Skip to main content
email·digit

Why your campaign recipient count should add up exactly

The number a campaign tool shows before you send should be the number that sends, and you should be able to see how it got there: who is in the audience, who was removed, and why. In Email Digit the Audience step lists each removal on its own line, and the lines add up to the send, because the preview and the real send list come from the same calculation.

How a recipient count drifts

A preview count that differs from the real send is rarely a dramatic bug. It is usually one of these:

  • Two different calculations. The preview runs one query and the sender runs another. Each is right on its own terms, and they disagree at the edges.
  • Removals applied late. Unsubscribes or a frequency limit are applied only while sending, so the preview never shows them.
  • People counted twice. Someone who is both unsubscribed and in a segment you excluded is subtracted once for each reason, so the preview is too low.

Nobody notices, which is the problem. Every rate in the report afterwards, reply rate, click rate, bounce rate, is divided by a number you never saw.

The rule that makes counts add up

Put the reasons for removing someone in a fixed order and remove each person by the first reason that applies. Each person then appears on exactly one line, and the lines sum to the starting total. Then use that same calculation, not a copy of it, to build the list you send to.

The Audience step in Email Digit

You pick a segment, and the Audience step starts from everyone in it. It then removes people in this order:

  1. Suppressed. Anyone on your do-not-send list: people who unsubscribed, bounced permanently or reported a spam complaint.
  2. Excluded by segment. Anyone in a segment you chose to leave out of this send.
  3. At the weekly cap. Anyone who has already received 5 marketing emails in the last 7 days. Campaigns and sequence emails both count toward it, so two teammates scheduling sends in the same week cannot stack them on the same people.

An example:

LineContacts
In the segment4,200
Suppressed (unsubscribed or bounced)minus 180
In excluded segmentsminus 320
Already at the weekly capminus 200
Will receive it3,500

A person who unsubscribed and is also in an excluded segment is counted once, on the suppressed line, because that check comes first. And 3,500 is what sends, because the same function produces the estimate and the send list. When the send goes out, the calculation runs again with the same rules, so anyone who unsubscribed in the meantime is left out on the same basis.

If the result is zero, because everyone is suppressed or capped, the Send button stays locked and says why. A campaign cannot go to nobody.

Why the denominator matters

Every rate in a campaign report is divided by the number of people who were sent the email. If 35 people reply to a send the preview called 4,200 but that actually reached 3,500, the reply rate is 1.0%, not 0.8%. Small differences like that decide which campaigns look like they worked, which subject lines you keep, and what you tell the rest of the team.

Leaving people out without building a new list

“Everyone except people who are mid-trial” is a common send, and building a separate list for it each time is how lists multiply. In Email Digit you pick the segment to send to, then pick one or more segments to exclude, such as Trial. Anyone in them is removed from this send and counted on its own line. Nothing about the contacts or the segments changes.

When the send is bigger than today allows

A new sending domain needs to build volume gradually; a large first send from a domain with no history is a common way to start badly. Email Digit applies its warm-up ceiling, which is tracked for your account as a whole rather than per domain, at the same point as the other rules and after every exclusion, so the allowance is spent on people who are actually going to receive the email.

If the audience is larger than the warm-up allows today, the extra recipients are held back from this send rather than pushed through it, and the composer shows how many on their own line. The held-back people are chosen in a fixed order, so the ones held back in the estimate are the same ones held back in the send.

Your plan’s monthly email allowance is checked separately at the review step. If the audience is bigger than what is left this month, the review shows how many go now and how many go when the allowance resets on the 1st.

Limits

  • The weekly cap is a platform setting of 5 marketing emails per contact per 7 days. You cannot change it per workspace.
  • The weekly cap counts campaign and sequence emails and decides who a campaign skips. Email your app sends, such as receipts and password resets, neither counts toward it nor is held back by it.
  • Recipients held back by warm-up are not queued for a later day automatically. Reach them with a later send once the ceiling has climbed, and keep time-sensitive sends from a new account small.

If your preview and your send can disagree, every rate in your report starts from the wrong number. See how Email Digit works for marketers, or read how to send a campaign.

Share this guideShare on XShare on LinkedIn