Market Researchers Should Torture Data Until It Confesses
by Bob Hull
Wednesday, 07 March 2012
Category A Day in a (MR) Life
0 Comments
Bob Hull
Guest has not set their biography yet
User is currently offline
A former colleague of mine used to tell us to “torture the data until it confessed”. In other words, don’t just stop your investigation at the first finding. But rather, keep poking, prodding, flipping and coercing until you feel you’ve uncovered all the data has to give. Ah…images of Jack Bauer doing his thing flash through my mind just thinking about our own data “torture” sessions.
All kidding aside, what my colleague was really trying to say was spot on. I’m sure we’ve all known researchers who habitually stop at the first find. They rarely take the time to consider different ways of looking at data, of considering the message within.
Tags: segmentation, Market Research, Choice
Later this month, those of us in the United States celebrate one of my favorite holidays, Thanksgiving. Officially, Thanksgiving is a post-harvest celebration that was brought to the Americas by European settlers in the 16th or 17th century (depending on which historian you believe). Unofficially, it's the day where families and friends gather to feast, take naps and watch football. Oh my, even as I type this my mouth is watering...turkey, potatoes, stuffing, cranberry sauce, peas and the like, with chasers of pumpkin, apple and other assorted pies. All delicious, but I particularly love eating turkey on Thanksgiving.
My seven year old son gets a $2 per week allowance. He doesn't really do anything to earn this money. Rather I give him (and his brother) an allowance to teach them how to save for things that they want. Implied, and in fact part of the bargain, is that they can't hassle me for Pokémon cards, or Wii games, or anything else they "need", because they have their own money. Well, about a month or two ago my seven year old mandated that I start paying him with a $2 bill. Yikes! Where was I going to get even one $2 bill, let alone one every week?
Prior to my current tour at TRC I was a partner at a small data mining boutique that had a simple objective: support the sales and marketing goals of our clients by helping them stem attrition. While the goal may have been simple, how we set out to do this was not. See, we took upwards of 30 months of time-series data on all of their customers and applied some mind-numbing statistical techniques that identified patterns in the data that preceded an eventual behavior. In most cases that behavior was a customer terminating their relationship with our client. Once our system identified the patterns that preceded attrition, we would then continue to feed it customer data each month and it would dutifully output a list of customers that were likely candidates to attrite. In addition, for each customer on this golden list we provided the prediction "trigger", or the thing that they did that was responsible for the system flagging them as a high-risk customer.
