Note from Jeanne: Happy Throwback Thursday! It was 2 years ago that I did this work with a client (and published this article on ClickZ) — but the lessons on smart list management are timeless. Read on to learn how we did it — and how you can do it with your own email marketing campaigns.
A lot of companies looking to improve their email marketing focus on creative, like subject lines and the copy, design, and/or layout in the body of the email. This isn’t wrong, but there’s another element you should be looking at: your list selections.
Here’s a case study showing how I helped one of my clients nearly triple their ROI, taking it from $0.94 to $2.77, primarily by focusing on which lists they sent to.
This is the tale of two sends, mailed just about a month apart.
As you can see, for the first send we had one control and two test creative executions. We generated just $0.94 for each $1 we spent on the send – meaning that we lost money overall. But the control was profitable, generating $1.52 for each $1 it cost.
This is a good result; we have a profitable creative execution. In most cases you’d simply take the creative winner, develop a new test, and do another A/B split test the following month to the same lists.
But we wanted to see if we could boost performance even further with some list analysis.
Thanks to the client’s data team we were able to match revenue generated back to the specific list selects.
For the first send we used nine different selects from four different lists. Here’s the breakdown:
Let me take a minute to talk about list segmentation and selects. I often get asked about the best way to segment a list. There’s no single answer to this question: it depends on (a) how you’re going to use the list and (b) what data is available to segment by.
In this case, the client is a large organization with more than 20 business units and too many products to count. In their jargon, “list” means the source of the data. In this example:
- List A is a list of industry names they purchase each year from a legitimate list broker.
- The broker provides the titles (along with other data points) associated with each email address, so for this list they often segment that way.
- List B is a list that was developed in-house by this business unit – they’ve collected the email addresses of purchasers as well as those who have browsed their e-commerce website while logged in.
- While we would be able to segment this list by title, past experience has shown that it’s more effective to look at engagement and recency, so we took current and previous year purchasers along with those who have explored the e-commerce site this year.
- Lists C and D are lists that have been developed by other business units within this organization.
- While we don’t have specific titles here, we know what part of the organization these names are from (I’ve called it “department” here); we’ve also taken a small portion of List D for which there’s very little data (Miscellaneous), on the off chance that, since there’s little information about them, they won’t be mailed much and we’ll get a good response.
Going back to the grid above, you can see that only three of the nine list segments had an ROI over $1, meaning that they were profitable. The other six were not profitable.
For the second send we decided to go to only the three profitable lists. This cut our send quantity down dramatically, from more than 250,000 addresses to just over 100,000 names.
With the second send, our ROI nearly tripled. We went from earning $0.97 for each dollar spent to earning $2.77 overall – solidly profitable.
The control (which was the same as the control in the first send) saw a similar lift in ROI, from $1.52 per dollar spent to $4.10. Total revenue for the second send was 12 percent higher than for the first, even though our mail quantity was less than half the first send.
When we looked at list performance, all three lists were profitable on the second send.
And, while we were testing creative on both sends, the lifts we saw were the result of list management and selection – not of creative gains. Because the control remained the control over both sends and our test creatives could not beat it.
One more note: this type of list management helps guard against over-mailing, which can cause erosion in list performance. By looking at ways to mail less but earn more for each email you send, you can improve performance without fatiguing your list.
Until next time,
Jeanne
