Fresh Azure tenants: 100 mailboxes you warm yourself

Warming a hundred mailboxes is not warming three, a hundred times over

This is the part that catches people. Almost every warming tool charges per mailbox. Three mailboxes for three weeks is a rounding error. A hundred mailboxes for the better part of two months is a real bill, and it arrives every month until you are done.

Work that number out before you order, not after. Take what your warming tool charges per mailbox, multiply by a hundred, multiply by two months. Compare it to the gap between this and a prewarmed tenant. For a lot of teams the gap closes completely, and some find that warming it themselves costs more than buying it warmed.

The other cost is the calendar. It is not a hundred separate 51-day clocks, it is one clock running on all hundred at once, and nothing you buy can make it shorter.

One reputation, and nothing to fall back on

Every mailbox in a tenant shares one reputation, and on a fresh tenant none of them have any yet. There is no history to absorb a mistake, which is the strongest argument for filling a first tenant with 25 mailboxes rather than 100: the price is the same and a bad first month costs you a quarter as much.

When you buy in packs of three and something goes wrong, you lose three inboxes. Here you can lose a hundred at once. That is the trade you make for the cost per inbox, and it is the same trade the prewarmed tenant asks you to make.

The difference is that a prewarmed tenant has already sent and received real mail for three weeks and survived it. A fresh one has proved nothing. You are taking on the concentrated risk and the unproven start at the same time, which is why this is the most advanced thing on the site rather than just the cheapest.

Bring them online in batches, not all at once

Do not switch on a hundred brand new mailboxes on the same morning. Start about twenty, let them run, watch what happens, and add the next twenty when the first are behaving. A hundred accounts appearing from nowhere and immediately sending is a pattern worth avoiding.

Verify every list before it touches the tenant, and check your bounce rate daily for the first fortnight. Bounces are the fastest way to lose a tenant, and on a hundred shared mailboxes they are the most expensive.

Plan the rotation before the first send rather than after. Ten to fifteen emails per mailbox per day across a hundred mailboxes is a thousand to fifteen hundred a day, which is plenty, and going higher buys you very little while risking all of it.

On a three-month commit, the warmed one is the same price

This is the fact that should decide it. A fresh tenant is $50. A prewarmed tenant is $69 a month, or $50 on a three-month commit — so if you were going to keep the tenant for a quarter anyway, the warmed one costs you nothing extra and arrives having already done the 51 days. At that point buying fresh is paying the same money to wait.

Fresh only wins on price if you genuinely need the tenant for a month or two. Below three months it is $50 against $69, and you are trading $19 a month for seven weeks of not sending.

If you have never run a tenant, do not start with a fresh one. The prewarmed version exists because the first two months are where tenants die, and having someone else take that risk is most of what you are paying for.

Buy fresh when all three of these are true: your start date is more than two months out, you already run warming infrastructure and know what it costs, and you have done a staged tenant startup before. If any one of them is false, the prewarmed tenant is cheaper than it looks.

And if you are somewhere in between, split it. Run a prewarmed tenant for what is live now and warm a fresh one behind it for next quarter. That is what teams sending at this volume actually do.