Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

Disconnect between CIOs and LOB managers weakens data quality

Clint Boulton | Feb. 9, 2016
Despite the promise of big data, most analytics projects suffer from bad data syndrome, the result of tension between CIOs and line of business managers.

cio business bad data quality

Having inaccurate or incomplete data is bad for business. Yet substandard data quality is a staple of many organizations, largely because of a disconnect between CIOs and line of business managers, according to a 451 Research survey of 200 senior IT and business leaders from large enterprises. The research bolsters the case for a chief data officer (CDO), a C-suite executive who can both serve as a liaison between CIOs and business managers and boost data quality.

Only 40 percent of C-level executives are "very confident" in the quality of their organization's data, says Carl Lehmann, the 451 Research analyst who wrote the report, commissioned by Blazent, a company that provides what it describes as data intelligence platforms. That 40 percent figure is a concerning sign at a time when 94 percent of senior IT leaders say that poor data quality impinges business outcomes, resulting, for example, in lost revenue or bad strategic decision-making.

“A big chunk of respondents thought that a considerable amount of value is lost, which doesn’t justify a laissez-faire attitude,” Lehmann says. "You'd expect the respondents to say ‘we need to really get on top of this game’ instead of 'I'll deal with the data I've got and make do." CIOs know that bad data is bad for business, but they aren't putting enough time or resources behind their data quality efforts. That’s dangerous at a time when businesses are seeking competitive advantages with their data.

Big data + big problems = big headaches

Corporations such as TD Ameritrade, Navistar and Merck Animal Health are on integrating and managing data for business insights, a cornerstone of digital transformations orchestrated by the CEO and the rest of his or her C-suite. The dark secret is that most of the data is poor. Data migrations and integrations often introduce errors, yielding data gaps or duplicate information.

dqm 2
Data quality issues are caused by a combination of  human errors and technology failures. (Click for larger image.)

As a result, Lehmann says that only 50 percent of organizations surveyed say that their data quality management, as well as the quality of data overall, was only slightly better than satisfactory or “good enough in general” – far from a ringing endorsement. And 8.5 percent say they don’t have a data quality management strategy at all. Instead, they “hope for the best,” he says.

Corporate data is captured in myriad ways, including manual entry and automatically via an array of computers, mobile devices and software. Next, the data is relayed to larger data warehouses for processing by IT. The CIO’s team processes and cleans the data, weeding out inaccuracies and duplications, a process known as normalization. IT staff typically builds reports or sets up self-service analytics systems for the business lines. Yet, despite this process, data quality problems still arise from poor data entry by employees or customers, data migrations or conversions, changes to source systems, and the introduction of new systems.


1  2  Next Page 

Sign up for CIO Asia eNewsletters.