In today’s Social Velocity interview, I’m talking with Isaac Castillo, Director of Outcomes, Assessment, and Learning at Venture Philanthropy Partners, where he leads VPP’s approach to data collection, data reporting, and outcome measurement.
Prior to coming to VPP, Isaac served as the Deputy Director for the DC Promise Neighborhood Initiative (DCPNI). At DCPNI, Isaac led efforts to improve outcomes in the Kenilworth-Parkside community in Ward 7 of the District of Columbia through the strategic coordination of programmatic solutions and research-based strategies. Prior to his time at DCPNI, Isaac served as a Senior Research Scientist at Child Trends where he worked with nonprofits throughout the United States on the development and modification of performance management systems and evaluation designs. In addition, Isaac was also the Director of Learning and Evaluation for the Latin American Youth Center (LAYC) where he led the organization’s evaluation and performance management work.
You can read interviews with other social change leaders here.
Nell: You have spent your career using data to improve the performance of the nonprofits for which you worked. Why do you think performance management is so important for nonprofits? Do you think all nonprofits should pursue performance management? When does it make sense and when doesn’t it?
Isaac: I believe that every nonprofit should pursue some form of performance management because they owe it to the clients they serve. Most nonprofits will assume that they are making a positive difference in people’s lives, but in the vast majority of cases they are just guessing. Using some form of performance management will allow every nonprofit organization to confirm this thinking and to identify areas that can and should be improved so that the next cohort of participants can get better services than the last.
Unfortunately, one of the greatest challenges preventing a nonprofit from implementing some form of performance management isn’t a lack of resources, expertise, or time. It is fear. The fear that they will find out that their work isn’t having a positive effect. This fear is what nonprofit leaders need to overcome, not for the benefit of themselves or their organization, but because they owe it to the clients they serve today and the clients they will serve in the future. I believe that every nonprofit should strive to serve tomorrow’s clients better than today’s clients, and one of the only ways to ensure that this happens is the sustained use of performance management.
The type of performance management that each nonprofit should pursue should vary by the size and scope of their work. At a minimum, small nonprofits should be tracking basic demographic and attendance information on their participants, and hopefully at least one meaningful output or outcome. Whether this occurs in a computerized system or in a spiral paper notebook is up to the nonprofit. But it doesn’t have to be costly, and it doesn’t take expertise. It only takes the will and desire to improve as a nonprofit.
Nell: In the nonprofits in which you’ve worked how have you been able to secure resources to fund performance management? What is the case you and your colleagues have made to funders and what do you think it will take to get more funders investing in performance management?
Isaac: Raising funding for performance management work usually takes a mix of several different strategies and approaches for potential and existing funders.
First, I strongly encourage nonprofits to include some percentage (1 to 5 percent – possibly more) of funding in each grant submission or proposal dedicated to supporting performance management and outcome measurement work. By placing this small percentage into each proposal, a nonprofit can begin to raise funds for internal evaluation and performance management activities. It may not seem like a lot, but it can add up, and eventually generate enough funds for a half-time or full-time position to support in-house performance management work.
Second, I also strongly encourage nonprofits to engage in regular ‘funder education’ – where a nonprofit proactively meets with their funders to have ongoing conversations about outcome measurement and evaluation. This allows both the funder and the nonprofit to come to agreement on measurement expectations and to ensure that both groups are focused on the same concepts. I often suggest that the first of these types of meetings focuses on each group’s definitions of three commonly misunderstood terms: outputs, outcomes, and impact.
Finally, I would recommend that the nonprofit and funder have an honest discussion regarding expectations of results and the funding necessary to support the related evaluation work. If a funder is expecting an random control trial (RCT) to be completed to determine ‘impact,’ then the nonprofit should be willing to push the funder to support a large investment to pay for a high quality evaluation. If the funder is only willing to support a small amount for outcome measurement, then the nonprofit should clearly articulate what is possible.
Nell: Ken Berger and Caroline Fiennes recently argued that we may have gone too far by asking nonprofits to produce research about their own outcomes. What’s your response to that argument?
Isaac: I fully support Ken and Caroline in their argument that most nonprofits should stay away from trying to produce impact research. The desire for ‘impact’ is something that has been (and continues to be) pushed unfairly (and without financial support) by the funding community.
I honestly think a lot of confusion in this space comes from inconsistent use and understanding of the term ‘impact’. The term ‘impact’ has a precise definition among researchers but is often used in a much broader context among funders, nonprofits, and the general public. In the research and evaluation world, impact is used to describe the effectiveness of a program while eliminating as many potential confounding factors as possible. That is why the use of random control trials (RCTs) is usually the cornerstone of impact research – RCTs are the easiest way to control for and eliminate confounding factors.
When most non-researchers use the term ‘impact’ however, they are usually just asking if the program or organization works and if it is making a difference for its intended service population. That is a much lower bar to set, and yet it is a critical distinction in this discussion. If you are thinking about ‘impact’ as a researcher, you will need a large amount of resources and expertise to determine ‘impact,’ which usually means completing one or more formal evaluations. If you are thinking about ‘impact’ in the more general sense and less strict way, then pursuing some form of performance management system will allow a nonprofit to determine if their efforts have been successful.
I do think every nonprofit should pursue some form of performance management to ensure that their work is having a positive effect as a complement to existing research that others have done. Relying only on the use of others’ research does not guarantee that a nonprofit will provide effective services and achieve positive outcomes. This type of research is a like a recipe – it shows what has worked in the past and provides a guide for the nonprofit – but a recipe can still be ruined with poor implementation or planning.
Every nonprofit has an obligation to the people they serve (and not to their funders) to ensure that their programming is having a positive effect (or at the very least not causing harm). Without some form of performance management system in place (even one that just uses paper and pencil), a nonprofit will never know if they have strayed too far from the recipe provided by previous research.
I also think there are a growing number of very sophisticated nonprofits that should be using AND producing research on effective programs. Every year, I see more and more nonprofits that hire talented and unbiased researchers dedicated to internal evaluation and outcome measurement work. These individuals are just as talented and unbiased as their colleagues working in traditional research and evaluation organizations. They can, and should, produce original research that can help inform the nonprofit field. The real challenge comes in nonprofit organizations finding the resources to support the hiring and retention of these individuals. Not every nonprofit will have the resources or capacity to hire one or more of these individuals – but those that do should absolutely be trying to produce original outcome and impact research to provide ‘recipes’ for effective programming that nonprofits with fewer resources can use in the future.
Nell: Your former organization, DC Promise Neighborhoods, is part of the national Promise Neighborhoods Initiative launched by the US Department of Education in 2010 and modeled after the famous Harlem Children’s Zone. How successful has this national replication of a successful local model been? Have you been able to replicate outcomes? And what hurdles, if any, have you and other replication sites found?
Isaac: I think that there has been some initial success among the Promise Neighborhoods. Part of the challenge that all the Promise Neighborhoods face is that the Harlem Children’s Zone did not achieve their success overnight. They have been working in Harlem for decades, so it would be unrealistic to believe that the Promise Neighborhoods would be able to create large scale change in a matter of a few years.
However, there are signs of progress across all of the Promise Neighborhoods. Each of the Promise Neighborhoods started to address a few outcomes with the initial round of funding, and these outcomes varied. Some focused on math and reading proficiency for students, some focused on obtaining medical homes for young children, and others sought to increase the amount of healthy food consumed by residents. In DC, we focused on improving school attendance.
I do think that most of the 12 Promise Neighborhood Implementation grantees were able to make progress on the outcomes they identified as initial focus areas. However, the very nature of the work (creating community level change) doesn’t lend itself to the rapid accomplishment of multiple outcomes in a short period of time. Each of the Promise Neighborhoods had to prioritize certain outcomes for their respective communities, and only several years later are they able to claim success and begin to identify the next set of outcomes to be addressed. So while certain outcomes haven’t necessarily been replicated across all the Promise Neighborhoods, that is due to the differences in priorities and community conditions rather than any problem with the model itself.
Photo Credit: Venture Philanthropy Partners