Bouncing 5% of recipients in the first 10 minutes of a campaign signals a stale list. Receivers slow you down to limit damage. Solution: verify before send.
Spam complaint rate spike
Above 0.1% complaints triggers automatic throttling at Gmail and Yahoo. Solution: make unsubscribe obvious and instant.
Authentication intermittency
If some of your mail passes SPF/DKIM/DMARC and some fails (perhaps because of an upstream IP rotation), receivers throttle until they understand your sending pattern.
What to do when you are throttled
- Stop the in-flight campaign. Pausing minimizes additional damage.
- Diagnose the trigger. Check bounce rate, complaint rate, send velocity, authentication status. Usually one is obviously off.
- Fix the underlying issue. Verify the list, fix authentication, lower velocity.
- Resume slowly. Restart at 25% of your prior send rate. Build back over 3 to 7 days.
Reading throttling response codes
| Code | Meaning | Action |
|---|---|---|
| 421 | Service not available, try later | Pause + investigate |
| 421 4.7.0 | IP temporarily blocked | Stop sending from this IP for 24h |
| 451 | Local error, try again | Retry with backoff |
| 452 | Insufficient system storage | Slow down |
| 421 4.7.28 | IP rate-limited (Gmail) | Reduce send rate immediately |
Provider-specific throttling rules
- Gmail: rate-limits based on sending domain reputation. Up to ~10,000 messages per hour for trusted senders; less for new ones.
- Yahoo: throttles based on IP reputation. Aggressive limits for unknown senders.
- Microsoft (Outlook): throttles based on IP and domain. Will defer rather than reject for marginal senders.
- Corporate (Office 365, Exchange): often have aggressive internal limits set by IT.
FAQ
Will throttling resolve on its own?
If the trigger was a one-time spike, yes, within a few hours. If the trigger is ongoing (bad list, low engagement, bad auth), no, the throttling will deepen until you fix the cause.
How long does recovery take?
For acute throttling from a single bad campaign: 24 to 72 hours after fixing the cause. For sustained throttling from systemic issues: 2 to 6 weeks.
Can I switch ESPs to escape throttling?
Sometimes. The IP changes, your domain reputation does not. If the cause is bad list or low engagement, the new ESP will throttle you the same way once it sees the same signals.
Prevent throttling with hygiene
The most common trigger is high bounce rate from an unverified list. Run verification before every campaign and most throttling never starts.