How many mailboxes do you need for outbound?

How many mailboxes do I need is really a maths question working backwards from your goal. Guessing leads to either burned domains from overloading, or too little reach to matter. A few minutes of arithmetic gets it right.

Start from the outcome, not the tool

Begin with the number that matters: how many real conversations or positive replies you want in a period. Everything else follows from that, not from how many mailboxes you happen to have.

Working forward from tooling produces arbitrary volume. Working back from the goal produces the capacity you actually need.

The chain of estimates

Turn the goal into a mailbox count through a simple chain.

1. Target conversations, the outcome you want in the period.
2. Divide by your expected positive-reply rate to get emails needed. Use your own benchmark, not an internet average.
3. Spread that over the period to get emails per day.
4. Divide by the safe daily limit per mailbox to get the mailbox count.

A worked example of outbound capacity planning

Numbers make the chain concrete. Take a team that wants twelve real conversations a month from cold email.

1. The goal. Twelve conversations per month.
2. The positive-reply rate. This is the assumption you supply. Use the rate your own last campaign produced; if you have never run one, pick a deliberately pessimistic figure and correct it after the first hundred sends. For the sake of the example, call it 1.5%.
3. Emails needed. Twelve divided by 0.015 is 800 sends per month, counting follow-ups.
4. Emails per day. Roughly 21 working days in a month, so about 38 sends a day.
5. Mailboxes. At a conservative 30 sends per mailbox per day, that is two mailboxes at full stretch, or three with sensible headroom.
6. Contacts required. If the sequence runs four touches, 800 sends means around 200 new contacts entering the sequence each month.

That last line is the one people skip, and it is usually the line that breaks the plan. Two hundred verified, right-fit contacts a month in a narrow niche is real work, and no amount of mailbox capacity substitutes for it.

Notice how sensitive the whole thing is to the reply-rate assumption. Drop the estimate from 1.5% to 0.8% and the same twelve conversations need 1,500 sends and four to five mailboxes. Which is why you rebuild the plan on your own numbers after the first full cycle rather than carrying the initial guess forward.

What a safe daily sending limit looks like

There is no single official figure, and anyone quoting one precisely is guessing. In the campaigns we run, a new mailbox stays on a small daily trickle right through warm-up, and a mature one settles at a few dozen sends a day at most, lower for Microsoft 365 tenants than for Google Workspace in most cases. Treat that as a shape rather than a number, and let your own bounce and reply data set the actual ceiling.

The limit that matters is not the provider's technical cap. It is the volume at which your engagement signals stay healthy. A mailbox sending 25 a day with replies coming back looks like a person. The same mailbox sending 25 a day into a stale list with no replies looks like software, and the ceiling drops. Volume and list quality are the same problem viewed from two sides, which is why how many emails per day is safe always comes back to who you are sending to.

Respect the per-mailbox ceiling

The safe daily limit per mailbox is the fixed constraint. You do not raise capacity by exceeding it, that burns the mailbox. You raise capacity by adding mailboxes, which is exactly why inbox rotation exists. Plan the mailbox count around the safe limit, not around wishful volume.

Do not forget the list

Capacity to send is useless without enough right-fit contacts to send to. Check that your addressable list supports the volume, especially in a small market where the universe is finite.

Sending capacity and list size have to match. Excess capacity into a thin list just means repeating yourself or reaching poor-fit prospects.

Build in warm-up and headroom

New mailboxes need warm-up time before they carry full volume, so plan capacity ahead of when you need it. Leave headroom rather than running every mailbox at its absolute limit, which gives room to absorb variation without pushing anything into the danger zone.

Warm-up is measured in weeks, not days. A mailbox created on Monday cannot be part of Tuesday's plan. Build the infrastructure two to four weeks before the campaign that needs it, on a separate sending domain so nothing you do here touches the reputation of your main corporate domain.

Capacity planning mistakes and how to fix them

• Planning from an internet benchmark. Reply rates published in vendor reports come from different markets, offers and list quality. Fix: run one cycle, measure your own rate on an outbound reporting dashboard, then replan. Until then, plan pessimistically.
• Counting first sends only. A four-touch sequence quadruples your daily volume against the same contact list. Fix: build the plan on total sends, follow-ups included, not on new contacts.
• Buying mailboxes on the day the campaign starts. Fix: order and warm the infrastructure while the list and the copy are still being built.
• Ignoring bounce rate in the maths. A list with a 6% bounce rate does not just waste 6% of your sends, it degrades the reputation that carries the other 94%. Fix: verify before sending and treat bounce rate as a capacity input, not an afterthought.
• Running every mailbox at its ceiling. A single provider hiccup then pushes you into the danger zone with nowhere to shed load. Fix: plan at roughly 70% of the safe limit and keep the rest as reserve.
• Never checking the plan again. Deliverability drifts. Fix: keep monitoring in place and rebuild the capacity plan every quarter.

Capacity planning in small markets

The arithmetic assumes an addressable universe large enough to feed it. In the Baltics, that assumption breaks quickly. In Lithuania, once you filter by company size, sector and role, the named contacts who genuinely fit a given B2B offer are often the entire market rather than a sample of it. Build and count that list before you plan volume, because in a small market the count decides the plan rather than the other way round.

When the market is finite, capacity planning inverts. You are no longer asking how many mailboxes support the volume you want, you are asking how slowly you should spend a list you cannot replace. Reaching every relevant contact in the country in six weeks with mediocre copy is not efficiency, it is burning the market.

The practical response is to widen geographically rather than sending harder. A Vilnius-based team running campaigns into Germany, Poland and the Nordics has an addressable universe an order of magnitude larger, which makes the volume side of the plan straightforward and moves the difficulty to language and localisation instead. DACH campaigns in German, Baltic campaigns in Lithuanian or English, and one shared reporting line is a more workable shape than pushing a small home market past what it can absorb.

In either case the constraint that binds is list quality, not mailbox count. Mailboxes are cheap and quick to add. Right-fit contacts are neither, which is the whole argument behind quality over quantity in outbound.