A provider limit is a quota, not a judgement. When you hit one you get a specific rejection, your mail stops, and nothing about your reputation has changed. This matters because the fix is completely different: a quota wall is solved by adding mailboxes, slowing your rate or raising a limit, while a reputation wall is solved by sending less mail to better recipients for several weeks. Teams routinely spend a month on the second problem when they had the first.
The limits worth knowing
| Provider | Practical limit | Notes |
|---|---|---|
| Google Workspace | 2,000 messages per user per day; 500 for trial accounts | counted per recipient, and external recipients counted separately |
| Gmail (free) | far lower, roughly a few hundred recipients a day | not a business sending platform |
| Microsoft 365 | 10,000 recipients per day per mailbox, with a per-minute rate limit around 30 messages | the rate limit bites long before the daily cap |
| Amazon SES | sandbox: 200 per day, 1 message per second, verified recipients only | production access and higher quotas are requested per account |
| SendGrid, Mailgun, Postmark | plan-dependent, enforced per account | check your own plan; do not assume a number |
Treat vendor-published numbers as the authority and re-check them before you build a schedule around one — these change. The shape, though, is stable: mailbox providers cap per user per day, relays cap per account with a rate limit, and new accounts everywhere start lower than established ones.
Tell a quota apart from a block
The rejection text is explicit if you read it.
Quota: 550 5.4.5 Daily user sending limit exceeded 452 4.5.3 Too many recipients 454 4.7.0 Too many messages per minute Reputation or policy: 550 5.7.1 Our system has detected an unusual rate of unsolicited mail 421 4.7.0 ... try again later (pace, at the receiving provider) 550 5.7.1 Service unavailable; blocked using <list>
A limit error names the limit. A filtering rejection talks about unsolicited mail, policy or a blocklist. If you are not sure, the giveaway is scope: a quota stops everything from one mailbox at once and resets on a clock; filtering is per-recipient-provider and partial.
How limits shape the ramp
Build your warm-up plan backwards from the ceiling, not forwards from zero.
- Work out capacity first. Target daily volume divided by a safe per-mailbox rate gives the number of mailboxes you need. Safe means well under the provider cap — if Google allows 2,000, plan on a few hundred, because the cap is not a target and sustained sending near it is itself unusual behaviour.
- Add mailboxes rather than pushing one harder. Ten mailboxes at 40 a day is safer and more resilient than two at 200. Each mailbox still needs its own warm-up.
- Respect the rate limit as well as the daily cap. Microsoft's per-minute limit means a queue that drains in bursts will fail even when the daily total is modest. Spread sends across the working day.
- Remember recipient counting. Most providers count recipients, not messages, so one email to fifty people spends fifty units.
- Account for everything else on the mailbox. The quota is shared with the human's own mail, your CRM, your support tool and your calendar invites.
What not to do
Do not plan to send at the published cap. The number is a hard ceiling with no headroom, and a cap-hugging pattern on a new mailbox is exactly what filtering treats as suspicious. Do not create ten mailboxes on day one and send from all of them immediately — unwarmed mailboxes on a new domain, all starting at once, is the clearest possible signal of bulk outreach. Do not use distribution lists or BCC to stretch a quota; recipient counting defeats it and BCC-heavy mail filters badly anyway. Do not raise SES production limits and then immediately send at the new ceiling — the quota changed, your history did not. And do not treat a quota rejection as a reason to buy more domains; it is a reason to add mailboxes or slow down.
Interpreting results near a limit
If you were throttled by a quota, your placement data for that period is incomplete rather than bad, and comparing it to a normal week will mislead you. Mark the period and re-measure. Conversely, if volume was within quota and placement dropped, the cause is reputation or content and no amount of extra capacity will help.
Judge the reputation half on what is still measurable. Google Postmaster Tools retired its domain and IP reputation charts, so the available evidence is Gmail spam rate — under 0.3%, ideally under 0.1% — plus per-provider seed placement. And keep the ramp honest: volume ramping and genuine engagement are defensible, but fabricated interaction between controlled mailboxes is the behaviour Validity's Heatwave blocklist, launched 3 September 2026, exists to catch. It lists domains rather than IPs, launched with over a million of them, and its thresholds are not public, so no tool can promise you exemption.
MailPilot works within provider limits by design: it warms mailboxes over OAuth for Gmail and Outlook or over SMTP/IMAP for Amazon SES, SendGrid, Mailgun, Postmark and custom relays, every plan includes unlimited mailboxes so capacity comes from adding mailboxes rather than pushing one to its ceiling, and it measures real per-provider inbox placement and per-mailbox reputation daily.
Ready to reach the inbox every time?
Automated email warmup across real peer mailboxes. Live inbox placement monitoring. Free 7-day trial - no credit card required.
Start free trial