What’s the impact of frequency and volume on filter scoring?

Volume and frequency are two of the cheapest signals a filter has, and they get used hard. Mailbox providers track how much you send, how often, and whether today looks like yesterday. When the numbers move in ways that don't match your history, the filter slows you down or starts dropping mail into spam while it figures out what changed.

The classic trigger is a volume jump. If your IP normally sends 1,000 messages a day to Gmail and you push 100,000 tomorrow, Gmail does not assume you grew overnight. It assumes the account got popped or you bought a list. You will see 4xx temporary deferrals (the "try again later" responses), throttling, and a chunk of mail sitting in queue. For more on what those rejection codes actually mean, see 5xx permanent rejection vs 4xx temporary.

This is why IP warming exists. Warming is the practice of growing daily volume on a fresh IP slowly so providers can build a baseline. Most operational guides land on a four to six week curve: start at a few hundred sends per provider per day, roughly double every two or three days, and stop pushing once you hit your target volume. Google's Postmaster guidance explicitly tells senders to increase volume gradually and warns that abrupt jumps look like spam attacks. M3AAWG's sender best practices recommends staged ramps for any new IP, new domain, or major list growth.

Frequency matters too, separately from raw volume. A sender who mails Gmail every Tuesday for a year has a rhythm. The filter learns it. When that same sender goes dark for three months and then blasts the entire list, the inactivity itself is the problem. The list is older, engagement will be worse, and the spike does not match the established pattern. Reputation systems treat a return-from-silence blast the same way they treat a volume spike on a cold IP: with suspicion. This ties directly into how filters evaluate ongoing sender behavior. The model is built on the assumption that legitimate senders are predictable.

A few practical patterns we see in client data:

• Gmail starts deferring inside a single hour if you exceed roughly 5 to 10 times your trailing seven-day average for that IP and domain pair. You do not get a warning email.
• Microsoft (Outlook.com, Hotmail) is stricter on new IPs. Expect 4xx deferrals from the first send until you have a small but consistent volume trail.
• Yahoo applies similar logic and publishes its expectations in the Yahoo Postmaster best practices.

What "consistent" means in practice:

1. Send on a schedule. Daily, weekly, twice a week, pick something and hold it.
2. Grow volume in steps, not jumps. Doubling is fine. Going from 1k to 100k is not.
3. If you have to pause, do not return with a full-list blast. Warm back up: send to your most engaged 10 percent first, then expand over a week or two.
4. Split traffic by mailbox provider when you check averages. Gmail, Microsoft, and Yahoo each track you separately. A 5x spike at Gmail is a spike at Gmail, even if your total volume looks flat.

Filters use your sending pattern as a cheap fraud check. Predictable senders get the benefit of the doubt. Erratic senders get throttled, deferred, or routed to spam while the filter waits to see what you do next. If you want more on the watchful-eye logic, the answer on how filters decide email is spam covers it in depth.