Data quality should be everyone’s job

In a recent post I talked about how misinterpretations of big data and subconscious biases can undermine an analytics program in a way that could jeopardize the business.

Another potential pitfall comes from the data itself. Enterprises now can capture vast amounts of data, much of which (such as information from mobile apps or geo-location) didn’t exist not long ago. The sheer scale of data collected makes finding valuable information a challenge, which is why data analysts get paid a lot of money to write sophisticated algorithms than can sift through huge data sets — including unstructured and internal data — to gain new insights and competitive advantages.

But while big data is good, not all data is good. There is such a thing as bad data, and it can be costly: A 2014 study by Experian Data Quality concluded that “75% of businesses are wasting 14% of revenue due to poor data quality.”

So it’s important that data scientists and enterprise business leaders recognize when data is bad in order to avoid making poor decisions based on information that is faulty (and perhaps fundamentally at odds with reality). The first step toward making a commitment to data quality is to make a public commitment to data quality. Assign the responsibility for guarding against bad data to the chief data officer or another leader of your analytics team, and then announce it. It’s that important. 

Second, emphasize to employees that while this person is where the buck stops, data quality is everybody’s job. Call center workers, for example, play a critical role in ensuring data quality because they do the initial job of collecting and recording consumer personal information. If they generate poor-quality data, the business over time will waste money through its failure to deliver messages to the customer, its targeting of inappropriate customers (thus wasting marketing resources), and its efforts to recollect and upgrade data, and other activities that can misdirected due to faulty data.

As to available data-quality tools, an enterprise’s overall data-quality strategy, needs and goals should guide the choice of vendor software and services. That’s a topic for another post.

Does your enterprise have a data-quality strategy?

Speak Your Mind


This site uses Akismet to reduce spam. Learn how your comment data is processed.