How Many Sending Domains and Mailboxes Do You Need for Cold Email?

Most people ask this question backwards. They decide they want five domains, buy five domains, put two mailboxes on each, and then wonder why they can only send 300 emails a day. Domains aren't the input — volume is. You work out how many emails a day you actually need to send, convert that to mailboxes, and the domain count falls out of that.

The short answer: 2 to 3 mailboxes per sending domain, and never your main company domain. If you need 300 cold emails a day, that's roughly 10 warmed mailboxes and 3–4 domains. If you need 60 a day, that's 2 mailboxes and a single domain. Anything past 3 mailboxes on one domain concentrates risk you don't need to take.

{"h2":"Start with volume, not domains"}

We already worked the mailbox math in detail in How Many Mailboxes Do You Need to Send 1,000 Cold Emails a Day? — the short version is that a fully warmed mailbox on Google Workspace or Microsoft 365 handles roughly 25–40 cold sends per day including follow-ups, and you should plan at 30 rather than at the ceiling.

Once you have a mailbox count, the domain count is arithmetic. Divide by 3 and round up.

Cost assumes $4–8 per mailbox per month depending on whether you use Google Workspace, Microsoft 365, or managed SMTP mailboxes, plus $10–20 per domain per year amortized. Nothing here is expensive in isolation. What's expensive is buying 30 mailboxes in week one and burning them all before they're warm.

{"h2":"Why 2–3 mailboxes per domain, and not 10"}

Mailbox providers and filters evaluate reputation at multiple layers: IP, domain, and individual sender address. The domain layer is the one that matters most for cold outbound, because it's the layer that follows you everywhere. A single bad mailbox gets its sends throttled. A domain that lands on Spamhaus DBL or picks up a poor Google Postmaster reputation takes every mailbox on it down at once.

So the question is really: how much of my sending capacity am I willing to lose in one event?

• One domain, 10 mailboxes. One blocklist hit removes 100% of your capacity. You're down for the 2–4 weeks it takes to warm replacements.
• Four domains, 2–3 mailboxes each. One blocklist hit removes 25%. You keep sending while you diagnose and replace.
• Ten domains, one mailbox each. Maximum isolation, but you now maintain 10 sets of SPF, DKIM, and DMARC records, 10 renewals, and 10 separate reputations that each build slowly because each one only sends 30 emails a day.

That last point is underrated. Domain reputation builds from consistent, engaged volume. A domain sending 30 emails a day accumulates signal slowly. Two or three mailboxes on the same domain sending 30 each gives that domain 60–90 daily sends of history, which establishes a stable pattern faster. Spreading yourself across too many domains means every domain stays permanently unproven.

Two to three is the balance point: enough isolation that no single failure is fatal, enough shared volume that each domain builds a real track record.

{"h2":"Your main domain is not one of them"}

None of this applies to the domain your invoices, password resets, and customer support emails go out from. Cold outbound generates complaints and unknown-user bounces at rates your transactional mail never will, and that reputation is shared. We covered the full argument in Should You Use Your Main Domain for Cold Email? — the summary is that the upside is marginal and the downside is your billing emails going to spam.

Buy lookalike domains instead. If you're acme.com, register getacme.com, acme.co, tryacme.com, or acmehq.com. Stay on .com, .co, or .io. Avoid the cheap TLDs — .xyz, .icu, .top, .click — because filters weight them heavily and you'll fight an uphill battle you created for yourself to save eight dollars.

{"h2":"Buy in waves, not all at once"}

The instinct when scaling is to buy every domain and mailbox on day one. Don't. Every new domain needs 3–4 weeks of warm-up before it carries real cold volume, and warming 11 domains simultaneously from a standing start creates exactly the pattern filters look for: a cluster of freshly registered domains, all at the same registrar, all with the same DNS provider, all starting to send at the same time.

A better cadence:

• Week 0: register 2–3 domains, set up DNS, create 2–3 mailboxes each, start warm-up.
• Week 4: those domains start carrying cold volume. Register the next wave.
• Week 8: wave one is at full volume, wave two is coming online, wave three is registered.
• Ongoing: keep one spare domain warm and idle. When something breaks, you swap it in the same day instead of waiting a month.

Registering domains 30–60 days before you need them also helps on its own. Domain age is a weak signal, but it's a real one, and the cost of registering early is a few dollars.

{"h2":"When you need more domains than the formula says"}

The 3-mailboxes-per-domain rule assumes one offer going to one audience. A few situations justify more separation:

• Multiple brands or products. If you're pitching two genuinely different things, the from-domain should match the pitch. Prospects who Google the domain should land somewhere coherent.
• Agencies running client campaigns. Every client gets their own domains, registered to a domain that matches their brand. Never mix clients on shared sending infrastructure — one client's bad list shouldn't cost another client their capacity.
• Different regions. If you're sending to both the US and Germany with localized copy, separate domains keep engagement signals from bleeding across audiences with very different reply behavior.
• High-risk experiments. Testing a new list source or an aggressive offer? Run it on an isolated domain first. If it burns, it burns alone.

And when you need fewer: if you're sending under 100 emails a day to a tight, well-researched list, one domain with three mailboxes is genuinely fine. Founders regularly over-build infrastructure for volume they never reach. Two mailboxes booking four meetings a month beats 20 mailboxes blasting a scraped list into a filter.

{"h2":"Every extra domain is extra DNS surface"}

There's an operational cost people forget. Each sending domain needs its own SPF record, its own DKIM key published and verified, its own DMARC policy, a working MX record, and ideally a live landing page and a custom tracking subdomain. That's six things to get right, multiplied by your domain count.

Misconfigured DKIM on domain 7 of 11 is the kind of problem that silently costs you 9% of your inbox placement for two months. Whatever tooling you use, you want automated per-domain checks, not a spreadsheet and good intentions. If you're adding domains faster than you can verify them, you're adding risk, not capacity.

{"h2":"Know when to retire a domain"}

Domains are consumables with a long but finite life. Retire one when:

• It appears on a major blocklist and delisting doesn't hold after 30 days.
• Google Postmaster shows its domain reputation at Low or Bad and it hasn't recovered after four weeks of reduced, clean sending.
• Inbox placement on seed tests drops below ~70% while your other domains on the same list and copy stay above 90%.
• Bounce rate on verified leads climbs above 3% for reasons you can't trace to the list.

A healthy domain that only ever sends to verified leads with honest copy and real unsubscribe handling can run for years. The ones that burn in three months usually burned because of the list, not the domain.

{"h2":"Where Sendvanta fits"}

Sendvanta is built for exactly this shape of setup: your own domains, your own mailboxes, no shared sending IPs. You connect Gmail, Microsoft 365, or SMTP/IMAP mailboxes across as many domains as you need, and Sendvanta scores deliverability health separately at the mailbox, domain, DKIM, and IP layers — so when one domain starts degrading, you see it as a domain-level signal rather than guessing from aggregate reply rates. If risk thresholds are crossed, sending slows or pauses automatically instead of grinding the domain into the ground.

Warm-up runs on real labeled messages inside the workspace, and seed-mailbox placement tests tell you where each domain is actually landing. Connecting your own mailboxes is free on every plan; the free plan covers 1,000 active leads, 3,000 emails a month, and up to 10 of your own SMTP mailboxes. On paid plans ($29 Starter, $59 Growth, $99 Pro), you can also register a sending domain for a one-time $20 and add managed mailboxes at $4/month each if you'd rather not run Workspace billing for a dozen aliases. More on how the layers work on the features page.

Methodology: written 12 June 2026, based on current Google Workspace and Microsoft 365 per-user pricing, typical registrar domain pricing, and per-mailbox sending limits observed across cold outbound campaigns. Volume figures assume verified lists and follow-ups counted against the same daily cap.