|
HOW I LEARNED TO STOP WORRYING AND LOVE DATA MINING 11.11.97 by Michael J. A. Berry D S * I first became involved with data mining and database marketing while working for a now defunct manufacturer of supercomputers. Most of our multi-million dollar massively parallel machines were installed at places like Los Alamos, Sandia, White Sands and the Naval Research Laboratory. A number were installed at Washington area agencies known by three-letter acronyms. Others were simply picked up by unmarked trucks and driven off to locations unknown. Because these exotic machines were difficult to program, we generally shipped applications engineers along with them to help the customers develop their parallel applications. For a time, I worked as one of these AEs. After returning from one stint as an AE at a French military research institute, I found myself in a very different environment helping one of our few commercial customers develop a massive database of charge card transaction detail records that could be used for developing targeted marketing campaigns. I found the work fascinating, but was unprepared for the reaction of many of my coworkers. Many of these people who spent their days helping design atomic weapons or improving the targeting of cruise missiles had moral qualms about my work improving the targeting of junk mail! At first, I thought that their concern was that I was aiding and abetting an industry that was already filling their mailboxes with lingerie catalogs and balance transfer offers. I would explain that if the application I was working on was successful, they should receive less junk mail, not more; that if we could figure out that they were vegetarian, we would stop sending them Omaha Steak coupons. None of this made people feel any better because the underlying issue was not junk mail, but privacy. The idea that someone is snooping through their credit card purchase histories makes most people nervous. If they knew that their supermarket purchases, their drug prescriptions, their magazine subscriptions and telephone records were also going into databases somewhere, they would be even more nervous, but most people do not give this much thought as they hand over their discount card and watch as their frozen peas and herbal tea bags cross the check-out scanner. Judging from the reaction of many of my friends and acquaintances to my recent book, Data Mining Techniques for Marketing, Sales and Customer Support, the only thing preventing a massive outcry against data mining in the service of direct marketing is the general ignorance of what is going on. Well, I do know what is going on and I am not alarmed. This article is an attempt to explain why not. It's Nothing Personal Despite all the rhetoric about "relationship marketing," no company is interested in you as an individual human being; they are interested in you as a potential customer. Information about you that sheds no light on your propensity to buy a certain product is simply not interesting to the purveyor of that product. Your age, income, sexual orientation, political affiliations, number of credit cards and fondness for lima beans may all be things you choose not to share with your neighbors. But even personal information that would set the neighbors tongues wagging for weeks on end is of no interest to the commercial data miner unless it can be shown to have predictive value for your likelihood to order from a catalog or default on a loan. If someone -- an ex-spouse or a collection agency, perhaps -- is really out to get you personally, they don't need data mining. The information that finds its way into marketing databases has always been available to those willing to look hard enough. In Massachusetts, where I live, if a car repeatedly blocks my driveway, I am free to go down to the registry of motor vehicles and look up the name and address of the owner. I can then walk over to the registry of deeds and see what his house is worth and I can look up his number in the phone book if I want to call him up and give him a piece of my mind. What is new with data mining is not the ability to see who owns a particular car, but the ability to scan thousands of automobile registrations looking for patterns that will help predict who is likely to spend $60,000 or more on a luxury sport utility vehicle. It May be Worth Something As consumers begin to realize the value of information about themselves and their habits, they will start charging for it. Already, many supermarkets pay you for this information. Until recently, most supermarket transactions were anonymous. The scanner collected data about what was purchased, but not by whom. Even anonymous purchase data is quite useful for market basket analysis, promotion planning, store layout and the like, but if purchase information can be tied to an individual, it is even more valuable -- valuable enough that the store is willing to pay you for it in the form of additional discounts offered to people who identify themselves each time they make a purchase. That is the purpose of the discount cards that have recently appeared in many supermarkets. The chain is now in a position to be an information broker, charging the manufactures of the products on its shelves to get a coupon or other promotion directly into the hands of the most likely purchasers. As consumers become more savvy, they will start expecting to be paid for this soft of information in other situations as well. Nobody Wants to Upset You Companies are only collecting data on you because they want to sell you things. If they misuse that information in ways that make you mad, you will be less likely to buy their products and services. That is why MCI asks you who your friends and family are instead of just figuring it out for themselves and calling them up. This fear of upsetting you is the strongest protection you have against the misuse of your personal data. In a later column: When Should We Be Worried About Data Mining --- Michael J. A. Berry, Naviant Technology Solutions, mjab@naviant.com +1 617 591-3041 http://www.data-miners.com |
|