eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
The problem with having data and the analytic power to crunch it is that options and policies that were previously infeasible can suddenly become quite practical—but also, perhaps, controversial.
For example, California regulators and insurance companies are now in the throes of an argument about the variables that ought to go into calculation of automobile insurance rates. A statewide ballot initiative, passed in 1988, required that rates be based mainly on driving record, number of years licensed and total miles driven by an insured driver each year, rather than being based largely on the location where the driver lived.
Only this month has a major insurer, the Automobile Club of Southern California, agreed to observe the 1988 law—a week before it would have been required to do so by new rules. Other insurance companies have successfully deferred this change by lobbying and courtroom battles, arguing that neighborhood variation in frequency of claims is important in assessing risk.
Groups opposing those insurers have suggested that rates based on postal ZIP codes have been tools for discriminating against minorities and have argued that the practice unfairly penalizes responsible drivers who happen to live in neighborhoods with historically high insurer costs.
Auto insurance has long been used as an example of what Ill call the “data dilemma.” In 1970, Robert Townsend (previously president and chairman of Avis Rent A Car) suggested that some crusading insurance CEO should selectively disclose crash statistics to top-tier customers in such a way that his or her company would be sued for failing to make that data broadly available.
In his book, “Up the Organization,” Townsend hypothesized that the CEOs goal should be to lose the case “in such a way that henceforth, all insurance companies must supply data to the public on exceptionally dangerous vehicles.”
Townsend wasnt talking, specifically, about the topic of this column; he was talking in broader terms about corporate social responsibility. He asked his readers two questions: “How would your industry be serving the public interest if it had just one honest chief executive officer?” and “Does it alarm you to know that your industry doesnt have a single honest CEO?” Feel free to apply those questions to your own domain.
In the meantime, its clear that many companies (as well as nonprofit entities) are in possession—whether they realize it or not—of information that their customers and clients might find either useful or, perhaps, offensive.
Using that information to set prices, or to allocate products to retail outlets, or to prioritize customer service calls is good business and is the main selling proposition of costly CRM (customer relationship management) technology, but there are also controversies waiting to erupt over the discovery that such things are happening.
For example, imagine the reaction if it became known that Apple had reduced the allocation of iPods to stores in inner-city neighborhoods where police departments (hypothetically) had reported high rates of muggings that appeared to target iPod wearers. Would this be considered a public-spirited act on Apples part or a disgraceful act of discrimination?
Imagine the reaction if a companys customer service phone-wait times were substantially less for callers in neighborhoods with above-average income and education. Data might well show that smart customers ask their questions more clearly and understand the answers more quickly and that rich customers represent a more valuable source of likely repeat business—so it would be only good business judgment to use caller ID and neighborhood demographic data to serve those groups more expeditiously. How would this spin, though, if the story came out?
I believe in collecting data, milking it for all its worth, and using the results to maximize efficiency and create competitive advantage. No company should fail to consider, though, the social perspectives from which the rational use of data might be perceived—and to factor in these questions before spending a lot of money to gather data and generate analysis that it might be afraid to use.
Technology Editor Peter Coffee can be reached at [email protected].