The Reddit thread that asks it best: “How I send 3,700+ cold emails per day (100,000+ per month).” That volume works, but only with infrastructure that most teams do not have: dozens of domains, hundreds of mailboxes, and a sending discipline that most people underestimate.
For everyone else, here is the honest math for cold email volume in 2026: how many per mailbox, how many per domain, how to ramp safely, and the warning signs that mean you are sending too much.
The per-mailbox ceiling: 50 emails per day
The widely accepted safe operating number for a warmed mailbox is around 50 cold emails per day. Below that, you are not maximizing infrastructure. Above that, the probability of a spam flag rises sharply.
The ceiling is not the platform’s limit. Instantly, Smartlead, and other senders will happily queue more. The ceiling is what the mailbox providers (Google, Microsoft) tolerate before their reputation systems flag the domain.
The per-domain cap: 2 to 3 mailboxes
Two to three mailboxes per domain is the sweet spot that practitioners converge on. One per domain is even safer, just more expensive. Beyond three or four on a single fresh domain, you are concentrating risk: if that domain gets flagged, every mailbox on it burns at once.
The math: if you need to send 1,500 cold emails a day, the calculation is:
1,500 sends ÷ 50 per mailbox = 30 mailboxes ÷ 3 per domain = about 10 domains.
That is the scale where teams get burned. Most teams need 100 to 300 emails a day, which is 2 to 6 mailboxes across 2 to 3 domains.
The ramp schedule
| Week | Sends per mailbox per day | What is happening |
|---|---|---|
| 1-3 | 0 (warmup only) | Reputation baseline building, no campaigns |
| 4 | 20-30 | Campaigns start, watched daily |
| 5 | 30-50 | Ramping if metrics are clean |
| 6+ | 50 | Full volume, monitored |
Warning signs that you are ramping too fast:
- Bounces climbing above 3 percent
- Open rates dropping across all mailboxes on the domain
- Replies landing in the recipient’s spam folder
- Your sending tool’s health score dropping
Any of these means: stop ramping, hold current volume, and investigate before continuing.
The volume math for your target
Work backwards from how many meetings you need:
| Daily target | Mailboxes (at 50/day) | Domains (at 3 per) | Safest (1 per domain) |
|---|---|---|---|
| 100/day | 2 | 1-2 | 2 |
| 300/day | 6 | 2-3 | 6 |
| 500/day | 10 | 4-5 | 10 |
| 1,000/day | 20 | 7-10 | 20 |
The “safest” column is for teams that have been burned before: one mailbox per domain contains the damage if something goes wrong.
The infrastructure cost
At 300 emails a day (6 mailboxes, 3 domains):
- Domains: 3 × $10-20/year = $30-60/year
- Mailboxes: 6 × $3-10/month = $18-60/month
- Sending tool: included in Instantly or Smartlead subscription ($30-100/month)
- Verification: $10-30/month depending on list size
Total: roughly $80 to $200 a month in infrastructure. That is enough to send 300 emails a day from a verified list with a real CRM and a working reply-detection loop.
The mistake hierarchy (from most to least common)
- One domain, many mailboxes. A reputation event on that domain kills all of them.
- Skipping the ramp. Going from 0 to 50 per mailbox in a week is the fastest way to get flagged.
- Using the main company domain. Never send cold campaigns from your primary domain.
- Unverified lists. Bouncing emails damage reputation faster than any other factor.
- Volume spikes. Going from 100 to 500 per day in a week reads as a spam pattern to providers.
If you fix these five, the volume question answers itself: as many as your infrastructure supports, ramped gradually, monitored daily, and never above the per-mailbox ceiling.
The bottom line
The sending math is boring and the discipline is the product. Fifty per mailbox, two to three per domain, ramped over weeks, monitored daily. When the floors are clean, you can add volume safely. When they are not, you are burning infrastructure that takes weeks to rebuild.
The sending volume is one piece of the infrastructure we build at KomsGro’s outbound marketing service. The mailbox math is in how many mailboxes per domain, the warmup timeline is in how long to warm up an email domain, and the “does warmup even work” debate is in does email warmup still work.
Common questions
How many cold emails can I send per day? The safe ceiling is 50 per mailbox per day after warmup. With 3 mailboxes, that is 150 a day. With 10 mailboxes across 4 domains, that is 500 a day.
What happens if I send too many cold emails? The domain’s reputation drops, emails land in spam, and recovery takes weeks. The first sign is a drop in open rates across all mailboxes on the domain.
How many cold emails does it take to book a meeting? At a 3 to 5 percent reply rate and a 25 percent positive rate, roughly 130 to 260 sends per meeting. A better-targeted list of 100 can book more meetings than a generic list of 1,000.
Can I send cold emails on weekends? Yes, but reply rates are consistently lower. Weekday mornings (Tuesday through Thursday) still produce the best results across the industry benchmarks.
Does the sending limit change with Sales Navigator? Sales Navigator does not affect email sending limits. It raises LinkedIn search and connection limits, which are separate systems.
The ramp schedule in detail (what “gradual” actually means)
The word “gradual” appears in every guide, but almost nobody says what it means. Here is the specific ramp schedule that works, based on running this for clients:
- Week 4 (day 22-28): start campaigns at 20 to 30 per mailbox per day. This is the first real volume after warmup. Watch bounces and spam placement daily.
- Week 5 (day 29-35): increase to 35 to 40 per mailbox if metrics are clean. No bounces, no spam flags, reply rate stable.
- Week 6 (day 36-42): increase to the 50 ceiling. This is full volume for that mailbox.
- Adding capacity: new mailboxes go on NEW domains, each with their own warmup cycle. Never add sends to an existing mailbox beyond 50.
The ramp is not a suggestion. It is the process that tells mailbox providers “this sender is building a legitimate pattern”. Every shortcut resets the clock.
The infrastructure setup (what needs to be in place before the math works)
The 50-per-mailbox ceiling assumes a properly configured infrastructure. Here is what that means in practice, node by node:
Domains. Buy secondary domains that look like your brand (komsgro-outbound.com, not k0m5gr0-biz.net). Register for 2+ years where possible. Set up a redirect to your main site. Never use your primary company domain for cold campaigns.
Mailboxes. Create 2 to 3 mailboxes per domain. Each mailbox has a profile photo, a display name, and a signature. Configure each mailbox separately in your sending tool.
Authentication. SPF, DKIM, and DMARC records set up and verified on every domain. This is the single most important technical step, and it takes 30 minutes. A missing DMARC record undermines everything else.
Warmup. 2 to 3 weeks per mailbox before campaigns start, through the sending tool’s native warmup or a standalone tool. The warmup builds the reputation baseline that the campaigns rely on.
Sending tool. Instantly, Smartlead, or a similar platform that manages rotation, warmup, and reply detection across mailboxes. The sending tool is the backbone; do not try to send from a CRM or a spreadsheet.
List verification. Every list verified before it is loaded. Bounce rates above 3 percent damage reputation faster than any other factor. Verification costs $0.001 to $0.005 per contact and is the cheapest insurance in outbound.
Monitoring. Check bounces, opens, replies, and spam placement daily. The floors are: bounces under 3 percent, spam placement near zero, reply rate at 3 percent or above. If any floor breaks, stop ramping and investigate.
All of this takes an afternoon to set up and a few hours a month to maintain. The discipline is the product.
What to do when the metrics break (the emergency playbook)
Every sending operation hits a bad week eventually. The difference between a warning and a burned domain is how fast you respond. Here is the playbook, in the order the problems usually appear:
Bounces spike above 3 percent. Stop the campaign immediately, do not just slow it down. Re-verify the entire list that is currently in flight. If the verification pass was older than a month, the list aged: people change jobs, domains expire, catch-all addresses stop accepting. Cut the bad rows, then restart at half volume for three days before returning to full.
Open rates drop across every mailbox on a domain. This is the classic pre-flag signal. Reduce that domain’s volume by half, check spam placement with a seed test, and confirm authentication records are still passing (DNS changes and expiring records happen more often than people think). If seed tests land in spam, pause campaigns on that domain for a week and let warmup-only traffic run.
Replies arrive but land in the recipient’s spam folder. Your domain’s reputation is degrading. The fix is not more volume; it is less. Cut to warmup-only, fix any authentication gaps, and rebuild for two weeks.
A single mailbox gets flagged. Disconnect it from the sending tool entirely. Do not keep it in rotation at lower volume; a flagged mailbox in the pool drags the domain’s collective reputation. Replace it with a new mailbox that starts its own warmup cycle.
The through-line in every case: the response to bad metrics is always to send less, never to send more. The teams that burn domains are the ones that try to push through a bad week to hit their volume target. The volume target is not the goal; meetings are the goal, and meetings require a domain that is alive.