Understanding the Arcane With Data Dictionaries and Governance Plans

This blog post is an excerpt from GovLoop’s recent guide “7 Tips to Transform Your Data Into Compelling Stories.” Download the full guide here.

Try to make it through the following sentence: Inculcating a dogma of data preeminence is an inordinately arduous and often enervating enterprise for agencies to undertake.

I know, ouch.

Whether you’re studying for the SAT or reading a pretentiously written article, having a dictionary on hand never hurts. But to many people, deciphering data can be almost as painful as deciphering the sentence above – which is why data, just like language, needs a dictionary for it to make sense. No matter what truths the content may hold, they’re useless if people can’t understand it.

The Defense Logistics Agency (DLA) recently finalized a data and analytics strategy and governance plan to get everybody on the same terms when talking about data. The military agency, which is in charge of supply chain logistics for the Defense Department, is also looking at reforming and adding fields to its data dictionaries – or repositories outlining the contents, formats and important elements of databases.

These efforts come on the back of an internal push from the agency’s chief data officer (CDO) to demystify data at DLA. The strategy sets up policies for standardizing data, establishing data roles and managing data to ensure quality.

“The waters can get a little muddied, and you suffer from poor data quality,” said Lindsey Saul, Lead Analytics Strategist at DLA. “We are taking measures to really ensure that our data is standardized, that we have tools to help the community with this effort and that we are all working from a single source of truth.”

Saul described DLA’s culture as generally appreciative of the value of data, but said translating raw numbers to business meaning could still be a challenge. For that reason, DLA recently acquired a data visualization tool that, once in full swing, will allow 26,000 users across 28 countries to have access to insights and user-friendly data. While business-side professionals will be able to look through read-only reports, DLA data scientists will have the chance to dig deeper and create their own reports.

Using a commercial data visualization tool, DLA Troop Support discovered “tremendous” cost savings by evaluating about 10% of its subsistence – or food and dining – contracts to find several pricing anomalies, Saul said. Although these insights can technically be discovered without data visualization tools, the technology removes painstaking and lengthy processes of aggregation and analysis by the user to show end results.

With new suites of technology to advance data insights, DLA is helping the military move materials more efficiently and with fewer mistakes. The agency is beginning to plan for predictive analytics and emerging technologies, and the new data strategy unveils a full-fledged vision for machine learning in the next five years, emphasizing the groundwork that will allow DLA to move its technology portfolio forward into next-generation technology.

But before going too far into disruptive technology, Saul emphasized that a foundation must be built to support big-name and big-money investments. That foundation is data quality and data standards, rooted in guiding documents for the organization, Saul said. She noted that the development of DLA’s data governance plan and strategy was the proudest accomplishment of her tenure there.

“At the core of it is data, and making sure that we have our data quality, data standardization, all the relevant themes that come up in our data governance plan,” Saul said.

Photo credit: Defense Logistics Agency’s Flickr