Your Warmup Inbox Count Can Read 100% While a Third Lands in Spam
TL;DR: A warmup "inbox" count can match the sent count while a large share of the warmup mail landed in spam, because warmup peers rescue emails from spam and, in Cedric Lauster's Smartlead stats, those rescued emails also count toward the inbox number. On 2026-10-13 one mailbox Cedric Lauster manages showed 274 warmup emails sent, 274 in the inbox and 100 in spam. Read the spam count or the warmup reputation built from it, never inbox divided by sent, and pull any mailbox under 99% from campaigns.
Key Takeaways
- A warmup inbox count can equal the sent count while a third of the mail landed in spam, because in Cedric Lauster's Smartlead stats emails rescued from spam are counted as inbox as well.
- On all 7 of 7 mailboxes in Cedric Lauster's fleet that had any warmup mail land in spam, the inbox count equalled the sent count.
- The number that tells the truth is spam divided by sent, or the reputation score built from it: (sent minus spam) divided by sent.
- Eric Nowoslawski of Growth Engine X pulls any mailbox under 99-100% warmup reputation from campaigns without further testing, and replaces it with a warmed spare.
- A perfect warmup score is necessary but not sufficient: it does not prove real prospects see your email in the inbox.
A warmup inbox count can tell you a mailbox is perfect while a third of its warmup mail went to spam. On 2026-10-13 I pulled the warmup stats for 44 Google Workspace mailboxes I manage. One of them had sent 274 warmup emails. The stats said 274 landed in the inbox. They also said 100 landed in spam. Both numbers were true at the same time.
If you only look at inbox divided by sent, that mailbox reads 100%. Its real reputation was 63.5%.
How can a mailbox show 274 sent, 274 inbox and 100 in spam?
Because the inbox number also counts mail that landed in spam first.
Warmup works by joining your mailbox to a pool of other mailboxes that send each other emails, open them and reply. Smartlead's help center describes what happens when a warmup email lands in spam: it is moved to the primary inbox, marked as important and given a priority reply. The point of that rescue is to show Gmail and Outlook that people want your mail.
The stats keep a separate daily count for those rescues, "saved from spam". In my stats, every rescued email also shows up in the inbox count. Smartlead does not document that, but the numbers leave little doubt: on that mailbox 100 emails landed in spam, 100 were saved from spam, and the inbox count still matched all 274 sent.
This is not a one-off. In that pull of 44 mailboxes, 7 had any warmup mail land in spam. On all 7 of them, the inbox count equalled the sent count. The inbox number simply could not see the problem.
Which warmup number should you actually read?
Read the spam count, or the reputation score built from it.
The calculation is simple: reputation equals sent minus spam, divided by sent. For the mailbox above that is (274 minus 100) divided by 274, which is 63.5%. Smartlead's own warmup reputation for the same mailbox read 66%, close to my calculation and nowhere near 100%. So the score is useful. The raw inbox count is what misleads.
Where this bites in practice:
- Anything that computes inbox divided by sent. A spreadsheet, a homemade dashboard, a script that pulls the warmup stats and does the obvious division. It will report 100% on every mailbox with rescued mail.
- Reading the counts by eye. "Sent 274, inbox 274" looks clean. You have to notice the spam column next to it.
- Averaging across a fleet. A fleet average built from inbox counts hides the one mailbox that is going under.
My rule for anything I build or read: the inbox count is never used to judge a mailbox. Spam over sent, every time.
Is an email rescued from spam a good sign?
No. It landed in spam first.
It is tempting to read "saved from spam" as warmup doing its job. In a sense it is: the network is working to repair your reputation. But the fact you need is that Gmail put the email in spam in the first place. A real prospect will not go into their spam folder, find your cold email and mark it as important. In a live campaign that same email simply stays in spam, unread.
So every rescue counts against the mailbox. A mailbox with 100 rescues out of 274 is telling you that roughly a third of its mail goes to spam when nobody is there to rescue it.
What warmup reputation should you pull a mailbox at?
Under 99%.
That threshold comes from Eric Nowoslawski of Growth Engine X. In his August 2026 video on cold email advice he treats anything under 99-100% warmup reputation as a pull trigger. His example: a mailbox at 92% is removed from campaigns automatically. It is not investigated further. A warmed spare takes its place.
In spam-count terms, 99% reputation means at most 1 spam email per 100 warmup emails sent. Over an 8-day window of around 270 warmup emails, that is at most 2 spam emails. The mailbox above had 100.
The reason to pull without investigating is cost. Diagnosing one mailbox takes time and keeps it sending while you look. A warmed spare costs nothing to swap in, as long as you keep spares warming. Eric recommends keeping about 50% extra capacity warming for exactly this.
Does 100% warmup reputation mean you land in the inbox?
No. It means the first check passed.
Eric is just as clear on this side: the rule only works in one direction. A low reputation is enough to pull a mailbox. A perfect one proves nothing on its own, because a mailbox can hold 100% warmup reputation and still land in spam with real prospects.
The reason is the audience. Warmup mail goes to other warmup mailboxes, which open, reply and rescue. Your campaign goes to strangers who do none of that, through content that looks nothing like warmup chatter. Different senders, different content, different engagement.
So I treat warmup as level one of a short sequence:
- Warmup reputation. Under 99%, the mailbox comes out of campaigns. This is the cheapest check and it runs on every mailbox.
- Seed placement tests. Copies of the real campaign email to test inboxes at Google and Microsoft. I read the Google result on its own and never act on one bad day. I wrote up how I read a single bad placement test.
- Reply rate per domain at full volume. The only number that comes from real prospects. If replies drop while the first two checks look clean, follow the order I use to diagnose a reply rate drop.
Each level catches problems the one before it cannot see. Warmup only ever answers one question: is the warmup network itself seeing spam placement?
How do you check your own mailboxes in 10 minutes?
- Open the warmup stats for every sending mailbox, not the fleet average.
- For each one, note sent and spam over the same window.
- Divide spam by sent. Anything over 1% fails the 99% gate.
- Ignore the inbox count completely, and ignore any column that was calculated from it.
- Pull every mailbox that fails from its campaigns and swap in a warmed spare. Leave the pulled one warming.
- For the mailboxes that pass, do not stop there. Check that your placement tests and per-domain reply rates agree.
If you run this and find several mailboxes on the same domain failing together, the problem is the domain, not the mailbox, and the whole domain comes out.
What is the next step if your warmup numbers do not add up?
Start with the spam column. If your dashboard shows 100% everywhere, check how it calculates that number before you trust it. If you want a second pair of eyes on a fleet, book a call with me and we will go through the numbers mailbox by mailbox.
Frequently Asked Questions
Why does my warmup inbox count equal my sent count?
Because emails that warmup peers rescue from spam also showed up in the inbox number in Cedric Lauster's Smartlead stats. Cedric Lauster found that on all 7 of 7 mailboxes in his fleet with any warmup mail in spam, the inbox count equalled the sent count. Check the spam count next to it.
How do I calculate warmup reputation correctly?
Subtract the spam count from the sent count and divide by the sent count. A mailbox with 274 sent and 100 in spam has a reputation of 63.5%, even if its inbox count reads 274.
What warmup reputation is too low?
Eric Nowoslawski of Growth Engine X pulls any mailbox under 99-100% warmup reputation from campaigns automatically, without further testing, and swaps in a warmed spare. 99% means at most 1 spam email per 100 warmup emails sent.
Is "saved from spam" a good sign?
No. A saved email landed in spam first. The warmup network moved it out, but a real prospect will not, so every rescue counts against the mailbox.
Does 100% warmup reputation mean my cold emails reach the inbox?
No. Warmup reputation is necessary but not sufficient. A mailbox can hold 100% and still land in spam with real prospects, so Cedric Lauster also runs seed placement tests and tracks reply rate per domain.
Should I pull a single mailbox or the whole domain?
If one mailbox fails, pull that mailbox. If several mailboxes on the same domain fail together, the problem is the domain's reputation, and the whole domain comes out of campaigns.
