ON THIS PAGE
By Efe Berke Colaker, Founder at GetleadReviewed by the Getlead editorial team for accuracy. Last updated August 2026.
A new domain and a new mailbox have no history. So filters have nothing to weigh you against except the shape of what you are doing right now. A first campaign at full volume is the largest thing that mailbox has ever done, which is precisely the pattern disposable spam infrastructure produces.
Warm-up fixes that by manufacturing the missing history. It is not a trick, it is evidence: small volumes, genuine opens and replies, growing steadily, until the account looks like an ordinary working mailbox.
Below is a four week schedule with daily numbers, what to check at each stage. And the points where the correct decision is to stop and go backwards. It assumes warm-up traffic is already running on the mailbox.
The four week schedule
Volumes below are per mailbox, not per domain. If you are running four mailboxes on a domain, the domain sees four times these numbers. That is fine as long as each individual account stays in range.
The ceiling in week five is not a hard limit, it is the point where further increases should be earned rather than scheduled. Providers allow far more, with Google Workspace permitting 2,000 messages a day, but that allowance is designed for mail people asked to receive.
Week 1: history with no risk
Nothing goes to a prospect this week. The mailbox exchanges warm-up traffic with other participating accounts, which open the messages, reply to some of them. And pull anything that lands in spam back into the inbox.
Those three actions are the signals that matter. An opened and answered message is the strongest evidence a filter has that a sender is wanted. And manufacturing a small amount of it early is what buys room later.
Before any of it starts, the authentication has to be in place. SPF, DKIM and DMARC published and aligned, an MX record so the domain can receive. And the domain registered in Postmaster Tools so you can watch reputation from day one.
Use the week to prepare the list as well. The first campaign in week three should be the cleanest segment you own, and cleaning it properly takes longer than people expect.
What week one buys you
Ten messages a day sounds pointless next to a target of hundreds. The value is not volume, it is the ratio. A mailbox where nearly every message is opened and a third are answered has an engagement profile that no cold campaign can produce. And that profile is what carries you through weeks three and four.
It also surfaces configuration mistakes while they cost nothing. If warm-up messages are landing in spam among participating accounts, the problem is authentication or the domain itself. And finding that out in week one rather than in week three is the entire point.
Week 2: the first real recipients
Warm-up continues underneath, and a small verified segment goes out on top of it. Small means tens of messages a day, and verified means confirmed valid rather than merely imported.
The point of this week is measurement rather than pipeline. You are looking for a bounce rate near zero, no complaints at all. And replies arriving in the inbox rather than the spam folder, which you confirm with seed accounts at Gmail, Outlook and Yahoo.
If bounces appear at this volume the list is the problem, not the ramp. Stop, re-verify, and repeat the week. In our verification data 23.9% of raw records are invalid. So an unverified segment produces bounces immediately regardless of how carefully the mailbox was warmed.
Choosing the test segment
Pick the segment you understand best, not the largest one. Twenty prospects where you know the company, the role and the reason for writing produces a cleaner signal than two hundred assembled from a filter you have not checked.
Recent records beat old ones here too. A segment sourced this month has had almost no time to decay. So a bounce in week two points at the sourcing rather than at the passage of time.
Warm-up is the practice of building a sending history on a new mailbox through small volumes of opened and answered mail before any campaign traffic runs through it.
For example, a mailbox in week two sends 20 messages a day, of which perhaps 15 are warm-up traffic that gets opened and replied to. That ratio produces an engagement profile no cold campaign can match, and it is what carries the account through weeks three and four.
Weeks 3 and 4: real campaigns, one variable at a time
Week three runs one campaign to one segment. Not three campaigns, not a mixed audience. The reason is diagnostic: when something moves you need to know what caused it. And a single segment gives you an answer instead of a suspicion.
Week four introduces the second segment and raises daily volume toward 50. This is also where follow-ups start landing, so total message volume rises faster than new prospect volume, which is easy to miss when planning capacity.
- Check bounce rate after every send, per segment rather than per campaign.
- Check Postmaster Tools twice a week for complaint rate and domain reputation.
- Re-run seed inbox tests when the volume step happens, not only at the start.
- Watch replies per thousand delivered, which is the metric that reveals placement problems early.
- Keep warm-up running the entire time rather than switching it off once campaigns begin.
The last item is the one teams drop first. Campaign replies are sparse by nature, so on a quiet week the only engagement signal your mailbox produces is warm-up traffic. Turning it off removes the evidence that you are a mailbox worth delivering to.
The signals that mean stop rather than continue
A ramp schedule is a plan, not a commitment. Four signals justify holding at the current level or stepping back, and none of them are ambiguous.
- Bounce rate above 2% on any send. The list is the problem. Re-verify before the next step, not after.
- Any complaint at week two or three volumes. At tens of messages a day, one complaint is a large percentage and a clear targeting signal.
- Seed messages landing in spam. Placement has degraded and more volume will accelerate it.
- A wall of 4xx responses from one provider. You are sending too fast for that receiver, so pace down rather than push through.
Stepping back costs a week. Pushing through costs a domain and the two to three weeks it takes to warm a replacement, plus whatever pipeline the pause removes.
The complaint signal is the sharpest of the four. The published ceiling for bulk senders is 0.3% of delivered messages with 0.1% recommended. And at 30 messages a day a single complaint dwarfs both.
Scaling after the ramp
Once a mailbox is at steady state, growth comes from two places and only one of them is per mailbox volume.
The third item on the right deserves attention because it is the most common strategic error in outbound. A poor reply rate is a targeting problem. And sending twice as much of the wrong message multiplies the complaint exposure rather than the pipeline.
Plan capacity from the meetings you need rather than from the volume the tool allows. At median reply rates, most teams need far fewer sends than they assume. And a smaller number of well targeted messages is also the cheapest way to keep every deliverability metric comfortable.
Sources and method
First-party data (Getlead, 2026): the verification split of 43.4% confirmed valid, 23.9% invalid, 16.7% catch-all and 16.0% unknown comes from 383,368 addresses analyzed through live SMTP verification. And the 0.51% bounce rate comes from 34,973 tracked sends, aggregated and anonymized at campaign level. Full method in our cold email benchmark study.
External sources: the 0.3% spam complaint ceiling, authentication requirements and reputation reporting come from the Google Workspace sender guidelines and Postmaster Tools, 2026; per account daily sending caps come from the Google Workspace sending limits documentation, 2026.
Figures were checked in August 2026 and provider policies change without notice.
Frequently asked questions
How long should I warm up a new email account?
About four weeks to reach useful volume. Week one is warm-up traffic only at 5 to 15 messages a day, week two adds a small verified segment at 15 to 25, week three runs the first real campaign at 25 to 35. And week four reaches 35 to 50 per mailbox per day.
How many emails can I send in the first week?
Five to fifteen a day per mailbox, and none of them to prospects. The first week is warm-up traffic only, which builds the opened and answered history that filters weigh when they have no other evidence about your domain.
Should I keep warm-up running after the ramp?
Yes. Campaign replies are sparse, so on quiet weeks warm-up traffic is the only engagement signal your mailbox produces. Switching it off once campaigns begin removes the evidence that the mailbox is worth delivering to.
What if bounces appear during the ramp?
Stop and re-verify. A bounce rate above 2% on any send means the list rather than the ramp is the problem. And continuing to increase volume on a bad list is how a carefully warmed domain gets burned in a single afternoon.
Can I speed up the ramp?
Not usefully. The shape of the increase is what filters read. And a sudden jump from tens to hundreds of messages looks exactly like a purchased list being mailed. Adding more mailboxes is the correct way to get more volume sooner.
How fast can I grow volume after week four?
About 20% a week, and only while bounce rate stays under 1%, complaint rate stays comfortably under 0.1% and reply rate is stable or rising. If any of the three moves the wrong way, return to the previous level for a fortnight.