August 12, 2014

Incremental Value Of Email Marketing

It's pretty common, folks ... you take a look at email marketing performance, and let's be honest ... it's tepid:
  • Average = 5 Campaigns Per Week.
  • Average Demand per Email Delivered = $0.08.
On an annual basis, you multiply the $0.08 by 5 Campaigns Per Week by 52 Weeks to get $20.80 of annual value.

Well, that's not too bad, is it? Let's pretend that the metrics above are for a 12-month buyer. Now we'll look at the 12-month buyer in total ... a 40% annual repurchase rate and 2 annual purchases and an AOV of $100 ... yielding $80 of future demand per 12 month buyer.

All of a sudden, email marketing is responsible for more than 25% of future value. Not too shabby.

Unfortunately, email marketing is measured via opens/clicks/conversions. There are two better methods for measuring email marketing performance.
  1. Holdout Tests. Don't mail a random sample of email subscribers for "x" weeks. You'll learn, immediately, if email marketing has true incremental value, or if email marketing is just cannibalizing the rest of the business.
  2. Modeling. I frequently create logistic regression models for annual email response, and regression models for spend. Here, you see all sorts of interesting trends. If you find that email subscribers are worth $10 and your opens/clicks/conversions suggest the email subscriber is worth $20, then you know that about 50% of email marketing is incremental, while 50% is cannibalized from online orders that would have happened anyway. If you learn that email subscribers are worth $30 and your opens/clicks/conversions suggest the email subscriber is worth $20, then you know that you are under-estimating the importance of email marketing!
Either way, it's pretty darn important to measure the incremental value of email marketing. Go do it!!