How Many Sending Domains Do You Need for Cold Email?

The short answer, then the math

If you want a number before you read the reasoning: plan for 2 to 3 mailboxes per sending domain, then back the domain count out of your volume. The chain is simple.

1. Start from your monthly send target.
2. Divide by the safe per-mailbox daily cap (turned into a monthly figure) to get the number of mailboxes you need.
3. Divide that mailbox count by 2 or 3 to get the number of domains.

So the domain question is really the last step in a volume calculation, not a standalone decision. The two upstream numbers (how much one mailbox can safely send, and how many mailboxes you need) are covered in depth elsewhere in this cluster, so this guide stays on the part everyone gets wrong: how many domains sit underneath those mailboxes, and why you don't pile them all onto one.

Here's the same idea as a one-line formula:

1domains = ceil( mailboxes_needed / mailboxes_per_domain )
2mailboxes_needed = ceil( monthly_target / (safe_daily_cap × sending_days) )

Plug in your own numbers and you have an answer in under a minute. The rest of this article explains why the ratio is 2 to 3 and not 10, why provider limits are the wrong starting point, and how risk diversification changes the math once you scale.

Start from your target, not the provider's maximum

The single most common planning mistake is sizing a cold email setup around what a provider technically allows. Those caps are huge, and they have nothing to do with what's safe for cold outreach.

Google Workspace lets a paid user send up to 2,000 messages per day, dropping to 500 for trial accounts, per Google's published sending limits. A free gmail.com account is capped near 500 emails per day, per Gmail's sending and receiving limits. Microsoft 365 (Exchange Online) enforces a recipient rate limit of 10,000 recipients per day and a message rate limit of 30 messages per minute, per Microsoft's Exchange Online limits.

Look at that throttle on Microsoft's default domain: outbound mail from an onmicrosoft.com address is limited to 100 external recipients in a rolling 24 hours. That's the platform telling you, in numbers, that you need a real domain to send anything serious.

But here's the thing. None of those ceilings is your planning number. A warmed cold email mailbox should send a small fraction of the provider maximum, because reputation, not the technical cap, is the real limiter. We cover the exact recommended figure in how many cold emails per day per mailbox is safe, and the official provider caps in detail in Gmail sending limits for cold outreach. For this article, just hold one idea: you plan from the safe cap, which is far below the technical cap.

Turn volume into mailboxes (the quick version)

Before you can count domains, you need a mailbox count. This is its own topic with its own worked examples in how many email accounts do you need for cold outreach, so I'll keep it to the formula here.

Pick a conservative per-mailbox daily cap. Operators commonly run cold mailboxes in a band well under the provider maximum, and for planning I'll use a round, conservative 30 emails per day across roughly 20 sending days a month. That's about 600 emails per mailbox per month. Use whatever figure the per-mailbox guide leads you to. The structure doesn't change.

1mailboxes_needed = ceil( 10,000 / 600 ) = 17

A 10,000-per-month target needs about 17 mailboxes at that conservative cap. Now we can finally answer the domain question.

The mailboxes-per-domain ratio (this is the real question)

So you have 17 mailboxes. Do you put all 17 on one domain? Two domains? Seventeen? This ratio is what most "how many sending domains do you need for cold email" searches are actually asking, and the answer is a narrow range: 2 to 3 mailboxes per domain for cold outreach.

Why not more? Two reasons, one of them straight from Google's own rulebook.

Reason one: reputation and risk concentrate at the domain. Mailbox reputation rides on domain reputation. If one domain picks up spam complaints or lands on a blocklist, every mailbox on that domain suffers together. Stack 10 mailboxes on a single domain and a bad week takes out 10 mailboxes at once. Spread those same mailboxes across 4 or 5 domains and a single blocklisting costs you 2 or 3, not the whole operation.

Reason two: Google counts subdomains under the primary domain. This trips people up. You can't dodge the math by spinning up mail1.yourbrand.com, mail2.yourbrand.com, and so on. Google's email sender guidelines FAQ states plainly that "messages sent from the same primary domain count toward the 5,000 limit," and defines a new domain as one that "hasn't sent more than 5,000 emails a day to personal Gmail accounts since January 1, 2024." Translation: subdomains share the primary domain's reputation and its bulk-sender accounting. They're a structuring tool, not a multiplier. (The subdomain-versus-root decision is its own question, covered in subdomain vs root domain for cold email.)

The sweet spot of 2 to 3 isn't a law of physics. It's the point where you've diversified enough to survive a single domain getting burned, without drowning in setup and warmup overhead. More on that tradeoff below.

How many sending domains do you need? Worked examples

Let's run three real-shaped scenarios using the same conservative assumptions: 30 emails per mailbox per day, about 20 sending days a month (so ~600 per mailbox monthly), and 3 mailboxes per domain.

Scenario A, the solo founder: 1,000 emails/month.

1mailboxes = ceil(1,000 / 600) = 2
2domains   = ceil(2 / 3)       = 1

One domain, two mailboxes. A founder testing a market doesn't need a sprawling fleet. One well-warmed domain with a couple of mailboxes carries this load with room to spare.

Scenario B, the small sales team: 10,000 emails/month.

1mailboxes = ceil(10,000 / 600) = 17
2domains   = ceil(17 / 3)        = 6

Six domains, roughly 17 mailboxes spread across them (call it 2 to 3 per domain). If you'd rather run a more cautious 2 mailboxes per domain, that's 9 domains. Either is defensible. The lower ratio buys resilience; the higher ratio buys lower cost.

Scenario C, the agency or scaled team: 50,000 emails/month.

1mailboxes = ceil(50,000 / 600) = 84
2domains   = ceil(84 / 3)        = 28

Twenty-eight domains at 3 mailboxes each, or about 42 at a 2-per-domain ratio. At this scale the fleet becomes infrastructure in its own right, and the per-domain blast radius is exactly why you don't cram those 84 mailboxes onto five domains.

Swap in your own per-mailbox cap and the table shifts, but the method holds. Notice how the domain count scales almost linearly with volume once you're past the trivial tier. That's the part teams underestimate when they budget.

Why spread across domains at all? Risk diversification

If one domain could safely carry your whole volume, you'd use one and skip the hassle. It can't, and the reason is how mailbox providers score senders.

Reputation is tracked per domain. Google Postmaster Tools reports a domain reputation signal, and a blocklist entry attaches to a specific domain or IP, not to your company as an abstract entity. So a single sending domain is a single point of failure. One spam-trap hit, one complaint spike, one careless campaign, and that domain's placement craters. If it was your only domain, your entire pipeline stops that day.

Spreading volume across multiple domains turns a catastrophic failure into a contained one. Lose one domain out of six and you've lost roughly a sixth of your capacity while you recover or replace it. That's the whole logic of domain risk diversification for cold outreach: no single domain should be load-bearing for the whole operation.

A few practical rules fall out of this:

• Don't share reputation across unrelated campaigns. Keep a risky, high-volume prospecting push on different domains than your steady, relationship-driven sends.
• Treat each domain as disposable but recoverable. When a domain's numbers slide, you pull it from rotation, diagnose, and either rehabilitate it or retire it. The playbook for that is in how to recover a burned cold email domain.
• Monitor every domain independently. Per-domain reputation, blocklist status, and bounce rates are the early-warning system. See domain health monitoring for the dashboard view.

This is also why a rotation layer matters. Once you have six or twenty domains, you need something to distribute sends evenly and pull unhealthy domains out automatically, which is the job of an inbox rotation strategy.

Keep cold outreach off your primary brand domain

Here's a rule that overrides the math: your main company domain, the one your website and billing and support run on, should never be a cold sending domain. If yourcompany.com gets blocklisted because of an aggressive prospecting campaign, you don't just lose outreach. You lose internal email, invoices, password resets, the lot.

So the domains in your fleet should be secondary sending domains bought specifically for outreach. Most operators register lookalike variations of their brand: getyourcompany.com, yourcompany.io, try-yourcompany.com, and similar. They're recognizably yours to a prospect, but they isolate all sending risk away from the asset you can't afford to lose.

A few guidelines for secondary domains:

• Buy them recognizable, not deceptive. A close variant of your brand builds trust; a random string looks like spam.
• Authenticate every one separately. Each domain needs its own SPF, DKIM, and DMARC records. There's no shortcut, and the setup is identical per domain. The full walkthrough is in the SPF, DKIM, DMARC guide.
• Point each one at your real brand. Redirect the secondary domain to your main site so a curious prospect lands somewhere legitimate.

If you're deciding between a dedicated outreach subdomain on your brand domain versus a separate registered domain, that tradeoff gets its own treatment in subdomain vs root domain for cold email. Short version: a separate registered domain gives you the cleanest reputation isolation for cold sending.

The hidden cost of more domains

Diversification is good until it isn't. Every domain you add carries real overhead, and over-provisioning wastes money and time without buying more safety.

Each new domain means:

• A registration fee, typically a low yearly cost per domain, plus renewal.
• Its own authentication setup, the SPF, DKIM, and DMARC records mentioned above, configured and verified per domain.
• Its own warmup runway. A brand-new domain can't send cold volume on day one. It needs weeks of gradual ramp before it's safe, and Google treats new domains on an accelerated enforcement timetable per its sender guidelines. Plan that runway with the email warmup complete guide.
• Its own slot in your monitoring and rotation. More domains, more dashboards to watch.

That's why the 2-to-3 ratio is a ceiling-and-floor, not a "more is always better" rule. Going to 1 mailbox per domain triples your domain count and your overhead for marginal extra safety. Going to 6+ mailboxes per domain saves money but reconcentrates the risk you were trying to spread. Two to three is where the curve flattens.

There's also a warmup-time argument for not over-buying. If you provision 30 domains and only need 8, you're paying to warm and maintain 22 idle domains, and idle warmed domains drift. Buy what your volume calls for, with a small buffer, and add more as you scale. The setup itself, provisioning mailboxes across those domains, is covered in setting up multiple Google Workspace accounts.

A simple rule of thumb

If you skip everything above, take this:

• Under ~1,000 emails/month: 1 domain, 2 mailboxes. Done.
• 1,000 to 10,000/month: size mailboxes from your volume, run 2 to 3 per domain, monitor each.
• 10,000+/month: treat domains as a managed fleet with rotation, per-domain health tracking, and a replacement pipeline.
• Always: keep your primary brand domain out of the rotation entirely.

And size the fleet from your safe per-mailbox cap, not the provider maximum. The whole calculation only works if the upstream number is honest.

Common questions

How many mailboxes per domain is safe for cold email?

Two to three is the standard recommendation. The limit exists because mailbox reputation rides on domain reputation, so the more mailboxes you stack on one domain, the more you lose at once if that domain gets blocklisted. Mature senders with tight monitoring sometimes run 4 to 5, but going higher reconcentrates the risk you were spreading.

Do subdomains count as separate sending domains?

No. Google's sender guidelines FAQ states that messages from the same primary domain all count toward the bulk-sender threshold, so mail.yourbrand.com shares reputation accounting with yourbrand.com. Subdomains are a way to structure and isolate streams, not a way to multiply your reputation footprint. Separate registered domains are what actually diversify risk.

Should I send cold email from my main company domain?

No. A blocklisting on your primary domain can knock out internal email, billing, and password resets, not just outreach. Register dedicated secondary domains for sending and redirect them to your main site.

How many sending domains do I need for 10,000 emails a month?

Using a conservative 30 emails per mailbox per day (~600/month), that's about 17 mailboxes, which lands at roughly 6 domains at 3 mailboxes each, or 9 at a cautious 2 per domain. Adjust the per-mailbox cap to your own figure and recompute.

Does each domain need its own warmup?

Yes. Warmup is per domain and per mailbox. A new domain has no sending history, and Google applies accelerated enforcement to new domains, so each one needs its own gradual ramp before it carries cold volume. The schedule is in the email warmup complete guide.

Can I just use one domain and send slowly?

For very low volume, yes, one domain is fine. The reason to add domains isn't only volume, it's resilience. One domain is one point of failure. If it burns, everything stops. The moment your pipeline can't afford a single domain going down, you need at least a second one.

Putting it together

Counting sending domains isn't a vibe, it's arithmetic with a safety margin. Back out your mailbox count from a conservative per-mailbox cap, divide by 2 or 3, keep your brand domain out of the line of fire, and add domains as volume grows rather than front-loading a fleet you can't warm.

The operational headache is running that fleet once it exists: distributing sends, watching per-domain health, pulling burned domains, and warming new ones. MailBeast handles the rotation and per-domain monitoring layer so a multi-domain setup behaves like one system instead of twenty spreadsheets. The broader picture of how these pieces fit together lives in cold email infrastructure.

Sources

• Google Workspace Help, Gmail sending limits in Google Workspace
• Google, Email sender guidelines
• Google, Email sender guidelines FAQ
• Google, Limits for sending & getting mail (Gmail)
• Microsoft Learn, Exchange Online limits