Tuesday, October 5, 2010

It's not personal. It's math.

If you drive a car, use a bank or a credit card, shop at a grocery store, rent movies from Netflix, have a cell phone, surf the web, or live in the 21st Century, you’re probably a denizen of a databank somewhere.

Like it or not, Google, Facebook, Walmart and the government are building profiles of you. They know who you talk to and what you search for online. They know what you purchase and how much you buy. They know your likes and preferences.

Always watching!Your car probably has a black box in it—just in case you’re ever in an accident—but also to track your position anywhere on the globe via GPS. Ditto your cell phone.

There’s no escaping scrutiny. You may keep a low profile, but these guys know more about you than even close family. You may hide your Mars Bars addiction at home, but these guys have your number.

And that’s just what you are to them: a vast series of numbers that describe your attributes.

Data mining is used to drill down through all those jumble of numbers to find the attributes (described by those numbers) that should resonate with your audience. Data mining is simply finding the patterns, relationships and correlations in all that stored information.
Needle-in-the-haystack
As marketers, you probably know that the problem is that there is simply so much data, it is hard to extract the exactly right bit of information to pinpoint your perfect target. And finding exactly the right bit is critical. It’s like looking for a needle in a haystack.

But fear not, gentle marketer. Exhaustive new programs are being written—even as I type—to simplify and clarify the process. Stay tuned. It is a brave new world out there. Soon even more information about your customers/donors/prospects will be available. At a price.

Have the money? If so, you can look like a marketing genius. After all, getting your message to someone who has a high likelihood of a positive response earns you better return on marketing investment. Your job is still to define the characteristics you are looking for. Then let the computers do the heavy lifting and sifting.

So what about individual privacy? Now there’s a quaint 20th Century concept. Forget about privacy in an era of Facebook and all-seeing cameras at every street corner. Sorry, it’s history.

Netflix and Amazon are using data mining today to increase sales. Enter their websites and they recognize you. They then recommend books and movies to you that “people like you” have enjoyed. The recommendations are based on the theory that you will enjoy what people like you have enjoyed. Aka, similar number strings like similar things.

It's Data Mining!Can beauty be in the eye of the beholder? Probably. If Netflix makes a movie recommendation and you like the pick, the information becomes useful. You come to rely on Netflix’ subsequent recommendations and the process becomes less like advertising and more like a friendly exchange between friends. Even robots and computers need friends, it seems.

It doesn’t always work perfectly. Sometimes you question what where they thinking. And sometimes you agree. But don’t take it personally.

Remember, it’s not personal. It’s math. It’s data mining.  And you are just a number in a computational labyrinth.

No comments: