Modernizing Data Sharing and Analytics Is Critical to Solving Our Nation’s Challenges
D3Damon/iStock
We must evolve the way we analyze and use the treasure trove of data at our disposal.
From day one, the Biden administration has emphasized the use of data to solve many of the challenges we’re facing today. Federal agencies were already required by law to have their own chief data officers, and many states have hired someone into this role as well. Biden has raised the bar further by issuing a Memorandum on Restoring Trust in Government Through Scientific Integrity and Evidenced-based Policymaking and deferring to scientific data for efforts around COVID-19, climate change, and more.
To tackle our most pressing challenges, we must evolve the way we analyze and use the treasure trove of data at our disposal. We must look for ways to break down data silos and modernize our analytics capabilities so that we can improve data sharing and accelerate solution building. And we must leverage technologies that allow us to proactively anticipate future needs so when the next crisis hits, we’re ready.
Here’s how to do it.
Build an Infrastructure for Data Sharing Among Agencies
This starts at the top. The Biden memorandum specifically calls for open and secure access to federal data and directs heads of agencies to make that data available and easily understandable. Now, it’s up to agency CIOs to take the data baton and pass it. There can be no egos here, no harboring territorial feelings about who owns what data. That adversarial mindset leads to incomplete analyses. Better decisions are made when data is shared and given context based on information that is collected across different agencies.
Yes, there are data sharing requirements to be considered, but the more information that can be shared and used appropriately, the more likely that federal agencies will be able to take the correct and decisive actions.
Create a Data Trust
The creation of an overarching data trust can make data sharing much easier while minimizing risk. Unlike a data-sharing agreement, which can be established for ad hoc data-sharing initiatives, a data trust is a formal agreement between several organizations that details what data will be shared, how it will be used and for what purpose. For example, a data trust between a health care agency and another government entity might state that data can be shared between the two, but only if the other entity agrees to abide by HIPAA regulations and delete any personally identifiable information.
The federal government can take cues from the states. Virginia, for instance, has established a data trust between various state agencies for its Framework for Addiction Analysis and Community Transformation program, which the Commonwealth has established to address its opioid crisis. Meanwhile, Massachusetts is sharing data among relevant government agencies to predict what office space and leasing requirements might be needed post-pandemic.
Leverage Technology for Greater Efficiency and Insights
Technology can make the modernization process easier and more efficient. By leveraging APIs, CIOs can provide seamless data sharing access across devices and platforms—much easier than emailing spreadsheets. And agile development approaches can help agencies implement new data analytics technologies faster and deliver incremental value along the way. The agile methodology relies on regular feedback from stakeholders, which means that developers are more likely to deliver the solution that was envisioned. In addition, smaller procurements are able to more easily reflect changes in direction than the long multi-year procurements of the past.
Once all of this initial groundwork has been laid, agencies can use their technology to more intelligently manage and use data. Predictive analytics, AI and machine learning can automatically look at millions of data points. Based on the pattern of the collected information, they can model potential outcomes and provide recommendations to proactively manage situations today and challenges tomorrow. User-friendly dashboards can be created to show these recommendations in ways that are easily understandable by anyone, not just data scientists.
Data in itself cannot help us achieve our objectives. Data must be supported by people and technology, with a special emphasis on processes that allow data to be more readily shared amongst different stakeholders. Without all these things, we are destined to continue to see only one side of the story.
Clarke Allen is senior director of strategic business at Qlarion.