Email Marketing 4 - Disaffection Index

I have mentioned the Disaffection Index in previous posts on email marketing but have not discussed it in detail. A year or so on I am still talking about it and use in training courses for the Institure of Direct Marketing (IDM) and so thought we should dive a little deeper.

The Disaffection Index (or DI) was first mentioned in a MediaPost publication wrtten by Melinda Krueger. It is effectively a better way or working with the unsubscribe rate, which is traditionally calculated by dividing the total number of unsubscribers by the total number of emails successfully sent. Although a very common calculation it is not a good metric to measure your performance on as it is skewed by all those people who did not choose to open that particular email.

Also, the unsubscribe rate from experience tends to be very consistent from one edition of the same newsletter to another, and only really changes with major format or content-type changes. Looking back over some historical client data there can be as little a change as 0.1% over all newsletters sent in a year.

DI instead calculates itself by dividing the number of unsubscribers by unique clicks.


DI = (unsubscribes / unique click) *100

This offers a measure of delivery vs promise, or put more literal gives you the total number of people who have actually taken an action just to not hear from you again.

Let me give you a real example from an old client (a major UK men's lifestyle magazine), who send out a weekly newsletter (that is exceptionally popular) and then rent their list for solus broadcasts too.

The results from a typical month are below (July 2006):

(each value is the DI for that mailing, and the lower the number, the better it was received by the audience)

Newsletter 1: 2.9
Newsletter 2: 2.9
Newsletter 3: 2.9
Newsletter 4: 2.9

You can see from these numbers that their newsletter performs consistently well, but in the same month they also sent out 6 solus emails on behalf of advertisers. Bear in mind that solus advertising emails will receive a higher DI generally because of the very nature

So the month started off ok with 3 broadcasts:

Mobile phone offer: 5.4
Godfather DVD Release 1: 5.6
Godfather DVD Release 2: 7.8

The mobile phone offer was well received by the audience and the DI barely moved. Similar for the first DVD offer about the 're-release of the Godfather in a week's time' but it did lift a little when they reminded the audience the same DVD was released that day.

Online Gaming: 15.6

The next offer of online gaming saw a huge jump and if I was the data owner I would be telling my sales folks to prioritise other advertisers above anyone pushing gaming.

Mobile phone offer: 5.4

Amazingly another offer for a mobile phone contract got the lowest DI of all commercial broadcasts for the month, rare given it was the 5th one in just a few weeks.

Godfather DVD Release 3: 63.0

However, the month was finished off with a final promotion for the Godfather DVD reminding people it came out the week before, an interesting strategic decision anyway, and one that saw the DI score leap to 63.0.

Think about this for a moment - that means that 63% of all people who decided to take some action with their email took that action just to tell them never to content them again. This is powerful insight.

My recommenedation is to always include DI in your reporting and to calculate it based on historical data. As a stand alone metric for a single email it will tell you nothing but over a period of time it will become the most vaulable part of your defence acting as your own marketing early warning system.