Can ground-level research inform the development of strategy and policy at scale?

Behavior health disorders can impact how someone feels, how they think, or what actions they take. Leaders in the State of Washington weren’t seeing the behavioral health outcomes they wanted. 

In a report from 2024, a working group noted that the state had made some positive steps, but that they still ranked 40th in the nation for youth mental health. 

Acting on a legislative mandate, the state launched a collaborative effort, now called Washington Thriving, to align the state around a strategy to improve behavioral health outcomes. Washington Thriving focuses on young adults from before they are born (prenatal stages) through age 25. They also focus on families and caregivers. 

As part of their strategic planning process, Washington Thriving reached out to Bloom Works to see how discovery sprints could help.

Bloom Works conducted 4 discovery sprints around behavioral health topics in Washington. Very quickly, we realized this was more than a discovery sprint project. 

The sprint topics focused on specific behavioral health areas in Washington:

• Behavioral health during pregnancy
• K-12 school-based behavioral healthcare
• Transition-age youth (ages 16 to 25) with complex behavioral health needs
• Complex hospital discharge

We split the sprints into 2 sets of 2, which ran at the same time. Throughout the work, we communicated with our core partners at Washington Thriving. This kept them updated and got us real-time feedback and clarity. 

In total, we talked to around 200 participants across all 4 topics, including more than 50 people with lived experience. We talked to youth, caregivers, relevant agencies, community-based organizations, providers, and educators.

The goal of a discovery sprint is to deliver actionable recommendations in a short period of time (in this case, about 10 weeks per project). The goal isn't to solve the problem in that timeframe. A textbook discovery sprint does well when there is a clear implementation or planning problem. They work best at ground level with clear boundaries. Essentially, a discovery sprint is best when the problem is, “We don’t know why X is happening with our service and need to figure it out.” 

As we collected research, we realized that the problems we were researching wouldn’t have the clear boundaries of an ideal sprint. These problems didn't focus on a service or a cluster of services. These problems dealt with life experiences in particular settings (K-12 schools) or over a specific time (prenatal). Any of them could touch a huge number of challenges, services, and agencies. We were going to have to navigate between multiple ground-level challenges and 100,000 foot system issues at the same time. This back and forth made it difficult to maintain a scope and identify actionable opportunities through many of the tracks. 

We could have reacted to this complexity by avoiding it. That would mean treating discovery at its most basic and presenting the lived experiences of our target groups. But we knew that presenting “here’s what the people were experiencing,” although insightful, wouldn’t be helpful for addressing systemic challenges across an entire state. Our value was going to come from following problems up the chain. As best we could, we mapped how funding worked, how various agencies interacted, and how that created the outcomes we saw in our research. That process and those conversations with stakeholders informed what we were seeing on the ground.

Our second round of sprints was informed by the first. We realized that in order to make actionable recommendations, we were going to have to find new ways to focus our scope and discuss implementation. We learned — and got our partners to understand — that this was a broad set of challenges at multiple levels at the same time. This was a broad strategy project and there was no way we would be able to get as specific about a particular problem as we had assumed. We also re-tooled internally. We moved one team member to focus exclusively on strategy and stakeholder engagement across the topics. We expanded the roles and scope of our team subject matter experts to better inform research in behavioral health. We introduced more knowledge sharing across teams. This helped us see and act on the connection points across the different domains we were exploring. 

We never lost our ground-level approach, and we remained rooted in centering lived experience. Informing the strategic plan with lived experience was Washington Thriving’s goal. But we had to map the system and embrace strategic and policy advice to provide actionable recommendations. In fact, talking to real people about lived experiences made our strategic policy recommendations smarter and more powerful. It helped us better understand the system. Often, a strategic process will start from the top down — stakeholders around the table talking about their views of the problem. Our research-first approach mapped the system from the bottom up.

One of the early challenges was the sheer number of stakeholders. Our direct partners were Washington Thriving, who are responsible for a strategic plan. Dozens of agencies and organizations across Washington State would be directly involved in that plan. Even more would be affected by the plan. Our research quickly took us into each of those organizations' areas of influence. This kind of exploration can often be met with resistance or defensiveness. For some of these agencies, we were the first introduction they had to Washington Thriving and its work. We also knew from experience that creating recommendations without engaging with the agencies on their needs and capacity limits the chance for real change. 

We met with the relevant agencies early on to understand their perspectives. As we prepared to release our report we met again with every agency and team mentioned to socialize recommendations and talk about what it means. This added another layer of complexity, as each new stakeholder needed slightly different things out of our work (or didn’t know yet what they needed). It also added a lot of time. Traditional discovery sprints tend to socialize and onboard people by their very nature. Doing discovery this broadly meant that we had to invest a lot of additional time into getting people aligned.

Throughout the research we saw the same patterns pop up from topic to topic. The system was fragmented. Every challenge involved multiple state agencies with ambiguous responsibility. We heard from people across the system about the collective action problem this created and how it had stymied previous attempts to address each challenge. 

We ended up having to understand how the system operates at a statewide scale. That was the only way our insights could have the necessary impact. It was also how departments and organizations could implement the recommendations. We knew we needed to share that back to our partners.

It wasn’t a formal deliverable or even something they had asked for, but near the end of the project we sent Washington Thriving a memo outlining the systemwide challenges and practical recommendations for improving the alignment of the system. Internally, we called it our “how things work” memo. We had no idea how these observations and recommendations would be received. They brought us in to provide user research, and this was a critique and analysis of the entire system. 

The team at Washington Thriving received this very well. The memo crystallized some of what they had been thinking themselves. As a third party, we were also better positioned to say some of what they knew needed to be said. When we started, we thought we were doing discovery sprints. And we delivered insights from those sprints. But we also gave them a ground-up view of the problems with the system and frameworks for addressing those problems.

For each topic, we delivered a deck, a report, and a participant memo laying out our research methodology, key findings, and recommendations for Washington Thriving to consider in the strategic plan, as well as opportunities for state agencies to take action in the near term. We also shared our “how things work” memo mentioned above.

Washington Thriving is working with an overwhelming amount of data and inputs. We recommended a clear and well-researched hypothesis and framework about what to do next.

In our research, we developed strong connections in the Washington state behavioral health space. Multiple agencies and hospitals partnered with us and commended our work. Some stakeholders are already implementing changes or adjustments based on that work, before the strategic plan is even published. Now, the Washington Thriving strategic plan is set up to integrate lived experience and ground-level realities, along with statewide data and stakeholder priorities.