Skip to main content

How much is data quality worth?

Richard Jones Archive

There's no doubt that when compiling massive clusters of data to analyze in an effort to gain greater wisdom about their industries, companies must emphasize accuracy. There's no substitute for superior data quality - if corporations use address management tools to ensure that their information is updated and verified, they can proceed with confidence toward decisions based on solid business intelligence.

Some skeptics, however, maintain that there is a risk of over-emphasizing data quality. While quality is of course important, corporate officials must always be aware of the amount of time, money and effort they're devoting to the endeavor. If they go too far, they may ultimately be doing their businesses more harm than good.

According to IT Business Edge, the tech industry spent around $994 million on data quality last year alone. That's a lot of money, and according to The Information Difference, it marks a 5 percent increase from one year prior. Data quality now makes up 30 percent of the cost of a typical data management project. Is that too much?

Not necessarily. The news source also cites another survey from The Information Difference - this one found that 80 percent of business IT officials see a need for more data quality, saying it's of "key importance" to their big data initiatives.

What the skeptics say
Those who caution against "too much" data quality say it's impeding actual progress. Rajan Chandras, a columnist for Information Week, cited humanitarian efforts that have taken place amid natural disasters internationally, such as in Haiti and Japan, and domestically, as when Hurricane Sandy struck the East Coast in 2012. By spending too much time on data quality, Chandras argued, organizations lost their ability to act quickly.

While speed is also a value, there's also the grave risk of acting too rashly without data quality initiatives. Imagine if an aid organization went to offer assistance after a tornado, but because of poor-quality data, had the wrong location or brought the wrong supplies. The consequences in this scenario could be just as detrimental as the wasted time skeptics warn about.

Chandras recommends that anyone working with big data, whether they're a large corporation or small nonprofit, "take the occasional step back and ask yourself what business value can be obtained from data as is." Taking that step, however, could potentially be harmful. Acting hastily can be just as bad as never acting at all.