Matchbacks Are Bothering Me. My New Year’s Resolution.
It’s the end of the year and my New Year’s Resolution for 2016 will be focused on figuring out why match rates for the matchback are so low. Have you seen your match rate lately? Is the low rate bothering anyone else? (If you’re not familiar with matchbacks, look to the bottom of this blog and read the Sidebar.)
My curiosity in 2015 was increased when I started seeing match rates below 30%. That might be fine but the rest of my brain was thinking that the catalog database is fairly new, and a majority of the file is mailed (plus some modeling on lapsed buyers.) So if I’m mailing most of the healthy file, modeling the lapsed customers, and not mailing the inactive customers…where are the 70% unmatched originating?
Yes, I thought about the other marketing efforts happening as well as social media, mass media, emails and online search. And of course the more brand awareness, the more likely to garner a purchase. All of this makes perfect sense to me.
Now to the “what really bothered me” part.
If 70% of the data is unmatched, and I’m mailing the majority of the file, then I immediately draw two conclusions.
- The unmatched are new customers/first time buyers. OK great. Did the file grow by “X” number of customers?
- The unmatched are customers who are on the file but were not mailed. OK great. So the customers who are lapsed, inactive and didn’t qualify for the model are buying again?
I have more questions but those two are fairly big pieces of the puzzle, and I’d like to answer those first.
The other thing I did was run a correlation coefficient analysis. This is a statistic to reveal if the unmatched data is correlated to the matched. I wanted to have a better idea if the unmatched orders/revenues are random or is it somehow related (correlated) to the matched data. A coefficient of .50 or greater is considered correlated and has a strong relationship. The two charts below have a correlation coefficient of .8726 and .7640 respectively. Finding out the correlation coefficient is high, I’m now even more curious about the unmatched data and what, if anything, does it have to do with the catalog mailing. (The charts are small, but I just want you to see the visual relationship between the red and blue lines.)
As I close out 2015, I am resolute to finding the answers to the matchback mystery during 2016. Stay tuned for updates. Let’s bring on the data and ring in the New Year!
Sidebar Matchback: The matchback is an analysis of looking at what was mailed versus the orders received during the life of the mailing (e.g. postcard, catalog, direct mailer.) Usually this is performed by the service bureau (a third party) and typically the service bureau will NCOA (National Change of Address) and standardize all addresses. This keeps both sets of data (the mail file and the order file) current. The rest is a boring technical story but the outcome is determining how many of the orders matched the mailing.
Dean of Marketing, CatalogUniversity.com
President, Hemisphere Marketing, LLC
The Catalog Blog covers opinions and information on all things catalog. Have something to add? Leave a comment below. Catalog University is devoted to helping you get ahead in the fun and fascinating world of catalogs. If you want even more information about cataloging, including FREE Pub Talks, be sure to sign up for the Cat-U mailing list.