Backhoe Analytics, the Red Sox and Kevin Hillstrom

by Bill LaPierre on October 7, 2012

If you have never driven a piece of heavy construction equipment, don’t like baseball, or do not know who Kevin Hillstrom is, this posting may not make sense to you. But for the rest of you, here goes…

My backhoe is 40 years old this year. This is a full size, regular construction job backhoe. It weighs over 8 tons, and is amazingly handy. (How do the rest of you get along without one?)  Although I’ve owned it for 19 years, I still can’t say I know how to use it. My wife and I own a lot of forested land, with lots of trees. Lots of trees that now show the scars of being hit by the backhoe over the years. I even managed to take a chunk out of the trim on the corner of the house once. And as Mike Hayden (from 4Cite Marketing) often reminds me, when our house was under construction, I accidently ripped out the pipe to my septic tank – twice. Thankfully the system was not yet live.

My use of my backhoe only goes to prove my theory about catalog and website analytics – just because you have the tools, it doesn’t mean you know how to use them.

Mercifully, the Red Sox 2012 season is finally over. They managed to lose more than 90 games for the first time since 1966. How did they manage to fall so far, after winning the World Series twice in the last nine years? One reason lies with the loss of Theo Epstein, the young general manager who, starting in 2002, pioneered the use of statistics in baseball. Epstein used player’s statistics to determine which players to draft, trade, include in the starting line-up, etc.  The proper use of statistics was a large part of the reason the Red Sox won the World Series in 2004 and 2007.  (The other reason they lost 93 games this year is that they now have a team of slacker prima donnas – but let’s stick to statistics).

So what does all this have to do with catalogs, circulation planning and catalog analytics?

If you don’t subscribe to Kevin Hillstrom’s blog at, you are missing out on one of the great minds in cataloging and ecommerce today. Kevin is a rock star among geeky statisticians and analysts. In my opinion, he is one of the few people in our industry that can peer into the future. And he does it with statistics.

I’m not going to tout his blog – his writing speaks for itself. (How does he manage to publish daily? – I’m so envious.) But I will tout his work. Kevin gave a great presentation at this spring’s NEMOA conference in Boston explaining his modeling of the new catalog customer triad – Jennifer, Judy and Jasmine. These three personas represent the distinct sub-groups of catalog and ecommerce shoppers today which Kevin has identified. Check out Kevin’s blog for full descriptions of his theory.

Over the years, I’ve worked with other modelers – I’m sure many of you have too. The main problem I have with most of them is that they approach building circulation models as if it were a science project sponsored by NASA. Nothing is simple. Usually, nothing is understandable either.

That is where Kevin has set himself apart. First, there is a simple grace to his method – identify the three personas on a client’s customer file, and establish separate, profitable contact strategies for each. Second, his modeling code is equally simple. There is a ton of complex math and statistics behind it, but that is all transparent to the user.

Success is where Kevin truly sets himself apart – his models work. Kevin implemented his three persona model for one of our client’s this spring. The client followed Kevin’s instructions for hold-out panels, and control panels. I should note as well that this was for a client that had a relatively small percentage of Jennifer and Jasmine customers. The net result was that Kevin’s models predicted where the client could eliminate circulation without sacrificing demand, and ultimately be more profitable.

Going back to my theory of backhoe analytics (“just because you have the tools, it does not mean you know how to use them”) – there are a ton of knowledgeable statisticians and analysis in our industry.  There are very few like Kevin Hillstrom that have talent in addition to knowledge.

So if you are not a follower of Kevin yet, sign up for his blog. Better yet, talk to him about modeling your customer file. I am a firm believer now that his techniques can work for even the smallest of catalog companies.

Now, if only we could get Kevin to come back to New England more often.

PS: After having finished in dead last place in the American League in 1966, the Red Sox came back in 1967, behind Carl Yastrzemski, to make it the World Series – which they of course lost in 7 games. But still, anything is possible next year.

By Bill LaPierre

VP – Business Intelligence and Analytics


Previous post:

Next post: