After looking at simple metrics like new customers and performance by segment (evaluating performance at a 10 foot level), you need to step back, and evaluate performance at a 10,000 foot level.
I always start with what I call a "comp segment analysis". You've heard me say this several times, but I am surprised that I've yet to see one person perform this analysis in my travels over the past five years ... not one single person mentioning it on Twitter, not one person in the blogosphere, or at any conference I've been to ... not one.
I like to look at customers who purchased exactly two times last year ... in some cases, I'll add a 0-3 month recency filter to the query, but overall, I like to look at customers who purchased exactly 2 times last year.
Say we're looking at the month of February. I will freeze the file as of 1/31, and identify all customers with exactly two purchases in the year prior to 1/31. Then, I simply calculate the average amount each customer spent from 2/1 - 2/28.
It's not a difficult analysis.
It is a revealing analysis.
Here's an example. Tell me what you see going on here:
In this case, comp segment performance is down by about 6%, so that's not good. However, look at the "Existing Comp" and "New Comp" columns. Existing items are performing reasonably well, heck, they're even outperforming last year.
In this example, new items are the problem ... performance is bad, very, very bad, and has gotten worse each of the past six years.
Have you done this analysis?
Like I said, I've yet to run across anybody who does this. In this case, the marketer pinpoints the problem with performance. It isn't landing pages. It isn't offers in email campaigns. It isn't long-tail search performance. It isn't channel shift. It is a merchandising issue, one being driven by an inability to either create new items, or one being driven by terrible performance among new items.
Run the analysis. Let us know what you learn. If you don't have the resources to run this analysis, give me a holler, I'll do it for you.
In football, pro teams liberally borrow from colleges, and colleges happily borrow from high schools. Read this article for details (clic...
Look at the first four rows of our life table (values of 0/1/2/3). These are the first 12-15 weeks after a customer buys for the firs...
You probably run Life Tables for your customer file, right? Right? They've been around forever ( click here for a reference f...
If you don't like geeky math, please skip this post, because I am about to show you how the sausage is made! I have eight variables in...