Pandemic highlighted how messy data sharing is, Mastercard exec says

Dive Brief

  • Though existing systems and protocols were helpful in dealing with disruption, the pandemic pointed out “holes in the data ecosystems,” as data sharing across multiple stakeholders gets “messy,” said JoAnn Stonier, chief data officer at Mastercard, speaking Thursday at Forrester’s Data Strategy & Insights North America 2020 conference.
  • Protocols involved in safekeeping data, such as sandboxing, became critical during the response to COVID-19, but more evolved techniques around data governance will need to emerge in order to meet the challenges ahead, she said. 
  • The pandemic made it clear data governance protocols needed to become “more sophisticated and more automated” as complexity increases, Stonier said.

Dive Insight

With external disruptions, organizations relied more heavily on tech leadership, and asked new questions from data in order to respond.

Data became simultaneously more valuable and more vulnerable. Remote workforces became exposed to a record number of cyberattacks, bringing security protocols like zero trust front and center.

The global race to develop a COVID-19 vaccine is an example of a process that will require more advanced methods for data sharing, Stonier said. But complexity also increases when sharing specific kinds of data. 

There are legal considerations around the security and privacy of data related to healthcare, with federal laws restricting release of medical information dictating data’s treatment. 

The more partners need to communicate, the more intricate data-sharing becomes.

“If it’s the state of North Carolina to the U.S. federal government, we can probably figure that out,” said Stonier. “But when it’s all 50 states to the federal government, it gets more complicated. When it’s all 50 states to each other, that gets even more complicated.”

There’s additional pressure to enhance data governance structures with the rise of automation across critical business processes. Without access to data that powers techniques, such as machine learning or robotic process automation, AI models won’t be effective at increasing productivity.

“When people talk about the new normal, I laugh, because we’re not going backwards,” said Stonier. “I think we’re just moving forward into the next generation of data challenges that we now need to figure out.