Cold emails per inbox per day, warmup and sending limits that protect your domain

Mailbox providers judge reputation at several levels at once: the sending IP, the domain in the From address, the domain in the DKIM signature and, increasingly, the individual mailbox and the engagement it receives. When one inbox suddenly sends hundreds of messages to strangers, the pattern looks nothing like a person at work, and filters react to patterns. A person writing to new contacts sends a few dozen messages spread across the working day, gets some replies, and almost never hits a dead address. Your goal is to keep every sending inbox inside that human looking pattern, and to reach the total volume you need by adding inboxes rather than by pushing each one harder.

That framing changes the planning question. Instead of asking how many emails your team can send in a week, you ask how many inboxes you need to reach that number at a conservative daily rate. The answer is usually more inboxes and lower volume per inbox, which is cheaper than a burned domain and far cheaper than a quarter of meetings that never happen because your messages land in spam.

Most experienced outbound teams work within broad ranges rather than a fixed rule. The figures below describe common practice across the industry, not a guarantee from any provider, and you should treat the upper end as a ceiling rather than a target.

• Brand new inbox in its first two weeks: warmup traffic only, no cold sends or a handful at most.
• Inbox in weeks three and four: roughly 10 to 20 cold emails a day, rising slowly.
• Established, healthy inbox: roughly 30 to 50 cold emails a day is the range many teams settle on.
• Above about 50 new contacts a day per inbox, most practitioners add another inbox instead of raising the cap.

Two details are easy to miss. First, the cap usually counts every email that goes to a new or cold contact, including follow-ups, so a five step sequence that reaches 40 people a day on step one will eventually carry well over 40 sends a day once follow-ups stack up. Second, replies you send by hand to interested prospects do not need to fit inside the cold cap, but they still come from the same inbox, which is one more reason to keep the cold volume modest.

Start from the number of new contacts you want to reach per month and work backwards. Suppose you want to start conversations with 1 500 new contacts a month. With roughly 21 working days, that is about 72 new contacts per working day. If each contact receives a four step sequence spread over two weeks, the daily send load in steady state is close to four times the daily new contact count, so roughly 285 sends a day once the pipeline is full.

At a cap of 30 sends per inbox per day, 285 sends need about 10 inboxes. At a cap of 40, you need about 8. At a cap of 50, about 6. Run the numbers with your own sequence length, your own working days and your own cap, and you will see why a team that needs serious volume ends up with several sending inboxes across two or three domains. The arithmetic also shows the cost of overly long sequences: every extra step multiplies daily load for the same number of new conversations.

Your primary domain carries your invoices, your support mail, your password resets and the replies your customers expect. If cold outreach damages its reputation, those messages start landing in spam too. The standard practice is to register one or more closely related domains for outbound, for example a variation of your brand name with a different suffix or a short prefix, and redirect their website traffic to your main site so anyone who checks the domain lands somewhere legitimate.

• Keep the brand recognisable so recipients understand who is writing.
• Give each outbound domain a real website redirect, an MX record and a working inbox that receives replies.
• Spread inboxes across two or three domains so a problem on one domain does not stop all sending.
• Keep each domain to a modest number of inboxes, many teams use two to five per domain.

Separate domains are a reputation firewall, not a disguise. Every message still identifies your business, carries a physical address and offers a clear way to opt out.

Authentication tells receiving servers that a message really comes from the domain it claims. Large mailbox providers now expect bulk senders to authenticate properly, and unauthenticated cold mail is one of the fastest routes to the spam folder.

• SPF lists the servers allowed to send for your domain. Include your inbox provider and nothing you do not use, and stay within the limit on DNS lookups.
• DKIM signs each message with a key published in your DNS, so the receiver can verify the message was not altered. Use the signing set up by your inbox provider and make sure the signing domain matches your From domain.
• DMARC tells receivers what to do when SPF or DKIM fails and where to send reports. A common starting policy is p=none with a reporting address, moving to quarantine once you have confirmed that all legitimate mail passes.
• Add a custom tracking domain if you track clicks, so links do not point to a shared domain with someone else's reputation. Many teams skip open and click tracking entirely on first touch emails to keep messages plain.

Check all three records after every DNS change. A typo in an SPF record or a DKIM key that was never published can quietly undo weeks of careful warmup.

A new inbox has no history, and no history looks suspicious. Warmup builds a track record by sending a gradually increasing number of ordinary messages that get opened, replied to and, when they land in spam, moved back to the inbox. The provider sees normal, engaged traffic from that mailbox before any cold email goes out.

Warmup is not a shield that lets you send unlimited volume later. It establishes a baseline, and that baseline still depends on what happens once real prospects receive your messages. If your list is poor or your copy reads like a mass mailing, reputation drops no matter how long the inbox was warmed. Most teams keep warmup running in the background at a low level even after cold sending starts, so the inbox keeps a steady stream of positive engagement.

The exact schedule depends on the age of the domain and the inbox provider, but a typical ramp for a fresh inbox on a new outbound domain looks like this. Treat it as a starting point and slow down whenever metrics look off.

Raise the cold cap in small steps, for example by a few emails a day each week, and only when bounces stay low and replies keep arriving. If anything looks wrong, drop back to the previous step for a week rather than pushing through.

Hard bounces are one of the clearest signals a mailbox provider sees. Repeatedly sending to addresses that do not exist tells the receiver your list was not built carefully. A widely used rule of thumb is to keep hard bounces under about 2 percent of sends, and many teams aim well below that. Above that level, pause and fix the list before sending more.

• Verify addresses before they enter a sequence, and remove any that come back invalid or risky.
• Stop the sequence for a contact immediately after a hard bounce, and add the address to your suppression list.
• Be careful with catch-all domains, which accept everything at first and bounce later.
• Refresh data regularly, because people change jobs and work addresses go dead.

Complaints matter even more than bounces. Large providers publish guidance that spam complaint rates should stay well below 0.3 percent, with about 0.1 percent as a sensible working ceiling. A clear unsubscribe link, a one click unsubscribe header and a relevant message are the best protection.

Inbox rotation means a campaign sends from a pool of inboxes, with each new contact assigned to one sender who then handles the whole thread. Rotation keeps every inbox under its cap, spreads risk and lets you pause one inbox without stopping the campaign. Two rules keep it clean: a contact always hears from the same sender across all steps, and replies route back to the person who owns the conversation.

Rotation also makes problems visible. When one inbox shows higher bounces or fewer replies than its peers on the same campaign, you know the issue is that inbox or its domain, not your message, and you can rest it while the others continue.

Volume is only one part of the pattern. Timing and content matter as well. Messages that go out in bursts at the same second, at three in the morning in the recipient's time zone, with identical text and heavy links, look automated because they are automated in the worst sense.

• Spread sends across business hours in the recipient's time zone, with random gaps between messages.
• Keep first touch emails short, in plain text, with at most one link and no attachments.
• Personalise the first line with something real about the company, so each message is different.
• Avoid image heavy templates, link shorteners and spam trigger phrasing.

You cannot see your reputation directly, but you can watch the signals that reflect it. Review them weekly per inbox and per domain rather than only per campaign.

• Hard bounce rate per inbox, with a clear threshold that pauses sending.
• Reply rate per inbox compared with other inboxes on the same campaign.
• Spam complaints and unsubscribes, which tell you about list fit and message relevance.
• DMARC aggregate reports, which show whether anything is sending as your domain without authentication.
• Provider tools for postmasters, where available for your volume, for domain reputation trends.

When signals drop, reduce volume first, then look at list quality and copy. Raising volume to make up for lower reply rates almost always makes things worse.

If an inbox starts landing in spam, stop cold sending from it, keep warmup running, and give it one to two weeks before trying a small volume again. Check authentication records, look at recent bounces and remove the lists that caused them. If a whole domain is affected, move new campaigns to your other outbound domains while it recovers. This is exactly why the multi domain, multi inbox setup exists: a problem in one place slows you down but does not stop your pipeline.

• One or more separate outbound domains, each with a website redirect and a working inbox.
• SPF, DKIM and DMARC published and verified on every sending domain.
• Each inbox warmed for at least two to four weeks before meaningful cold volume.
• A per inbox daily cap, typically in the 30 to 50 range for healthy inboxes.
• Verified lists, automatic stop on bounce and a single suppression list across campaigns.
• Unsubscribe link and header in every campaign email, and your business address in the footer.
• Weekly review of bounces, replies and complaints per inbox.