The safe answer is 20 to 50 cold emails per day per fully warmed inbox, and 100 to 150 per sending domain across all its mailboxes. If you are asking how many cold emails per day you can send, that range is the working ceiling most experienced senders operate inside. Anything higher is possible but buys you risk you do not need.
Disclosure: This guide mentions GTM Bud alongside third-party tool categories. We have a bias toward our own product, but the volume guidance here is vendor-neutral. Use whatever fits your stack.
We are not guessing at these numbers. Our outbound agency, Referral Program Pros, has booked over 7,000 meetings for clients, and we built GTM Bud on the playbook that agency runs every day. The volume rules below are what we settled on after watching what happens on both sides of the line: domains that ramped patiently and kept producing for years, and domains that got pushed hard in month one and never recovered.
What is the safe daily cold email limit per inbox?
A fully warmed mailbox on an aged domain can send 20 to 50 cold emails per day indefinitely without reputation trouble, assuming a verified list and plain text copy. Below 20 you are leaving capacity unused. Above 50 the returns flatten fast while the risk climbs, because a single mailbox sending high volume to strangers who mostly do not reply is the exact behavioral signature filters are built to catch. The number that actually matters is not the send count but the engagement ratio behind it. Forty emails a day producing steady replies is a healthy mailbox. Eighty a day producing silence is a mailbox on a countdown. Treat 50 as a hard personal ceiling and spend your effort on adding mailboxes rather than raising it.
Three things move you within that 20 to 50 band:
- Domain age. A domain sending for two years tolerates more than one registered six weeks ago.
- Reply rate. Positive engagement is the strongest counterweight to volume. Campaigns that earn replies can carry more send volume than campaigns that do not.
- List quality. Bounces cost you more than volume does. A clean list at 45 per day is safer than a dirty list at 20.
Volume is not the thing providers punish. Unwanted volume is. Every send you make either adds engagement evidence or subtracts it, and the daily number only decides how quickly you accumulate whichever one you are producing.
Why are Google Workspace and Microsoft 365 limits not cold email limits?
Because published provider limits are anti-abuse ceilings for normal business mail, not safety guidance for cold outreach. Google’s documentation puts a paid Google Workspace account at 2,000 messages per day, with 3,000 external recipients per day and 10,000 total recipients per day. Microsoft’s Exchange Online limits set a recipient rate limit of 10,000 recipients per day per mailbox and a message rate limit of 30 messages per minute. Those numbers describe the point at which the provider stops you at the door. They say nothing about the point at which spam filters, which are a separate system with separate signals, quietly start routing you to the junk folder. Sending 2,000 cold emails from one Workspace mailbox is technically permitted and operationally suicidal.
| Source | Published number | What it actually governs |
|---|---|---|
| Google Workspace (paid account) | 2,000 messages per day | Hard send cap before a 24 hour suspension |
| Google Workspace (trial account) | 500 messages per day | Hard send cap on unpaid trials |
| Google Workspace external recipients | 3,000 per day | Distinct recipients outside your domain |
| Microsoft Exchange Online | 10,000 recipients per day per mailbox | Recipient rate limit on a 24 hour sliding window |
| Microsoft Exchange Online | 30 messages per minute | Throttling of burst submission |
| Gmail sender guidelines | 5,000 messages per day to Gmail defines a bulk sender | Triggers the strict authentication requirements |
| Gmail sender guidelines | Spam complaint rate below 0.3 percent, ideally under 0.1 percent | The one published reputation threshold |
Read the full numbers at Google’s Workspace sending limits documentation, Microsoft’s Exchange Online limits, and Gmail’s email sender guidelines. The gap between the published cap and the safe cold email volume is roughly forty to one, which is why quoting the provider limit as your sending plan is the most common way new senders burn a domain in their first month. The authentication side of this, meaning SPF, DKIM, DMARC, and how reputation is scored, is covered separately in our cold email deliverability guide.
The week-by-week ramp schedule for a brand new inbox
A new mailbox has no sending history, which means providers have no evidence either way and will assume the worst if you behave like a bulk sender on day one. The ramp exists to manufacture that evidence. Here is the schedule we run.
| Week | Warmup emails per day | Cold emails per day | What you are doing |
|---|---|---|---|
| 1 | 5 to 10 | 0 | Establishing that the mailbox exists and sends mail |
| 2 | 15 to 20 | 0 | Building reply and engagement history |
| 3 | 25 to 30 | 5 to 10 | First real sends, smallest and cleanest segment only |
| 4 | 30 to 40 | 10 to 15 | Confirming bounces near zero before adding volume |
| 5 | 40 | 15 to 25 | Steady climb, watching reply rate hold |
| 6 | 40 | 25 to 35 | Approaching working volume |
| 7 | 40 | 35 to 45 | Near ceiling, checking Postmaster reputation weekly |
| 8+ | 30 to 40 ongoing | 40 to 50 | Steady state, warmup stays on permanently |
Two rules make or break this table. First, never skip a week to catch up. If week four shows a bounce spike or a reply collapse, you hold volume flat until the cause is fixed rather than continuing the climb on schedule. Second, keep warmup running forever. Turning warmup off once you hit full volume removes the positive engagement that was offsetting your cold sends in the first place. The mechanics of warmup traffic are covered in our email warmup guide for cold outreach.
How to scale volume the right way
You scale cold email by adding inboxes and domains, never by raising the per-inbox number. This is the single most consequential structural decision in outbound, and it is counterintuitive because the per-inbox lever is the one sitting right in front of you in the software. Raising one mailbox from 50 to 150 per day triples your volume and concentrates all of that new risk on one asset that can be flagged in an afternoon. Adding two more mailboxes achieves the same total at a volume each mailbox can sustain, and if one gets flagged you lose a third of capacity instead of everything. Plan roughly two to three sending mailboxes per domain, and keep every domain a secondary domain so your primary business domain is never exposed.
| Target daily sends | Inboxes needed (at 40/day) | Domains needed (at 3 inboxes each) | Realistic monthly reach |
|---|---|---|---|
| 40 | 1 | 1 | ~880 sends |
| 120 | 3 | 1 | ~2,640 sends |
| 240 | 6 | 2 | ~5,280 sends |
| 400 | 10 | 4 | ~8,800 sends |
| 800 | 20 | 7 | ~17,600 sends |
Monthly figures assume 22 working days and no weekend sending. Note how quickly the infrastructure requirement grows: 800 emails a day is a twenty mailbox operation across seven domains, each one needing its own warmup, monitoring, and rotation. That coordination is the actual work, and it is what a cold email automation tool is for. Comparing your options is easier with our roundup of the best cold email software in 2026.
Working backwards from a meetings target to a daily send number
Most teams pick a daily volume first and hope meetings follow. Reverse it. Start with the meetings you need, apply your own funnel rates, and the required volume falls out as arithmetic instead of a guess. The chain is: sends produce replies, replies produce positive replies, positive replies produce booked meetings. If you have historical numbers use those; if you are starting cold, run a small test batch for two weeks and measure rather than borrowing someone else’s benchmark, because reply rates swing enormously by industry and offer.
The calculation, using your own measured rates:
- Meetings needed per month, divided by your booking rate from positive replies, gives positive replies needed.
- Positive replies needed, divided by your positive reply rate, gives total sends needed per month.
- Total monthly sends divided by 22 working days gives your required daily volume.
- Required daily volume divided by 40 gives the number of warmed inboxes to provision.
Worked through with placeholder numbers, if you need [X] meetings and your funnel converts at [Y] percent, step three usually lands somewhere that makes the infrastructure requirement obvious well before you have built it. That is the point of doing it in this order. Use your own measured reply and booking rates here rather than a borrowed benchmark, because the whole point of the exercise is to size infrastructure against reality. If your math says you need volume you cannot support, the fix is usually improving the conversion rates rather than adding mailboxes. Our guide to reply rate optimization for cold outreach is the cheaper lever in almost every case.
Warning signs you are sending too much
Volume damage announces itself before it becomes permanent, and every one of these signals is visible inside a week if you are watching. Open rate declining against your own baseline is usually the first, because filtered mail is delivered but unseen. A bounce rate climbing past 3 to 5 percent means you are burning through addresses faster than your data can keep up. Spam complaints approaching the 0.3 percent threshold Google publishes are the most urgent signal on the list, because that number is enforced rather than advisory. Reply rate collapsing while nothing about your copy or list changed is the clearest sign of all: your emails stopped reaching people. Any one of these means cut volume immediately and diagnose before resuming.
What to do in each case:
- Open rate slides. Cut per-inbox volume by half, raise the warmup ratio, and confirm placement with seed tests.
- Bounces spike. Stop the campaign and re-verify the list. Bounces are a data problem, not a volume problem, but they compound volume damage.
- Complaints climb. This is targeting, not throughput. Narrow the audience before you touch the send count.
- Replies dry up. Rotate to backup mailboxes, pause the affected ones, and check domain reputation in Postmaster Tools.
The most expensive mistake in cold email is treating a reputation warning as a copy problem. Rewriting subject lines while a mailbox is quietly filtered wastes weeks, because the messages you are testing were never seen.
Why list quality caps your real volume regardless of infrastructure
You can build twenty warmed mailboxes across seven domains and still be capped at a fraction of that capacity by your data, because volume you cannot fill with qualified contacts is volume that damages you. Every send to a stale address, a wrong-fit prospect, or a spam trap subtracts from the reputation your infrastructure spent weeks building. This is why the answer to “how many cold emails per day” is never purely a number. The honest ceiling is the smaller of what your infrastructure can safely carry and what your list can defensibly justify, and for most small teams the list runs out first. Verifying before every send is non-negotiable, and our comparison of the best email verification tools in 2026 covers the options. A tighter list at lower volume consistently outperforms a loose list at high volume, on both reply rate and mailbox lifespan.
Frequently asked questions about cold email sending limits
How many cold emails per day can I send from a brand new inbox?
Zero on day one. A brand new inbox should send only warmup traffic for its first two to three weeks while it builds a sending history. After that, start real campaigns at roughly 5 to 10 emails per day and add about 5 to 10 per day each week until you reach 20 to 50. Mailbox and domain age matter more than the tool you send from, so buying secondary domains early and letting them sit is the cheapest advantage available to a new sender.
Does sending cold emails slower during the day help?
Yes. Spreading sends across working hours with randomized gaps looks more like a person and less like a script. Microsoft applies a message rate limit of 30 messages per minute per mailbox in Exchange Online, and bursts near any rate ceiling are exactly the pattern filters watch for. Sending 40 emails evenly across an eight hour window is safer than sending the same 40 in five minutes, even though the daily total is identical. Most schedulers handle this automatically, including the ones covered in our Instantly alternative comparison.
Should I use one domain with many inboxes or many domains?
Many domains, with a small number of inboxes on each. Two or three sending mailboxes per domain is a common ceiling. Reputation is tracked at the domain level as well as the mailbox level, so twelve inboxes on one domain concentrates all your risk in a single asset. Spread the same twelve across four or five secondary domains and one bad campaign costs you a fraction of capacity instead of all of it.
Do email sending limits apply to follow-ups too?
Yes. Every message leaving the mailbox counts toward your daily total, including second, third, and fourth touches in a sequence. This surprises people who plan volume around new prospects only. A four step sequence means each new prospect eventually consumes four sends, so a mailbox at 40 per day is contacting far fewer new people than 40 once the sequence reaches steady state. Build your volume plan on total sends, not on new contacts.
Is LinkedIn outreach subject to the same daily limits?
No, LinkedIn runs on entirely separate limits and a different failure mode. Connection requests and messages are throttled per account by LinkedIn itself, and the penalty is account restriction rather than a spam folder. Because the two channels have independent ceilings, running LinkedIn alongside email is the most reliable way to raise total daily touches without pushing any single email inbox past a safe volume.
Send the right number and keep your domains for years
The number you came for is 20 to 50 per warmed inbox and 100 to 150 per domain, but the more useful takeaway is the shape of the answer: safe volume is a function of mailbox age, engagement, and list quality, and you grow it horizontally by adding inboxes rather than vertically by pushing one harder. Provider limits are ceilings, not targets. Ramp over eight weeks, keep warmup on permanently, watch open rate and bounces weekly, and work backwards from meetings instead of forwards from a send count.
If you would rather not manage twenty mailboxes and seven domains by hand, GTM Bud runs the ramp, the rotation, and the daily limits for you across both email and LinkedIn, with setup in about 15 minutes and a 7-day free trial that needs no credit card. It is the same volume discipline our agency used to book over 7,000 meetings, applied automatically so you never have to choose between growth and keeping your domains alive.