Yesterday, for the first time ever, the Dow Jones Industrial Average closed above 12,000. Two days earlier, the U.S. Census Bureau announced that the United States has now officially became a nation of 300 million people.
What do the Dow Jones industrials and the U.S. population have in common, other than having both set new records? For one thing, both are highly dynamic.
Of course, stock prices and markets fluctuate on a daily basis. And while the U.S. population only grows larger with each passing day, as indicated by the parking situation at my local megamall, demographic statistics point to a continuous state of flux, as well. This year, for example, there will be over 40 million residential moves. There will also be four million births, two million marriages, a million divorces, a million retirements and 2.5 million deaths, give or take a few hundred thousand.
For those of us immersed in the world of database marketing, these population shifts speak to the importance of Customer Data Integration (CDI) as a key enabler of precision marketing—and, ultimately, increased revenue growth. It also speaks to the importance of CDI in the context of bottom-line cost reduction (e.g., sending merchandise catalogs to dead people is never a good idea; no matter how compelling the offer, you’re unlikely to elicit a favorable response).
Following years of experimentation, often with less-than-satisfactory results, companies today have come to view CDI as a critical success factor. And more of them are taking advantage of recent innovations in CDI solutions to better propagate a single view of the customer across the organization. Aaron Zornes, founder of the aptly-named Customer Data Integration Institute, predicts that the market for CDI solutions (software and services) will exceed $1 billion by 2008.
I'm told by my colleagues at Fair Isaac that the two primary criteria in CDI solutions are match rate – the proportion of records that are successfully matched to a reference database ID – and precision, the accuracy with which those matches truly reflect the individuals and households concerned. These two criteria tend to trade off against one another in that being overly aggressive on matching by loosening rules to achieve a hit against a reference database can lead to poorer accuracy.
For example, James Brown (the Godfather of Soul) and Jim Browne (his next-door neighbor) may be viewed as one and the same because they’re “close enough”. But in fact they’re two different people, only one of whom is included in the reference database. By combining these two names, you increase the match rate but decrease the precision.
In marketing to potential consumers in situations where the offer has low risk and low cost, companies tend to err on the side of a higher match rate, sacrificing precision in the process. In situations where there is significant downside –- for example, a pre-approved offer of credit –- companies tend to err on the side of caution and focus on precise matching over volume of match rates. They'd prefer to throw Jim Browne away rather than risk sending him an inappropriate offer.
In short, most applications tend to cluster around two universes: a marketing-oriented universe and a risk-oriented universe. Providers of CDI solutions generally choose to build their solutions around one of these clusters. Those with a focus on marketing solutions may sacrifice precision for match rates while those with a focus on risk may emphasize precision over match rate.
At Fair Isaac, we seek the middle ground, balancing match rates and precision to achieve an optimal balance. That said, certain business situations certainly benefit from turning the dial one way or another. A client whose stated business goal is “customer acquisition at all costs” will be more inclined to seek higher match rates and risk lower precision. An example might be a company aggressively building market share in long distance telephony. On the other hand, some clients (e.g., a credit card company or catalog merchant, for whom precision marketing is a mantra and tailored offers are the norm), may be more inclined to sacrifice match rates for accuracy in identification.
In any case, CDI is merely a means to an end. Resolving discrepancies in the spelling of customer names and addresses, and the variations in their phone numbers, birth dates and other numerical identifiers, reduces mailing costs while increasing response rates. Bigger picture: CDI provides the foundation for deploying decision management systems that enable companies to deliver highly relevant customer experiences, services, messages and offers. For marketers, it’s a prerequisite for doing business in a nation of 300 million people.
