Organizations have vied for years to be data-driven, leveraging spreadsheets, analytics solutions, and machine learning technology to draw fresh insights from their data. Indeed, the focus in recent years has been to amass data from every nook and cranny of the corporate digital infrastructure to eke out all possible benefits from first-party data.
But not all data is good data, and bad data can lead to outright disaster — potentially with organizations walking right off the cliff while patting themselves on the back for a job well done.
A data disaster in space
One story that exemplifies the disastrous effect of bad data would undoubtedly be the crash of NASA’s Mars Climate Orbiter. Built at a cost of USD125 million, the 638kg robotic space probe was launched in 1998 to study the Martian climate. Unfortunately, the probe burned up in the Mars atmosphere after a miscalculation placed it too close to the planet.
A blog on IT Chronicles summed up the situation: “The problem was that one piece of software supplied by Lockheed Martin calculated the force the thrusters needed to exert in pounds of force — but [the] second piece of software, supplied by NASA, took in the data assuming it was in the metric unit, newtons.”
In a nutshell, a critical bit of bad data at the wrong time literally destroyed the efforts and work of hundreds of engineers and scientists.
According to a report by ZoomInfo, 71% of marketers report that their systems run on old, outdated data, and 57% have no standard operating procedure for keeping data up-to-date. While ZoomInfo would have you pay them for access to its database of business contacts, the findings sum up the state of most organizations’ data health.
A focus on quality data
This brings us to data debt, which is the accumulated cost associated with the suboptimal governance of data assets. While inaccurate data is easy to fix at the point of data entry, the cost of rectifying bad data balloons when not rectified. Ungoverned and ignored over time, it can culminate in catastrophic outcomes at worst or hurt data initiatives and data-driven decisions at a minimum.
The cost of poor data is real. According to Gartner, poor data quality costs organizations an average of USD12.9 million. Aside from an immediate impact on revenue, the analyst firm says poor quality data increases the complexity of data ecosystems and results in poor decision-making.
So how can organizations ensure that their data quality is on par? The first step, according to Gartner, is to establish how improved data quality impacts business decisions. This starts with the creation of a list of data quality issues and how they are impacting business KPIs and revenue.
Only then can data and analytics leaders start building a targeted data quality improvement program. The organization will need to clearly define the scope, the stakeholders, as well as a high-level investment plan.
Of course, any plan will require the definition of an acceptable standard of data deemed to be “good enough”, before pushing it out across all business units. On this front, Gartner recommends the creation of a special interest group for data quality that operates across business units.
Building a data culture
Keeping data clean is a journey and empowering a special interest group is but a start. Long-term success calls for an organization-wide data culture where employees are encouraged to value and practice the use of data – data must matter to everyone, not just a select few.
The building of a data culture begins with commitment and trust, says JY Pook of Tableau. The first starts by putting the right data in the hands of employees and treating it as a strategic asset that leaders and employees alike seek to maximize.
Trust is equally important and works both ways: From the leadership to the employee, and vice versa. With transparent access to accurate data and the approval to utilize it, employees can take on greater responsibility and accountability to deliver superior outcomes with data.
Ultimately, we know that data quality is directly linked to the quality of decision-making, while organizations with mature and aligned cultures experience up to four times more revenue than those without.
The time to get started on your data debt is now.
Paul Mah is the editor of DSAITrends. A former system administrator, programmer, and IT lecturer, he enjoys writing both code and prose. You can reach him at [email protected].
Image credit: iStockphoto/Christian Horz