How Many Cold Emails Per Day Per Mailbox?
Send 20–30 cold emails per day per mailbox on a fully warmed, dedicated sending domain. For the first two to three weeks of a new mailbox's life, send 10–15 per day. The technical ceilings are far higher (Google Workspace documents about 2,000 messages per day per user, Microsoft 365 about 10,000 recipients per day), but those are platform abuse limits, not deliverability-safe limits. Inbox placement, not the provider's quota, is what caps your real volume.
The rest of this article covers where the 20–30 number comes from, how to ramp to it, how to tell when you've crossed the line, and how to work out exactly how many mailboxes your target volume requires.
How many cold emails per day per mailbox is actually safe?
A practical range, assuming a dedicated sending domain (not your main company domain) and a mailbox that has been warming for at least three weeks:
- Days 1–7 (new mailbox): 0 cold emails. Warmup only.
- Weeks 2–3: 5–15 cold emails per day, raised in small increments.
- Week 4 onward: 20–30 cold emails per day, steady state.
- Aggressive ceiling: 40–50 per day, only if reply rates are healthy and spam complaints are near zero. Most senders who try this see placement degrade within a month.
Two things make that range move. Engagement moves it up: a mailbox that gets genuine replies, forwards, and "not spam" actions builds a reputation that tolerates more volume. Anything that looks automated moves it down: identical message bodies, bursts of sends in a five-minute window, high bounce rates, or a domain registered last Tuesday.
If you are sending to a list you scraped without verification, 20 per day is still too many. Volume is the last variable to tune, not the first.
Why not send 500 per day if Google allows 2,000?
Because the provider's published limit and the receiving side's tolerance are two different systems.
Google Workspace's roughly 2,000 messages/day limit (lower on trial accounts, and applied on a rolling 24-hour window rather than resetting at midnight) is an anti-abuse throttle on your own account. Hitting it gets your sending suspended for 24 hours. It says nothing about whether the recipient's Gmail, Outlook, or corporate filter will place your message in the inbox.
Inbox placement is decided by the receiving mail system, using signals like:
- Spam complaint rate. Google's Postmaster Tools guidance treats 0.1% as the level to stay under and 0.3% as the point where filtering visibly worsens. At 30 emails/day, a single complaint in a week already puts you near 0.5%. Volume amplifies the complaint math brutally.
- Bounce rate. Sustained hard bounces above ~3% signal list scraping. Higher volume on a bad list gets you there faster.
- Sending pattern. A mailbox that sends 20 messages spread across eight business hours looks human. One that sends 200 in ninety minutes does not.
- Domain and IP reputation history. New domains have no history, so early volume carries disproportionate weight.
Microsoft tightened this further in 2025. Senders above 5,000 messages/day to consumer Outlook domains now need SPF, DKIM, and DMARC alignment, which mirrors the bulk-sender requirements Google and Yahoo introduced in 2024. Authentication is table stakes rather than an optimization.
What is a realistic warmup and ramp schedule?
Warmup and cold sending are separate activities that share one quota. A workable four-week ramp for a brand-new mailbox on a brand-new domain:
- Week 1: Warmup only, ~10–20 warmup messages/day. Zero cold sends. Let SPF, DKIM, and DMARC records propagate and the domain age past its first week.
- Week 2: Warmup ~20–30/day, plus 5–10 cold emails/day.
- Week 3: Warmup ~30–40/day, plus 10–20 cold emails/day.
- Week 4+: Warmup dialed back to ~20–30/day as a maintenance baseline, plus 20–30 cold emails/day.
Two rules matter more than the exact numbers. First, increase gradually. A jump from 5 to 50 is a stronger negative signal than a high absolute number. Second, never stop warmup entirely. A mailbox running only cold outreach has an engagement profile of nothing but unanswered messages to strangers, which is exactly what spam looks like.
One note on responsibility here: Inboxlogy provisions the mailboxes and handles the DNS and authentication side, but warmup itself runs inside your connected sending tool, whether that's Instantly, Smartlead, or ReachInbox. You configure the warmup schedule there, and the mailbox connects to it over standard OAuth or SMTP.
Does warmup count toward the daily limit?
Yes. Warmup messages are ordinary outbound mail sent from your mailbox, so they consume the same provider quota and shape the same reputation. A mailbox running 30 warmup messages and 30 cold emails is sending 60 messages a day, not 30.
This trips up a lot of people who set warmup to an aggressive setting and then wonder why a "20/day" campaign is throttling. When you plan capacity, add the two together and keep the combined total comfortably under the platform ceiling. In practice, well under 100/day per mailbox is plenty conservative.
How many mailboxes do I need for my target volume?
The arithmetic is simple, and it's the calculation most people actually came here for:
Mailboxes needed = (target cold emails per day) ÷ (safe sends per mailbox per day)
At a conservative 20 sends per mailbox per day:
- 200 emails/day → 10 mailboxes
- 500 emails/day → 25 mailboxes
- 1,000 emails/day → 50 mailboxes
- 2,000 emails/day → 100 mailboxes
Then add a working-days adjustment. Cold email should go out on business days, so ~22 sending days per month. At 1,000/day across 50 mailboxes, that's roughly 22,000 sends per month, not 30,000.
Cost follows directly. Inboxlogy mailboxes start at $2.80/mailbox/month with $0 setup, billed monthly, so 50 mailboxes for a 1,000/day program is about $140/month in infrastructure, before your sending tool, data, and verification costs. Running the same volume at 5 emails per mailbox per day would need 200 mailboxes and cost four times as much, which is why the safe-volume number is a budget decision as much as a deliverability one.
How many mailboxes per domain?
Two to three sending mailboxes per domain is the standard, and it's a good one. Domain reputation is shared: if one mailbox on a domain generates complaints, every mailbox on that domain inherits the damage.
So a 1,000/day program needs roughly 50 mailboxes across ~17–25 domains, at 20 sends per mailbox. That gives you about 40–60 sends per day per domain, which is a sane per-domain load, and it means a single burned domain costs you 4–6% of capacity rather than everything.
Buy domains that are plausible variants of your brand (getacme.com, acme-hq.com), set up redirects to your real site, and keep your primary corporate domain entirely out of cold outreach. Your primary domain carries your invoices, contracts, and support mail. It is not worth risking for prospecting.
How do I know if I'm sending too much?
Watch leading indicators, not revenue. By the time pipeline dips, the reputation damage is weeks old.
- Reply rate falling while everything else is constant. The single most reliable early warning. If copy, list, and offer are unchanged but replies drop by half, you are landing in spam.
- Google Postmaster Tools. Free, and the only direct read on how Gmail sees your domain. Watch spam rate (target well under 0.1%) and domain reputation. It requires a DNS TXT record and meaningful volume to populate.
- Microsoft SNDS and JMRP. Free equivalents for the Outlook side. JMRP gives you complaint feedback directly.
- Bounce rate over 3%. Usually a list problem, but it accelerates reputation loss. Verify every list before loading it.
- Seed testing. Send to a small set of real mailboxes you control across Gmail, Outlook, and a corporate Microsoft 365 tenant, and check placement weekly. Imperfect, but it catches a spam-folder slide fast.
When any of these turn, cut volume per mailbox by half and hold for two weeks before doing anything else. Adding mailboxes on top of a reputation problem just spreads it.
Do Google Workspace and Microsoft 365 mailboxes have different safe limits?
The safe cold-sending range is similar, roughly 20–30/day for both, but the failure modes differ.
Google Workspace enforces its quota hard and fast: cross it and sending stops for 24 hours. Gmail's filtering is also more engagement-driven, so reply rate influences placement heavily. Google Postmaster Tools gives you real visibility.
Microsoft 365 has higher documented limits but a stricter per-minute throttle (around 30 messages/minute), so burst sending fails before daily volume does. Microsoft's filtering leans harder on authentication, content signals, and IP/domain reputation, and gives you less visibility. You often find out from silence rather than a dashboard.
Most serious programs run both, because a meaningful share of B2B prospects sit on Microsoft 365 and inbox placement is better when the sending platform matches the receiving one. Inboxlogy provisions both authorized Google Workspace and Microsoft 365 mailboxes on dedicated US or EU IPs, with SPF, DKIM, and DMARC configured automatically at setup, so the authentication layer isn't something you hand-configure per domain.
What breaks deliverability before volume does?
Honest answer: for most senders, volume is not the actual problem. In rough order of impact:
- List quality. Unverified, poorly targeted lists produce bounces and complaints that no volume discipline can offset.
- Relevance of the offer. Complaints come from people who cannot understand why you emailed them. Tight targeting is a deliverability control.
- Authentication and DNS. Missing DKIM or a DMARC policy that doesn't align will hurt you at any volume.
- Copy patterns. Link-heavy messages, tracking pixels on every send, image-heavy HTML, and identical bodies across thousands of recipients all raise filter scores. Plain-text-style messages with at most one link perform better.
- Volume and pacing. Real, but it's the amplifier. It multiplies whatever the above is already doing.
If reply rates are poor, cutting from 30 to 15 per mailbox will not fix it. Fix the list and the message first. Volume tuning is what you do once the fundamentals work.
What should I do with a mailbox that's already burned?
Say a mailbox or domain has clearly lost placement: replies near zero, seed tests land in spam, Postmaster reports a bad domain reputation. Here's the sequence.
- Stop cold sending on it entirely. Not reduce, stop.
- Keep warmup running in your sending tool for three to four weeks. Recovery is possible but slow, and it depends on rebuilding positive engagement.
- Restart at 5–10/day and ramp more slowly than the first time.
- Retire the domain if it doesn't recover. A burned domain is a sunk cost of a few dollars. Weeks spent nursing it are not. This is the argument for many cheap domains and mailboxes over a few precious ones.
One practical caution when replacing mailboxes: make sure you actually own them. Some providers resell mailboxes on tenants they control, which means you cannot export data, cannot audit configuration, and lose everything if the provider disappears or the tenant gets flagged for someone else's behavior. Inboxlogy gives you 100% ownership and admin access to every mailbox, plus a full API for provisioning and teardown, so replacing a burned domain's mailboxes is a scripted operation rather than a support ticket.
Frequently asked questions
Can I send 50 cold emails a day from one mailbox?
Technically yes, and some senders sustain it on aged domains with strong reply rates. But it roughly doubles your exposure per mailbox, and most accounts that start at 50/day on a new domain see placement degrade within four to six weeks. Reaching 50 is safer than starting there: get to 20–30, run it for a month, verify placement with seed tests, then increase in steps of 5.
How long does a new mailbox need to warm up before cold sending?
Two to three weeks minimum before you reach normal volume, with the first seven days as warmup only. Domain age matters as much as mailbox age, so buy sending domains a few weeks before you plan to launch.
Should I use my main company domain for cold email?
No. Use dedicated lookalike domains that redirect to your main site. Cold outreach carries real reputation risk, and your primary domain handles mail you cannot afford to have filtered: invoices, contracts, support, recruiting.
Is it better to have more mailboxes at low volume, or fewer at high volume?
More mailboxes at lower volume, almost always. It distributes risk, keeps each mailbox's pattern closer to human, and lets you lose a domain without losing capacity. At $2.80/mailbox/month, the cost difference between 25 mailboxes at 40/day and 50 mailboxes at 20/day is about $70/month, far less than the cost of rebuilding a burned sending program.