How to Build a Learning Culture Within Your Agency

What do government agencies celebrated for their leadership in evidence-based decision-making do to create a culture of learning and continuous improvement? It’s a question more agency executives are asking these days.

That’s due in part to the movement around using evidence and data catalyzed by the Bush and Obama administrations. More recently, the 2018 Foundations for Evidence-Based Policymaking Act puts a new focus on agencies’ use of evidence to drive better program results and cost-effectiveness.

From my work with federal, state and local organizations, I’ve concluded that a good way to summarize what evidence-focused organizations do is this: They think, “Yes, and…” The term comes from improv comedy (see video clip below), but today the technique is used more broadly as a catalyst for effective brainstorming in organizations. The idea is that participants should take what their colleagues say and expand on it, rather than limiting the dialogue or shutting it down entirely.

The idea is a bit different, however, when applying “Yes, and …” to building an evidence culture within government. Leading agencies recognize that any strategy to build or enhance a learning culture has its limits. As a result, they draw on multiple strategies at the same time—employing the “Yes, and…” technique—to create a comprehensive approach to harnessing the power of data and other forms of evidence.

But comprehensive doesn’t have to be complicated. Rather, it’s about understanding the strengths and weaknesses of different strategies and combining them to best meet an agency’s needs. Leading federal agencies apply “Yes, and…” in at least four ways.

1. Use top-down and bottom-up strategies.

A learning culture requires clear and consistent messages from leadership that data and evidence are valued and will be used to inform resource allocation and programmatic decisions and to drive continuous improvement. And… agencies use bottom-up efforts, including peer-to-peer initiatives led by staff. An example is the Education Department’s Evidence Planning Group, made up of staff from across the department who work with grant programs to find ways to better integrate evidence into grant-making processes.

2. Focus on shorter-term and longer-term ways to gather insights.

Too often, people think of evidence-building, including randomized controlled trials, as a slow process. Yet leading agencies use rapid experimentation and data analytics to produce insights for program managers quickly and often cheaply. And… those agencies also create research agendas around key issues that may require multiple studies, over many years, to produce meaningful insights about longer-term questions about what works. Those robust research agendas aren’t quick, but they are better suited to informing major policy changes.  

3. Harness existing and new data.

An important step in advancing an evidence culture is making existing agency data more accessible to program managers, while protecting privacy, to better understand program dynamics and outcomes. And… agencies can partner with other federal agencies to link their data to explore broader questions and deepen their analytical capabilities. For example, the partnership between the Housing and Urban Development and Health and Human Services departments that links housing and health data has produced useful, actionable insights about the relationship between health outcomes and different housing options.

4. Strengthen internal capacity and create research-practitioner partnerships.

Finally, leading agencies create robust chief evaluation offices, and they ensure that those offices’ leaders are in the room when important top-level program decisions are made. That allows those evidence experts to be “truth tellers” for leadership about how much evidence (or how little) is behind different policy or program options. And… agencies harness external evidence expertise by forming partnerships with university researchers to tackle high-priority questions. A good example is the Internal Revenue Service, which has partnered with University of Texas economist Day Manoli. He spends part of his time in Washington, working within the IRS, to generate insights for tax administration from big data, such as how to improve outreach around the Earned income Tax Credit.

A “Yes, and…” approach to creating a learning culture is simply another way of saying a multifaceted one. It’s a good reminder that there are no silver bullets for creating and nurturing results-focused government: Leaders need to use complementary strategies. To its credit, the Foundations for Evidence-Based Policymaking Act encourages a multipronged approach. Agency leaders who take the legislation seriously and adapt its requirements to create valued new initiatives will be on a useful path forward. Even then, will there be more to do to advance evidence-informed government? Of course. Think “Yes, and…”

Andrew Feldman is a director in the public sector practice at Grant Thornton and also hosts the Gov Innovator podcast. He served as a special advisor on the evidence team at the White House Office of Management and Budget in the Obama administration.

Image via Christian Bertrand / Shutterstock.com.