How Can You Tell When a Mailbox Is Ready for Cold Outreach?

Almost everyone asks the readiness question in days. "Is two weeks enough?" "My tool says 21 days, can I start?" Days are the easiest thing to measure and the least informative. A mailbox that sat idle for 20 days on a domain with a broken DKIM record is not ready. A mailbox that's been having real, threaded conversations for 12 days on a properly authenticated domain, landing in Primary at every major provider, almost certainly is.

Readiness is a set of observable signals, not a countdown. Below are the six checks worth running before you point a mailbox at a cold list, what thresholds actually mean something, and what to do when one of them fails. If all six pass, start sending. If one fails, fix that specific thing — sending anyway just converts a small, cheap problem into an expensive reputation problem.

1. Authentication and DNS are boring

This is a pass/fail gate, not a gradient, and it's the one people skip because they "set it up weeks ago." Send a message to a test address, open the raw headers, and read them. You want SPF to pass with your actual sending source included and the record under the 10-lookup limit. You want DKIM to pass, signed by your sending domain rather than a platform's domain, at 2048-bit where the provider supports it. And you want DMARC to pass with alignment — meaning the DKIM d= domain or the SPF return-path domain matches the domain in your From header.

Alignment is where most people get caught. A message can show SPF pass and DKIM pass in the headers and still fail DMARC because neither one aligns with the visible sender. That failure is invisible in most sending dashboards and it quietly costs you placement at Google and Yahoo, both of which now expect bulk senders to be authenticated properly. Fix authentication before day one of warm-up, not before day one of sending — failures during the ramp teach receiving providers exactly the wrong thing about your domain at the moment they're forming an opinion.

While you're in DNS, check two more things: that any custom tracking domain you use isn't a shared or previously burned hostname, and that you haven't left a catch-all enabled on the sending domain. A catch-all accepts garbage, inflates noise, and makes your own bounce data useless.

2. Seed placement lands in Primary, not Promotions, not spam

This is the single most predictive check you can run, and it's the one that separates "the tool says I'm warm" from actually being warm. Send a realistic message — the same copy style, signature, and links you plan to use in the campaign, not a stripped-down test — to a set of seed mailboxes spread across Gmail, Outlook.com, Microsoft 365, Yahoo, and at least one corporate Google Workspace tenant.

Ready looks like zero percent spam folder placement across all seeds. Not "one out of ten." Zero. A single spam hit in a ten-seed test implies roughly 10% of your list is heading to junk, and at 500 sends that's 50 people who never see you and a handful who report you. Overall inbox placement should be 90% or better, Gmail seeds should land in Primary rather than Promotions, and Microsoft seeds shouldn't be arriving with a high spam confidence level.

Read the failure patterns, because they point at different fixes. Gmail Promotions placement is almost always a copy and link problem, not a warm-up problem — too many links, image-heavy HTML, marketing phrasing. Microsoft junking you while Gmail inboxes you usually means domain age plus link reputation. Universal spam placement means authentication or blocklists. Seed tests do have limits: seed mailboxes have no engagement history with you, so results skew slightly pessimistic at Gmail and slightly optimistic at aggressive corporate filters. Treat them as a directional gate. Here's how to run one properly.

3. Warm-up traffic shows real conversation shape

The volume of warm-up emails a mailbox has exchanged means very little on its own. The shape of that traffic means a lot. Three numbers are worth pulling out of your warm-up dashboard before you declare a mailbox ready.

Reply depth comes first. Threads that run two or three messages deep signal that this sender gets responses from humans. A warm-up system that fires one message and marks it read is generating a much thinner signal than one producing actual back-and-forth. Second is the spam rescue rate — the share of warm-up mail that landed in spam and had to be pulled out. On a healthy new mailbox this typically starts somewhere between 5% and 20% and should trend under 2% by the end of the ramp. If you're still at 10% on day 21, more warm-up will not fix it; something upstream is wrong and you need to go back to checks one and two. Third is the folder placement of the warm-up mail itself. If your warm-up messages are sitting in Promotions or Updates, cold mail written in the same voice will do the same.

One honest caveat: warm-up networks that shuttle mail between thousands of unrelated mailboxes produce a pattern that filters can and do recognize. Warm-up that happens with real labeled messages inside real workspaces holds up better than pooled network traffic. The mechanism matters more than the message count.

4. The volume history is a curve, not a cliff

Receiving providers build their model of you from a trend line, so the shape of your recent send history is itself a readiness signal. A mailbox that sent 5, then 8, then 12, then 18, then 25 messages a day across three weeks has taught Gmail and Microsoft something coherent. A mailbox that sent 40 a day for three days, went silent for a week, then jumped back to 40 has taught them nothing except that it's unpredictable.

Pull the last 14 days of send history and ask three questions. Was there any day where volume more than doubled over the previous day? Were there gaps of two or more days with no sends? Did sending happen at hours that make no sense for the recipient's timezone? A yes to any of those means the curve is broken, and your first week of real sending should be noticeably gentler than you planned — start at half the volume you intended and rebuild the slope. Our 30-day warm-up schedule shows what a clean ramp actually looks like day by day.

5. Provider-side signals are clean

Your sending tool's dashboard tells you what it did. The provider-side signals tell you what the receiving side thinks, which is the only opinion that matters. Check Google Postmaster Tools first: domain reputation should read Medium or High. "Low" means you're already being filtered and you shouldn't be adding cold volume. "Bad" means stop entirely and work on recovery. Reported spam rate should sit under 0.10% — Google's stated ceiling is 0.30%, but at cold-email volumes you want a wide margin because a single bad day can spike the average.

Then read your SMTP logs for 421 and 4.7.x deferrals. Temporary throttling from Google or Microsoft is the most direct feedback you will ever get: it means you're moving faster than they're comfortable with. If you're seeing deferrals during warm-up, you are not ready. On the Microsoft side, check the Restricted Entities list in the Defender portal — a restricted mailbox is a hard stop rather than a slow-down, and continuing to push through it makes the restriction stickier. Finally, check Spamhaus, SURBL, and any link domains you plan to use in the copy. A perfectly clean sending domain paired with a listed link domain will still get filtered, and people lose weeks chasing the wrong variable.

6. Bounce and complaint history is near zero

If the mailbox has sent anything real — even to a small test segment — look at the results. A hard bounce rate above 2-3% means your list hygiene is going to sink the mailbox regardless of how good the warm-up was. Complaints should be effectively zero; one spam complaint per 1,000 sends is already at the edge of what's sustainable, and providers weight complaints far more heavily than they weight volume.

This is where verification pays for itself twice over. A mailbox can pass all five previous checks and still get wrecked in its first week by a list with 12% dead addresses. Readiness is a property of the mailbox plus the list, not the mailbox alone.

A quick readiness scorecard

What "ready" actually permits

Passing all six checks does not mean unlimited sending. A ready mailbox on Google Workspace or Microsoft 365 can typically start at 20-30 cold sends a day and climb roughly 20% per week to a practical ceiling of 40-50. If your plan needs 300 a day out of a single mailbox, no amount of warm-up makes that safe — you need more mailboxes and more domains, not a longer ramp.

Readiness also expires. A mailbox that passes today and then gets pointed at a badly targeted list will fail inside ten days, because the complaint and non-engagement signals overwhelm whatever goodwill the warm-up built. Re-run the seed test every two or three weeks and immediately after any change to copy, link structure, or daily volume.

Red flags that mean wait, regardless of the calendar

• Domain registered less than 14 days ago — new domains get filtered hard for their first few weeks no matter how well the mailbox behaves
• A purchased or pre-aged account whose prior sending history you can't inspect
• A mailbox that was warmed and then sat idle for 10+ days; restart at about half your previous volume
• Warm-up running on a subdomain whose parent domain has an active blocklist listing
• Drafts scoring above 3.0 on SpamAssassin before you've sent anything — that's a copy problem that will follow the mailbox everywhere you move it

How Sendvanta answers this question for you

Checking six signals by hand stops scaling somewhere around five mailboxes. Sendvanta scores readiness in layers — mailbox-level behavior, domain reputation, DKIM and authentication status, and IP health — and rolls them into a health score you can look at before attaching a mailbox to a campaign. Warm-up runs with real labeled messages inside your own workspace rather than a shared pool, drafts get SpamAssassin-scored before send, and seed inbox-placement tests run against real mailboxes at the major providers.

The part that matters most day to day: when a health signal crosses a risk threshold mid-campaign, Sendvanta slows the mailbox down or pauses it automatically instead of letting it keep burning while you're asleep. Everything sends through your own Gmail, Microsoft 365, or SMTP mailboxes — nothing is proxied through shared IPs. Connecting your own mailboxes is free on every plan, including the $0 Free tier (1,000 active leads, 3,000 emails a month, up to 10 SMTP mailboxes active). Paid plans start at $29/mo. See the full feature list.

The short version

Stop counting days. Ask instead: do seeds land in Primary with zero spam hits, does authentication align, is the warm-up spam rescue rate under 2%, does Postmaster show Medium or better, is the send history a smooth curve, and is the bounce rate near zero? Six yeses means send — starting at 25 a day, not 250. Any single no means fix that thing first, because every additional day of warm-up on a broken foundation is just teaching the filters something you'll spend months undoing.