Email throttling: controlling your sending rate

Email throttling is a hard cap on how quickly a sender can deliver messages to a receiving mail server. Providers apply it to protect their infrastructure from overload and to filter out spam-like traffic. Senders apply it voluntarily to avoid damaging their IP and domain reputation.

Why providers throttle senders

A major mail provider processes billions of messages a day. When one sender suddenly pushes tens of thousands of messages per minute, it stresses the receiving servers and looks like a spam attack, even if the campaign is entirely legitimate.

Providers respond in a few ways:

• Temporary rejection (4xx) — the server replies with code 421 or 452, meaning "try again later." The sending MTA queues the message and retries after a set interval.
• Connection slowdown — the server accepts the connection but deliberately slows its SMTP responses, stretching each session.
• IP block — if the limits are exceeded badly enough, the provider blocks the IP outright for anything from a few minutes to several days.

Approximate limits at major providers

Exact thresholds are not published, but practitioners have mapped rough values over time:

• Gmail — for a new IP with no sending history, stay under 200-500 messages per hour to gmail.com addresses. An IP with a solid reputation can handle tens of thousands per hour without friction.
• Yahoo/AOL — throttles aggressively on volume spikes. A new IP can typically send 100-200 messages per hour without triggering 421 responses; a sudden jump will get you queued immediately.
• Mail.ru — limits depend on the domain's standing in the Mail.ru Postmaster panel. New senders start with tight caps that loosen as positive history accumulates.
• Outlook/Hotmail — Microsoft throttles hard on unestablished IPs. Error codes 421 TS01 and TS02 signal a temporary block because the message volume looked suspicious.

IP and domain warming

Throttling connects directly to the warmup process. A new IP or domain has no sending history, so providers set the lowest possible limits until they can form a judgment. The goal of warming is to build that history gradually, giving providers time to see that your mail gets opened rather than complained about.

A typical warmup schedule: start at 50-100 messages per day, double the volume every 2-3 days, and reach your target throughput in 2-4 weeks. Send to your most engaged subscribers first — people who will open the message. High engagement early on builds the reputation that loosens the throttle.

Voluntary throttling: why limit yourself

Experienced senders configure throttling on their own side before providers force it. There are practical reasons:

• Sending a million messages in 10 minutes will saturate your own outbound MTA before you even reach the provider.
• Spreading delivery over several hours avoids peak-load pressure on the receiving side, which means fewer temporary rejections.
• A sudden volume spike is a red flag for spam filters. Steady, predictable rates look normal.
• Slower sends give you time to notice problems: if bounce rates or complaint rates start climbing, you can stop the campaign before it damages the whole list.

Configuring throttling in your ESP or MTA

Most email service providers — SendGrid, Mailgun, Amazon SES, Postmark — let you set a per-hour or per-minute sending cap. Some handle throttling automatically: when 421 codes arrive, the platform backs off and retries without any manual adjustment.

On a self-managed MTA, the setup is manual. In Postfix, the relevant parameters are default_destination_rate_delay and default_destination_concurrency_limit. PowerMTA goes further, letting you set per-domain limits independently so you can throttle Gmail tighter than a corporate domain that tolerates higher volume.