Email warm-up: what it builds, and what it cannot

Email warm-up means sending and receiving mail from a new mailbox in small, rising volumes so that providers see a history before your first real campaign arrives. Every cold email tool now offers it, usually free, and the feature is sold with more confidence than the mechanism deserves.

It does something real. It also cannot do the thing most people buy it for.

What warm-up actually builds

A domain registered last week has no sending history. Providers have no basis to distinguish it from the thousands of throwaway domains registered that day for spam, so the safe assumption is the unfavourable one. Warm-up gives them evidence: mail leaving the domain steadily, being opened, replied to, and moved out of spam rather than reported.

Mechanically, a warm-up service connects your mailbox to a pool of other mailboxes. Yours sends to them, they open and reply, and the exchange is logged as ordinary correspondence. Volume rises over weeks — a typical ramp starts around 10 to 20 messages a day and climbs over four to eight weeks.

That produces genuine signals. Sending consistency, reply rates and an absence of complaints are all things receivers weigh, and a domain with six weeks of quiet, well-behaved history is treated differently from one with none.

The part that is oversold

Gmail’s filtering is heavily personalised. A large part of the decision about whether your message reaches a particular person is about the relationship between your domain and that recipient — whether they have opened your mail before, replied, archived it, or marked it as spam.

A warm-up pool cannot generate that. The mailboxes replying to you belong to strangers your prospects have no connection to. You are building a domain-level reputation, which matters, while the recipient-level signal — often the decisive one — stays empty until real people engage with real mail.

This is why warm-up cannot rescue a bad campaign. If your first real send goes to a poorly targeted list and people ignore or report it, the reputation you spent six weeks building is spent in a week. Warm-up buys you a fair hearing. It does not decide the verdict.

Pools can be detected

Warm-up networks are a known pattern, and the pattern is detectable in principle: the same mailboxes corresponding with thousands of unrelated new domains, replying at implausible rates, on predictable schedules. Vendors compete on exactly this point — Warmup Inbox, for instance, argues its network of real human inboxes differs from “shared pools of fake mailboxes” and “scripted engagement providers learn to filter”.

Whether any particular network is distinguishable from ordinary mail is not something a buyer can verify from outside, and not something we can tell you yet either. It is on the list to measure, because it is the difference between a feature that works and one that theatre.

Do you need a dedicated tool?

Probably not, and this is the practical question. Woodpecker, Smartlead and Instantly all bundle warm-up at no extra cost. Warmup Inbox sells it standalone from $15 per inbox per month — and that per-inbox pricing matters, because a cold email setup runs many mailboxes, so eight inboxes is $120 a month for a feature three of your alternatives include free.

The case for paying separately rests entirely on the bundled version being worse. That is a measurable claim and an untested one. Until somebody measures it, the default should be to use what you already have.

What to do alongside it

  • Authenticate first. Warm-up on a domain with broken SPF or DKIM is wasted effort — see the DMARC guide.
  • Clean the list. Bounces undo warm-up faster than warm-up builds. List cleaning costs a few dollars.
  • Ramp real sending too. Going from warm-up volumes to a full campaign overnight is its own spike. Increase gradually.
  • Watch the complaint rate. Google and Yahoo publish a ceiling of 0.3% for bulk senders and name 0.1% as the level to stay under. Three complaints in a thousand is the limit, and it is stricter than most people assume.
  • Keep warm-up running. Reputation decays. Most tools continue it in the background alongside live campaigns, which is the right default.

How long before sending

Two to four weeks is the common recommendation for a new domain, longer for one that has been burned. There is no published threshold because providers do not publish one, and any tool quoting an exact number of days is quoting its own preference rather than a rule.

The more useful signal is what Google Postmaster Tools reports about your domain reputation once you have enough volume for it to display data. That is real mail to real recipients, which is the only measurement that counts.