← Back to blog
Domain Warmup

Domain warmup: what most guides get completely wrong

A generic warmup schedule assumes every list behaves the same way at every step. Mine never do, and neither will yours. Following a fixed day-count schedule on a list it wasn't built for can do more damage than sending no warmup at all.
6 min read · Updated July 2026

Search "email warmup schedule" and you'll get the same shape of answer everywhere: day 1 send 50, day 3 send 100, day 7 send 250, ramp up over 30 days. It reads like a settled fact. It isn't. That schedule was built to be generically applicable, which means it was built around no specific list at all, and applying it to your specific list is closer to guessing than diagnosing.

I don't warm up domains on a fixed schedule. I warm them up against a moving check: does engagement hold as volume increases, at every step, before I take the next one. That distinction is the entire difference between warmup that builds reputation and warmup that quietly damages it.

What a fixed schedule actually assumes

A generic 30-day plan assumes your list has uniform, moderate engagement, that your ESP's sending infrastructure behaves the same as every other ESP's, and that mailbox providers score volume increases the same way regardless of what's driving them. None of those assumptions hold in practice.

A list built from years of engaged newsletter subscribers can absorb a faster ramp than the generic schedule allows, because the engagement signal backing it is already strong. A list rebuilt after a platform migration, with addresses that haven't seen mail from you in months, will get punished by the exact same schedule, because volume increased faster than trust did. The schedule can't tell the difference between those two lists. I have to.

Why ramping too fast is worse than ramping too slow

This is the part that surprises people. Warming up too slowly costs you time. Warming up too fast costs you reputation, and reputation is the thing that's expensive and slow to rebuild once it's damaged.

Here's the mechanism: mailbox providers watch the relationship between your sending volume and your engagement rate as volume changes. If you scale from 500 to 5,000 sends per day and your open and click rates hold steady, that reads as a real, growing, wanted mail stream. If you scale the same way and engagement drops or complaint rate ticks up, that reads as a sender who just started blasting a list that doesn't actually want the mail, and that's a much worse signal to correct than simply having sent nothing yet. A brand-new domain with zero history is neutral. A domain that just taught Gmail its own recipients don't want its mail is not.

How I check it At every volume step, I compare open rate, click rate, and complaint rate against the prior step, not against an industry benchmark. A meaningful drop at any step is the signal to hold volume, not increase it, regardless of what day the schedule says I'm on.

What should actually set your ramp rate

None of these show up in a generic day-count schedule, because a generic schedule was never built to account for them. It was built to be a number you could follow without knowing anything about your own list.

What I actually do instead

I set a starting volume based on list quality and infrastructure history, then increase only after confirming engagement held at the prior step, not on a calendar. Some domains ramp faster than any 30-day guide would allow, because the list backing them earned it. Others need to hold at a low volume for longer than any guide suggests, because the evidence says the list isn't ready yet, and pushing forward on schedule would cost more than the delay does.

This is the same logic behind Ramp Planning in the diagnostic: the ramp gets calibrated to what your engagement signals actually show, not to a schedule that assumes every sender looks the same.

Part of the deliverability diagnosis guide
See how warmup and engagement fit into the full diagnostic framework →

Questions I get asked a lot

How long should domain warmup take?

There's no fixed number of days that applies to every sender. It depends on how engagement holds as volume increases. A list with strong existing engagement can ramp faster than a fixed 30-day schedule assumes. A cold or unengaged list needs longer than any generic schedule accounts for.

Can a warmup schedule actually damage a domain?

Yes. Ramping on a fixed day-count schedule without checking whether engagement is holding at each step teaches mailbox providers that increased volume correlates with recipients not wanting the mail. That's a worse signal than warming up too slowly.

Do I need to warm up a domain if I'm only sending a small volume?

Yes. Volume alone isn't what warmup protects against, sending history is. A brand-new domain or IP has no trust built with Gmail, Outlook, or Yahoo regardless of how few emails you're sending. Low volume just means the ramp itself can move faster, not that it can be skipped.

A schedule isn't a ramp plan.
The free diagnostic gets you a ramp calibrated to your actual list, not a calendar built for someone else's.
Start the free diagnostic →
Julian Turgelski
The Diagnostic Blog
hello@julianturgelski.com
Julian's diagnostic console