The perfect data management approach doesn’t exist, according to government experts. And because nobody’s perfect, they advised organizations to stop striving for perfection in data management and instead settle for “good enough” to reach effectiveness.
At ATARC’s Data Management Breakfast Summit on March 30, the experts explained how a dreamy data management utopia is not a reality, and data scientists should focus on achievable outcomes.
“Don’t let perfection be the enemy of the ‘good enough,’” said Hengyi Hu, chief data analytics officer at the U.S. Product Consumer Safety Commission. “And sometimes a ‘good enough’ is a 20 percent solution – it’s just to get your feet wet.”
Hu said that organizations should experiment and not be “scared of what’s to come,” as far as new technologies go.
“You can’t be agile in a perfect environment, there has to be some leeway to make mistakes, there has to be some leeway to pursue new things,” he added. “Let’s not try to pursue that perfection, because I have never seen it – that technology is way too new to be able to do that. All I can say is, let’s get involved.”
Jason Kane, deputy assistant director in the Office of Investigations at the U.S. Secret Service, agreed that a perfect data utopia doesn’t exist, and noted that “we’ll be having this conversation much longer.”
As for state and local government organizations, one data manager explained how at the state level, they face the same data management challenges, “and yet still haven’t quite solved the problems.”
“As optimistic as I like to think myself to be, there’s no such thing as perfect. Not even as people, we always have room for improvement – and the same applies for data,” said Amberle Carter, enterprise data manager in the Information Technology Division at the Texas Department of Public Safety.
“You’ll always have messy data that needs to be cleaned up, you’ll always have data policies that need to be revised to address new standards or requirements, you’ll always have new architecture to build,” she continued. “There are so many components that will just continue to come our way, and we’ll continue to improve it and we’ll get our practices better, but never perfect.”
Instead of pursuing perfection, Lauren Pavlik, chief of the Data and Software Services Division at the Army’s Enterprise Cloud Management Agency, encouraged organizations to pursue effectiveness.
Pavlik explained how the end user doesn’t see the data model, but they do see the information and the end result.
“If we’re not looking at that in terms of effectiveness, the data management approach doesn’t really matter,” she said. “We have to be able to adapt the data, the data management, and the data model approach that we’re going after to fit the need of what our end state value is to our end user – whatever our use case may be for that particular subject matter.”