Email List Segmentation by Engagement: A Simple Framework

Sending the same message to everyone costs you reputation. Segmenting by engagement protects it.

The four tiers

1. Active: clicked, replied or purchased in the last 30 to 60 days. Send your full programme.
2. Warm: engaged in the last 3 to 6 months. Send less often with your best content.
3. Cooling: no engagement for 6 to 12 months. Send a re-engagement series.
4. Inactive: no engagement for a year or more. Suppress or remove.

How to build it

• Use clicks, replies, purchases and logins, not opens, as signals.
• Add one more tier outside the four: risky addresses (accept-all, role) from your verification.
• Refresh the segments automatically each month.

What it gives you

• Better inbox placement because most sends go to engaged people
• Higher click rates and revenue per send
• Early warning when a segment starts to decay

Start with clean data

Segmentation only works when the list is real. Verify first with the bulk verifier, then build tiers.

Building the tiers from data you already have

You do not need a complex tool. Most platforms record the last click or last open and the sign-up date. Use a rule such as:

• Active: clicked or replied in the last 90 days.
• Warm: clicked or replied 91 to 180 days ago.
• Cooling: no click in 181 to 365 days.
• Dormant: nothing for over a year.

Because open rates became unreliable, base the tiers on clicks and replies rather than opens, and count purchases or logins too if you have them.

How to mail each tier

• Active: your normal programme, including promotions.
• Warm: slightly lower frequency and your most useful content.
• Cooling: a short re-engagement series, then back to normal or out.
• Dormant: a final "do you still want this?" message, and removal if there is no answer.

Numbers to compare

Look at the unsubscribe and complaint rate per tier. They are almost always higher in the cooler tiers, which is why mailing them at full frequency hurts your reputation more than it earns.

Start from clean data

Segments built on a list with dead addresses make the cold tiers look colder than they are. Verify the list first with the bulk verifier so a "dormant" contact is a real person who stopped reading, not a mailbox that no longer exists.