A Crisis of Data ConfidenceBy Michael Vizard | Posted 02-03-2014
By Michael Vizard
Becoming a truly digital business involves leveraging data to create a sustainable business advantage. Clearly, there is an incredible amount of interest in creating a new generation of IT systems that leverage big data from different sources. However, while IT has never had more tools available for deriving business value from its IT investments, the biggest impediment may well be the fact that business executives often doesn't trust the data that IT has collected.
A recent survey of 442 business executives conducted by Harvard Business Review Analytics Services at the behest of QlickTech, provider of QlickView business intelligence software, finds that only 16 percent of the executives surveyed were confident in the accuracy of the data they used to make business decisions. Another 42 percent said they were not confident in their decisions simply because they couldn't get access to all the relevant data they needed.
The good news is that with the rise of technologies, such as Hadoop, it's now possible to cost-effectively save every piece of data, says QlikTech CTO Anthony Deighton. Organizations need to adopt this strategy not for compliance reasons, but more importantly because the data may become valuable in the future. When all data is being saved, the next step is make that data accessible in a way that doesn't necessarily require the skills of a data scientist or data analyst.
"Business users are usually a lot better than IT as identifying patterns and outliers in data," says Deighton. "What they don't want to do is having to sit there knitting SQL tables together."
Tools that make it easier for end users to visualize data not only help make organizations more productive, they increase confidence in the veracity of the presented information, Deighton says.
That's critical because one of the primary reasons Green Mountain Coffee and other organizations are investing in analytics is to enable end users to self-service their own business needs in a way that also improves productivity.
"We're trying to empower people to do things in a way that doesn't always require manager approval," says Nate Isham, a network engineer for Green Mountain Coffee.
Ensuring the data quality is reliable is of paramount importance because many business decisions are irreversible once they are made. Interestingly enough, Chris Moyer, CTO for Hewlett-Packard Enterprise Services, notes that as organizations become more sophisticated about using data, an interesting side benefit starts to emerge that helps ensure the quality of their data: because it's now easier to access multiple sources of data inside and out of the organization, the overall quality of the organization's data starts to increase.
"Organizations are not locked into a single data source," says Moyer. "They can now correlate information across multiple sources to make sure the data they have is accurate."
Moyer says mastering those multiple sources of data is critical because IT organizations are being asked to leverage existing investments in systems of record to create applications aimed at serving the unique needs of customers, partners and suppliers.
Many of these applications are going to be of a self-service nature, says Moyer. But if those applications are based on unreliable or otherwise faulty data, CIOs may be doing more damage to their organization's brand than if they had opted to do nothing at all.
About the Author
Michael Vizard is a contributing writer for CIO Insight. To read his previous article for CIO Insight, "Broncos Ride Broadband Network Into the Playoffs," click here.