In Focus

Making an Impact, Collectively

Children break ground on the Rainbow de Colores playground, a community revitalization project undertaken in eastern North Philadelphia by the Sustainable Communities Initiative, led by the Local Initiatives Support Corporation. The playground—once unsafe and underutilized—has been a catalyst for neighborhood change.

Collective Impact, or CI, systematizes an approach to cross-sector collaboration—an idea for which Metis has long advocated. Now the idea is gaining currency in a variety of areas where people are seeking solutions to chronic problems and are working toward social reform. This type of work, in sectors such as education, housing, re-engagement of disconnected youth, and neighborhood revitalization, among others, goes back several decades. Collective Impact is the idea that multiple, diverse agencies and groups must work together—not in their separate silos—in pursuit of a shared goal to achieve effective and sustained change. What does Collective Impact mean, and is it a new idea or an old idea whose time has come? Here is a primer.

What Is Collective Impact?

The CI concept was coined by the Foundation Strategy Group (FSG), an organization that seeks solutions to major social problems. When FSG looked in depth at a number of successful social-change projects, they found that they shared certain essential components. John Kania, managing director, and Mark Kramer, founder and managing director, of FSG have defined these components as:

  • Common agenda
  • Shared measurement
  • Mutually reinforcing activities
  • Continuous communication
  • Support of a “backbone” organization

To Metis, these components bear strong similarities to a number of elements embedded within urban-reform programs of the past, several of which still exist and in which we are involved. As Melody Barnes, former director of the White House Domestic Policy Council, says, “Collective Impact is not necessarily new, but it’s an idea whose time has come.”

Collective Impact takes institutional collaboration several steps forward, formalizing the relationships among the participants, their goals, and respective duties. It also centralizes the key responsibilities with a backbone organization, so no one is left wondering who is doing what. A key tenet of CI is collective and continuous learning, which allows new solutions to emerge during the course of a project. In fact, CI embraces ambiguity and sees solutions as emergent rather than predetermined.

A characteristic example of Collective Impact can be seen in the United Way–supported multi-site Strive Network, which was begun in Cincinnati, Ohio, and northern Kentucky to promote student achievement. The Strive Network has brought together leaders not only from education but from the nonprofit, philanthropic, business, and civic communities to develop cradle-to-career supports for young people. The first five years showed very promising results in students’ academic success and system-wide improvements in education.

What Are Some Examples of Collective Impact Projects—Then and Now?

Over the years, Metis has worked with a number of large community collaboratives that, in our view, would fit the criteria for Collective Impact:

  • A precursor to many of today’s community collaborations, the Annie E. Casey Foundation’s five-city New Futures Initiative, begun in 1988, acknowledged that improving the public school system was not an issue only for schools but for the whole community, and especially for child-serving agencies. The initiative identified strategies needed to address interrelated youth problems, such as dropping out of school, teen pregnancy, and unemployment, and established an evaluation framework with agreed-upon outcome indicators. It was participants’ willingness to share data that enabled the New Futures collaboratives to demonstrate their success.
  • Metis has been working since the mid-1990s with the Georgia Family Connection Partnership, which works in all 159 Georgia counties to improve conditions for children. Each county has a community collaborative (read: “backbone organization”) that monitors and promotes services to children. Participating groups, including school systems, the child welfare system, businesses, youth-services providers, and families, have a common agenda and are committed to a shared theory of change and a data-collection and measurement structure.
  • Metis also has worked since the 1990s with the Living Cities initiative, a collaborative of 22 foundations and financial institutions, which aims to improve the ways in which major citywide systems interact and respond to community needs. Today, in Newark, New Jersey, Metis is evaluating the progress of the Living Cities Integration Initiative, led by the Strong Healthy Communities Initiative. This “backbone” organization has brought multiple stakeholders to the table, including city government, local foundations, nonprofits, and even corporations.
  • In Philadelphia, Metis is evaluating the Local Initiatives Support Corporation (LISC) project known as the Sustainable Communities Initiative (SCI). SCI is taking aim at two low-income neighborhoods, engaging a wide range of community partners in seeking solutions to crime, inadequate educational opportunities, and poor health.

The federally funded Promise Neighborhoods initiative also calls for inter-agency collaboration to support young people in schools and through other community support channels. “Although the government does not dictate what the partnerships should look like, Collective Impact would be an appropriate framework for Promise Neighborhoods coalitions,” says Manuel Gutiérrez, Metis vice president and senior research scientist.

Mayor Dana Redd of Camden, New Jersey, talks with neighborhood children about their dreams and what they want to be when they grow up. She is working closely with the Center for Family Services and a broad coalition of community partners to ensure student success in school and to combat violence in the community.

Metis is helping the Promise Neighborhoods “backbone organization,” Center for Family Services, in Camden, New Jersey, to generate the data it needs across domains to demonstrate the program’s outcomes. Metis also assisted the Promise Neighborhoods grantee CAMBA, in the Flatbush neighborhood of Brooklyn, to establish a sample for a community survey and then to analyze the survey data. And, in Brooklyn’s Sunset Park neighborhood, Metis is providing broad technical assistance to Lutheran Medical Center, a Promise Neighborhoods initiative leader, to prepare to use the Efforts-to-Outcomes software system and track student indicators along the educational continuum.

On a smaller scale, the Tackling Youth Substance Abuse (TYSA) initiative on Staten Island, for which Metis provided research assistance, has pulled together a coalition of organizations, government agencies, and individuals to find ways to reduce alcohol abuse and misuse of prescription drugs among Staten Island youth. By uniting the various parties, from hospital-based staff to YMCA staff to school-based counselors, the initiative quickly noted a need for better screening and referral services. It has identified and is currently promoting the use of an evidence-based screening tool, which it believes could lead to earlier intervention and better referrals. TYSA has used the data that Metis collected on Staten Island residents’ attitudes, behaviors, and perceptions about youth substance abuse to inform its public-education campaigns, plan targeted interventions, and establish a baseline to track progress and impact. Kania and Kramer highlight this initiative as an example of the power of “collective seeing, learning, and doing.”

How Do You Evaluate a Collective Impact Project?

According to Metis President Stan Schneider, Collective Impact requires an integrated theory of change—a cross-agency logic model that is designed to funnel resources toward an agreed-upon set of outcomes. One of the best tools to develop a consensus for shared goals is a logic model—which makes explicit the theoretical pathways to expected outcomes. By hypothesizing the links between resources invested, activities, and outcomes—both near- and long-term—the logic model can become the framework for evaluation.

“A theory of change is a living document that changes as you learn,” says Schneider. “Metis’s tools and data systems can really help partnerships create the cycle of continual learning that CI requires. We get agencies to talk to each other and to communicate internally more effectively.”

The development of a theory of change brings into play the first three “conditions” of CI in Kania and Kramer’s rubric: the common agenda among participants, a shared measurement system, and agreement on mutually reinforcing activities. The other two conditions—continuous communication and the backbone organization—have been central to many of Metis’s evaluation projects. Metis could not agree more with Kania and Kramer’s assertion that continuous feedback (especially at the highest levels of the participating organizations) is a model of evaluation that is far superior to the retrospective and episodic evaluations that have been commonplace among nonprofits.

Georgia’s Family Connection Partnership (GAFCP) offers a good example of how a multi-component collaborative project can be evaluated. As one part of the evaluation of the 159-county Family Connections initiative, Metis, with its partners from Emstar Research and Georgia State University, identified the counties that had placed a priority on reducing teen pregnancy rates (one of the key, agreed-upon outcome indicators). Outcomes for these counties were compared with rigorously matched environments to see whether there were pregnancy-rate reductions that could be attributed to the interventions. The significant differences that were found provided strong and reliable evidence of the power of the collaborative to effect positive change.

“What is interesting,” Schneider says, “is that each collaborative consists of a range of organizations. The school system’s goal isn’t reducing teen pregnancy, nor is the housing or juvenile justice system’s, and so on. But what succeeded was combining all of the resources for the community into an agreed-upon set of activities focused on that positive outcome.” (For information on the specific strategies please see

What Are the Lessons Learned?

Stan Schneider says that he has learned a number of key lessons from three decades of work evaluating projects that fit the criteria for Collective Impact. One is that it is easier said than done for various types of organizations to align themselves around the same goal and to share their information—especially when these organizations have a history of competing for funding, and, in some cases, for populations to serve.

Another is that organizations must have a clear sense of their own missions and goals before they can productively participate in collaborative engagements with others. In some cases, especially for larger, multi-service organizations, it may be necessary to tear down some internal silos in order for the organization to achieve that clarity of purpose.

Yet another key lesson is that CI success requires a long-term commitment and stable leadership. One of the reasons that the Georgia Family Connection initiative has been so successful for more than 20 years, Schneider says, is that not only are there local backbone organizations in each of 159 counties, but there is a state-level backbone—the GAFCP—that keeps the process moving.

“There is sustaining power that comes from having a committed backbone organization that is always tweaking the theory of change, keeping relationships healthy, and providing technical assistance,” he says. “Although the initiative has worked better in some areas than others, certain key indicators, such as teen births, high school graduation rates, and instances of child abuse and neglect have shown substantial improvements over the past several years.”