Email Warmup Schedule: A Safe 30-Day Sending Plan

Most mailboxes that end up in spam were not killed by bad copy. They were killed in week two, when someone decided that 14 days of automated warm-up traffic entitled them to send 50 cold emails a day. Google and Microsoft do not score you on how long you warmed up. They score you on the shape of your sending curve, who engages with your mail, and how many people ignore or report it.

This post is the detailed schedule: what to send on which day, how much cold volume to layer on top, and when to stop ramping. If you need the fundamentals first — why warm-up works, what a warm-up pool actually does — read How to Warm Up an Email Mailbox for Cold Outreach and come back.

{"h2":"Do this before day 1"}

{"h2":"Before day 1: the checklist that makes the schedule work"}

A warmup schedule cannot fix a broken setup. If any of these are missing, the 30 days are wasted:

{"ul":"SPF, DKIM, and DMARC published and passing on the sending domain. Start with DMARC at p=none. See the [plain-English setup guide.","The domain has existed for at least 7-14 days with DNS resolving and a real website on it. Brand-new domains registered yesterday get treated as throwaways.","A full mailbox profile: real first and last name, photo, signature with a phone number and physical address, and a working reply-to.","A separate sending domain if you are protecting your main one. Most teams should — here's the trade-off.","Tracking pixels off, or at least off for the first two weeks. Open tracking adds an image call to a redirect domain nobody has ever seen before. It is the single easiest thing to remove.","One-click unsubscribe (List-Unsubscribe headers) enabled from your very first cold email, not added later."]}

{"h2":"The 30-day schedule"}

Volumes below are per mailbox, per weekday. "Warm-up sends" are conversational emails inside your warm-up network or between your own team's real inboxes, expected to be opened, replied to, and moved out of Promotions. "Cold sends" are campaign emails to prospects.

{"table":{"headers":["Days","Warm-up sends/day","Cold sends/day","What matters this stage"],"rows":[["1-2","2-4","0","First traffic ever from this mailbox. Reply to every warm-up thread manually if you can."],["3-4","5-6","0","Vary send times across a 6-8 hour window. No identical subject lines."],["5-7","8-10","0","Confirm DKIM passes on received copies. Check Postmaster Tools is collecting data."],["8-10","10-12","5","First cold sends. Pick your 5 best-fit, most likely-to-reply leads. No links."],["11-14","12","8-10","Add one link maximum. Watch bounce rate — it must stay under 2%."],["15-17","12","15","First real read on reply rate. Under 2% replies means fix copy before adding volume."],["18-21","10","20","Enable follow-up steps in the sequence. Follow-ups count toward daily volume."],["22-25","8","25","Run an inbox-placement test. If you're landing in Promotions or spam, hold volume here."],["26-30","5-8","30","Settle at your ceiling. Keep a small warm-up trickle running permanently."]]}}

Weekends: run warm-up sends at roughly a third of your weekday number, or skip them. Do not send cold email on Saturday and Sunday during the ramp. A mailbox that sends 30 emails every single day including holidays looks like software, because it is.

Pacing inside the day matters as much as the daily total. Spread sends across 8-10 hours with randomized gaps of 8-25 minutes. Thirty emails fired in six minutes at 9:00am is a burst pattern; thirty emails trickled from 8:30am to 5:00pm is a person.

{"h2":"Where you land on day 30"}

Thirty to forty cold emails per mailbox per weekday is a sustainable ceiling for a Google Workspace or Microsoft 365 mailbox on a dedicated sending domain. Some teams run 50 and get away with it for a while. Almost nobody runs 100 per mailbox without watching reply rates collapse within a month. If you need more volume, add mailboxes — do not raise the per-mailbox number. We covered the math in How many cold emails can you send per mailbox per day.

And plan the ramp for each new mailbox separately. Adding five mailboxes at once and starting them all at day 1 on the same domain produces a domain-level volume spike even though each mailbox looks tame. Stagger cohorts: two mailboxes this week, two the following week, two the week after.

{"h2":"What warm-up sends should actually look like"}

The single biggest difference between warm-up that helps and warm-up that does nothing is whether the traffic looks like correspondence.

{"ul":["Multi-turn threads, not one-shot blasts. A reply on an existing thread carries more weight than a fresh email.","Plain text or very light HTML. No images, no tracking, no footers full of links.","Human subject lines: \"quick question re Thursday\", \"notes from the call\". Not \"Warmup email #14\".","Real mailbox actions on the receiving side: opened, replied, marked important, dragged out of Promotions or spam into the primary inbox, labeled.","Some variety in recipients — a mix of Gmail, Outlook, and custom-domain destinations, not 100% Gmail-to-Gmail."]}

Warm-up should also never fully stop. A mailbox that sends 30 cold emails and receives almost nothing back has a terrible send-to-engagement ratio. Keeping 5-8 conversational sends a day running indefinitely costs you nothing and keeps that ratio survivable, especially in the weeks when your campaign reply rate dips.

{"h2":"The numbers that override the schedule"}

The table above is a plan, not an instruction. These thresholds beat the plan every time. Check them at the end of each week.

{"table":{"headers":["Signal","Healthy","Hold volume","Stop and diagnose"],"rows":[["Hard bounce rate","Under 1%","1-3%","Over 3%"],["Reply rate (cold)","3%+","1-3%","Under 1% after 100+ sends"],["Spam complaints","Under 0.05%","0.05-0.1%","Over 0.3% (Google's red line)"],["Postmaster domain reputation","High/Medium","Low","Bad"],["Seed inbox placement","80%+ primary","50-80% primary","Under 50% or any spam hits"]]}}

If you hit a "hold" column, freeze at your current volume for a full week instead of advancing to the next row. If you hit a "stop" column, pause cold sending entirely, keep warm-up running, and work through the causes — bad list hygiene, a burned domain, spam-triggering copy. Why are my cold emails going to spam walks through the usual suspects in order.

{"h2":"Adjustments for common situations"}

{"ul":["Brand-new domain (registered this month): add 7 days of warm-up-only at the front. Start cold sends on day 15, not day 8, and cap at 20/day by day 30.","Aged domain with real business history: you can compress week 1 into 3-4 days, but do not skip it. The mailbox is still new even if the domain isn't.","Mailbox that previously got flagged: treat it as worse than new. Warm-up only for 14 days, cold sends capped at 10/day for two weeks after that, and expect a lower ceiling permanently.","SMTP/IMAP mailbox on a shared host: your IP reputation is not yours. Ramp slower, cap lower, and test placement more often.","Agency running client domains: ramp each client's mailboxes independently. Your good sending history does not transfer to their domain."]}

{"h2":"Mistakes that undo a clean ramp"}

{"ul":["Turning on follow-up steps mid-ramp without counting them. A 4-step sequence at 25 new leads a day can push actual sends past 60.","Uploading an unverified list on day 8. A 6% bounce rate on your first campaign erases three weeks of careful warm-up in an afternoon.","Adding open tracking back in once things look fine. It rarely helps and reliably costs placement.","Sending the same 900-character HTML template to everyone. Similarity across recipients is a filter signal on its own.","Ramping on volume alone while ignoring replies. Volume without engagement is exactly the pattern filters are built to catch."]}

{"h2":"Running this automatically"}

Executing this schedule by hand across six mailboxes is tedious enough that most people stop checking the numbers by week three. That is what Sendvanta automates: gradual warm-up using real labeled messages inside your own Google, Microsoft 365, or SMTP workspace, per-mailbox daily caps that step up on a schedule you control, and layered deliverability health scoring across mailbox, domain, DKIM, and IP that automatically slows down or pauses a mailbox when a threshold is crossed instead of after you notice. Drafts get SpamAssassin-scored before send, seed-mailbox tests report actual inbox placement, and engagement classification filters out security-scanner clicks so your reply numbers mean something. Connecting your own mailboxes is free on every plan, including the free tier ($0, 1,000 active leads, 3,000 emails/month, up to 10 SMTP mailboxes) — enough to run a full 30-day ramp before you pay anything. See features for the details.

{"faq":[{"q":"Can I shorten the 30 days if I'm in a hurry?","a":"You can compress week 1 to 3-4 days on an established domain, which gets you to roughly a 24-day ramp. Beyond that you're trading a permanently lower sending ceiling for a few days of speed. It is almost never worth it."},{"q":"Do I need a paid warm-up tool, or can I just email colleagues?","a":"Manual works for one or two mailboxes if you genuinely reply to threads and move messages into the primary inbox. Past three mailboxes the labeling and reply discipline breaks down, and that discipline is the part that matters."},{"q":"Should I keep warm-up running after day 30?","a":"Yes, at a low level — 5-8 conversational sends a day. It keeps your send-to-reply ratio reasonable during weeks when campaign replies are slow and gives you a live signal if placement starts drifting."},{"q":"What if my reply rate is zero after two weeks of cold sending?","a":"Stop adding volume. Zero replies across 100+ sends is either a placement problem or a targeting problem, and more sending makes both worse. Run a seed inbox-placement test first to find out which."},{"q":"Does warm-up help if my domain already has a bad reputation?","a":"Partially and slowly. A domain with a bad Postmaster reputation can take months to recover, and often the cheaper fix is a new sending domain with a clean 30-day ramp."}]}

Schedule and thresholds reflect Google and Microsoft sender guidance and observed placement behavior as of February 2026; recheck Postmaster Tools thresholds before treating any number here as fixed.