Does email warm-up actually work?
By Chris Aylott
Yes. With one condition that changes everything: the seed network providing the trust signal has to be made of real inboxes.
What warm-up actually does
ISPs make delivery decisions based on reputation signals. A brand-new domain has no reputation — it could be a legitimate company or a spam operation that registered this morning. ISPs treat new and unknown domains conservatively.
Warm-up builds a controlled sending history before you need the domain for real campaigns. Your tool sends low volumes of email to seed mailboxes in its network. Those mailboxes open the mail, engage with it, and signal positive history back to ISPs. Over four to eight weeks, the domain's reputation lifts from unknown to trusted.
That mechanism is real and it works. The question is: what are the seed mailboxes in that network?
Warm-up genuinely helps for:
- New domains created specifically for outbound campaigns
- Domains dormant for more than 90 days
- Domains migrating to a new sending IP or ESP
- Domains that have been blocklisted and are recovering
The question your warm-up tool probably cannot answer
Ask the tool you are currently using: "Are your seed mailboxes real customer inboxes or synthetic addresses created specifically for warm-up?"
This is not an obscure question. It determines whether the reputation signals your tool generates are recognised by ISPs as valid — or have been discounted as artificial.
Most warm-up tools cannot answer this favorably. Their networks are built from synthetic addresses — mailboxes created and owned by the warm-up provider, used for no purpose other than receiving warm-up traffic. Some of these networks contain 20,000 to 35,000 synthetic addresses receiving thousands of warm-up emails per mailbox per day from thousands of different senders.
ISPs see this pattern at scale. Their models learn it. The trust signals generated by a synthetic network gradually become untrustworthy — in the same way that fake reviews eventually stop moving purchase decisions.
Why network quality matters more than network size
A synthetic network of 35,000 addresses that ISPs have fingerprinted provides weaker trust signals than a genuine network of a few hundred real business inboxes that behave exactly as ISPs expect real inboxes to behave.
A real business inbox receives normal commercial email, newsletters, replies, forwards — and a handful of warm-up emails per week. The warm-up traffic is invisible within normal inbox activity. The engagement profile of that mailbox is genuine. The signal it generates is a real vote.
The question is not how many seed addresses the network has. The question is whether the addresses are real.
How InboxPlease's network is built
Every seed mailbox is a real customer inbox — with structural constraints no competitor uses.
Invite-only, real inboxes
Participation requires an invitation. Members are existing InboxPlease customers with active business inboxes — not synthetic addresses. They receive a 50% subscription discount.
Hard cap: 200/day
Every seed mailbox is hard-capped at 200 warm-up emails per day. Members typically receive around 3 per week alongside normal business mail.
Send-only permissions
InboxPlease requires only permission to send to the participating mailbox. Not full read/write. Not OAuth that hands over the full inbox. No other warm-up provider operates this way.
The question to ask any warm-up provider
- Are your seed mailboxes real customer inboxes or synthetic addresses?
- What is the daily email cap per seed mailbox?
- Do you require full read/write access, or only send permissions?
- What is the sample size behind the placement score on my dashboard?
If the answers are not published, the score on your dashboard is calculated against an audience you cannot evaluate.
What warm-up cannot fix
For established domains where email is going to spam, warm-up is almost certainly not the root cause.
List quality
Unverified lists and disengaged subscribers generate spam signals from real recipients that no seed-network warming can offset.
Content
Spam filters model recipient intent. A warmed domain sending irrelevant cold outreach will still generate spam placements.
Authentication gaps
If SPF, DKIM, or DMARC is misconfigured, warm-up will not compensate. Authentication is evaluated before reputation in most providers' delivery stack.
Volume cliff
Warm-up runs at 50–500 emails/day. Jumping immediately to 10,000+ for a campaign registers as a suspicious spike. The ramp must continue.
Does warm-up work? Yes — when the underlying network is real. If you have done everything correctly and email is still landing in spam, the first question to ask your warm-up provider is what their seed network is actually made of. If they cannot tell you, the score on your dashboard is measured against something ISPs have already stopped trusting.
Warm-up that's measured against real inboxes
InboxPlease's seed network is made entirely of real customer mailboxes — 200/day cap, invite-only, send-only permissions. Not a synthetic farm.
Start free