Backbone Organization or Backbone Function?

by Jeff Edmondson on December 3, 2013

As communities across the country engage in collective impact generally and the work of building cradle to career civic infrastructure specifically, one of the first issues that always comes up is stress and tension around the selection of a “backbone organization.”  This is a core component of the work that Kania and Kramer lifted up in the original Stanford Social Innovation Review article.    While we could not agree more that there is a need for the concept that is described, the power struggles that often occur among the various entities that want to play this role often get in the way of progress and can derail an effort early on as historical issues of turf quickly emerge.

We are learning that the concept of a single backbone organization may very well be flawed.  This has become clear as we worked with an array of different communities looking to navigate the often contentious discussions around where the organization should land.  Most of time, the different organizations engaged in these discussions locally bring very different skills, interests, and competencies to the table.  Sometimes they have a unique leader who could play the central executive director or “cat herder” role effectively.  Others, they have the capacity to do the critical data analytics.  Still other times they may really be interested in moving one or two outcomes, say early childhood and early grade reading alone, not the entire continuum of outcomes.

This has led us to the conclusion that what is likely needed is a “backbone function” not a “backbone organization.”  This may simply sound like semantics, but it leads to a completely different way to approaching the staffing of collective impact work.  This shift helps us to see that this work is not about a central power center that gets created in a traditional hierarchical paradigm, but instead is about a set of shared roles that need to be played as we look to connect the dots instead of recreate the wheel.  These roles, which simply have to be played by a host of organizations since no one new organization can lead collective impact work alone, include:

  • Ensuring there is a person who wakes up thinking about how best to act as a servant leader to a broad partnership to achieve a collective goal and move specific outcomes every day.
  • A core data analytics role that includes the development of an annual dashboard on critical community level outcomes and comprehensive data management systems, but even more importantly the building of local capacity to use data on a regular basis.
  • Facilitation of practitioners looking to take what they are learning from the analysis of local data to change how they serve their target population each and every day, building comprehensive action plans around what works to move a specific outcomes
  • Community mobilization work to get a diverse array of voices engaged in this work, building shared ownership for improvement and supporting practices that get results.
  • Convening investors so they begin to communicate about how to put resources behind what works and consider ways to incent the use of data for continuous improvement.

There are certainly other roles that emerge over time and need to be played, but this is a start.  And if we see that a host of organizations working in concert all can contribute to the overall backbone function any community needs to have played, it can and should reduce some of the power struggles that have emerged around this important piece of the work.

We have learned one additional lesson that deserves to be mentioned.  It is helpful, especially early in this work, to have all the key staff located in the same place even if they come from different organizations.  The importance of these staff sharing what they are learning on a daily basis, helps them practice the type of continuous improvement they are looking to promote across community partners.  The simple reality is there will be a need a fiscal agent and they have to sit somewhere.  We recommend communities not create a new 501c3 to house the staff since this work is primarily about leverage existing resources.  We term wherever they land as the “anchor entity”, but whatever it is called it need not cause conflict since it should become clear very quickly that there is joint ownership for the backbone function as a whole.

At our recent convening we had an outstanding plenary session with stories on how sites have “failed forward”.  This feels like an important example of us failing forward, learning and adapting at the national level, just as local leaders do this every day on-the-ground to achieve better results for children and families.  What do you think?

Jeff Edmondson About Jeff Edmondson
Jeff Edmondson is the Managing Director of StriveTogether, a subsidiary of KnowledgeWorks. StriveTogether is a national cradle-to-career initiative that brings together leaders in Pre-K-12 schools, higher education, business and industry, community organizations, government leaders, parents and other stakeholders who are committed to helping children succeed from birth through careers.

Related Posts Plugin for WordPress, Blogger...

{ 13 comments… read them below or add one }

Lucas Cioffi December 4, 2013 at 2:08 pm

You make a strong case here for backbone function vs. backbone organization. I’d expect that existing organizations (with other missions) sharing the backbone function will help bring those organizations together. Would you recommend that the specific roles for each organization are written down and agreed upon so that each can depend on the other? I would expect that without explicit agreements, organizations can have the best intentions of helping out, but when their core commitments get in the way, their commitments to shared backbone function will suffer, and once some organizations start pulling back, there can be a vicious cycle leading to everyone pulling back.

Also, from your experience, do you have any recommendations for how shared funding can get distributed in the “backbone function” situation as compared to the “backbone organization” situation?


Ann Mathieson December 4, 2013 at 3:09 pm

I couldn’t agree more with this analysis Jeff.
This has been our experience in Marin County.
We don’t have an organization that either wants to or meets the criteria to play the backbone role. We’ve created a “backbone staff” that is supporting the Partnership Council, driving the work, and will, when fully funded, live together in a space in one of our county-wide, well established organizations. We also have a fiscal agent who is supporting the backbone staff. Thanks for providing the words and rationale around this nuance.


Lori Hart December 4, 2013 at 5:21 pm

This is great Jeff – we struggle with the word organization as we tell people we’re not an organization, but rather a community partnership, but then use the words backbone organization when describing some of the functions we perform… “function” is more descriptive and helpful for us here in Waterbury, CT.


Steve Votaw December 4, 2013 at 7:32 pm

This is a great blog for discussion purposes. I am not sure there is an ideal backbone structure. Power struggles have been a fact of life within the education and human services fields for decades. These power struggles are seldom due to structural issues but are often driven by personalities. The success of the collective impact movement will be based on the willingness of key stakeholders to set aside their respective differences and egos to work in the best interests of children and students. Decentralizing backbone functions will not solve the power struggle issue. It is far more important to get the right people on the leadership team than the right organizations. The strength of the collective impact movement is to utilize data effectively and continuously improve. Honest dialog and commitment to continuous improvement by funders can help ease the power struggle issue. However, because of personalities, different structures may be needed in different communities. Functions of the backbone should be clear but it will be best to leave it up to each respective community to determine the appropriate structure.


Derran Wimer December 4, 2013 at 7:41 pm

We are an organization that solely functions as a backbone. Each of us wake up thinking about our work every day.

We think of (+speak of) ourselves as an “idea” not as a “place”. We have even toyed with the concept of having no “office” as it simply chews up budget. If we are doing our jobs well, we are out in the field much more than in the office.

Saying that, I support Jeff’s thoughts about function vs organization. However I deeply believe that success, whether in an organizational structure or in a functional/ distributed structure, is embedded in genuine, trustworthy, non competitive engagement with our educators to work “with” them on “on” them. When I talk with colleagues in other areas of the state and country who are working in this arena, this is the one barrier that seems to stop everything.


Jen Splansky Juster December 5, 2013 at 1:01 am

Great post, Jeff. Thanks for sharing these learnings. The research and work with communities that we at FSG have done also supports this perspective that collective impact initiatives must ensure that the backbone functions are being played by staff dedicated to the initiative and “where” the backbone sits is entirely context-specific. We have also seen backbone support sit within a broad range of organization types – from being shared across multiple organizations (Magnolia Place is a great example we mention in “Channeling Change: Making Collective Impact Work”), to sitting within other organizations such as funders, public agencies, or nonprofits. Your blog underscores the importance of adapting the collective impact approach to local circumstances and leveraging local strengths.

We have also been concerned with some of the misperceptions about the backbone that you reference, and jockeying for the role that has played out in some communities. In particular, I’d highlight your point that the backbone is not the power center for the initiative, but the servant leader focused on guiding the vision and strategy of the initiative, and the other functions you outline. Some research we have done and shared in blogs on SSIR highlights additional characteristics of the servant leader guiding the backbone and larger initiative. Look forward to surfacing and sharing more of these learnings from across the field on the Collective Impact Forum.


Scott Jones December 5, 2013 at 3:53 pm

Jeff….I find myself standing firmly on BOTH sides of this argument! In a smaller community such as Red Wing organizations that would collectively supply human resources to the backbone have limited ability to do so. This can be a barrier to this disaggregated approach. On the other hand, it is true that no one organization can do it all. We are a bit of a hybrid.

You mention your recommendation to not become a 501(c)3. We are seeing that becoming an issue in our fundraising as it is difficult to describe that the “anchor entity” is the fiscal agent but Every Hand Joined is not where their funding should go. Are others seeing this?



Leslie Maloney December 11, 2013 at 4:41 pm

This concept makes a lot of sense, even for us in Cincinnati where we are blessed to have “Strive” at Knowledgeworks be the backbone “organization.” However, as we know too well, this has actually caused a lot of confusion here locally about who is Strive, who is Strive staff versus the Strive Partnership, etc. I actually think that creating the backbone function with a group of individuals from a number of different organizations, each bringing their expertise to the table, is very reflective of the collective impact work we are trying to do and is easier to communicate to the community what this anchor entity is and the role it plays in supporting the work. This truly is neither semantics or a nuance but a very real concept that requires utmost transparency.


Jeff Edmondson December 11, 2013 at 5:41 pm

Love the perspectives being offered up here. Would invite others to join. I would agree whole heartedly that there is no one size fits all answer. There definitely does need to be people waking up thinking about this every day as Darren notes so accurately. If these folks anchor the work, it seems that part of what they do is align with partners to help fill out the full complement of roles needed for the backbone function. And to the question from Lucas, if this is indeed how it plays out in a community with multiple entities playign the overall backbone function, documenting the roles various partners are playing is critical. Angelo Gonzalez in Albuquerque with Mission Graduate has done this really well. I know others have to. Any other examples people know of out there to add to the discussion?


Tim Brokopp January 29, 2014 at 8:38 pm

We at Amoveo Group are working at being the catalyst (back-bone organization) for collective impact initiatives in poor communities in the developing world. We have found just the opposite to be true. There is a strong need for a separate, neutral org that is not involved in any aspect of the project other then the facilitation of strategic collaboration accessibility for all parties involved. In fact we have found that organizations are much more willing to participate when there is a separate backbone that is mediating (for lack of better terminology).


Jan Garbett February 3, 2014 at 3:44 am

In you MSNBC interview it was observed that a collective impact model could be applied to other social issues. We want to do that. Using data in your own back yard was one of 3 key things you didn’t have time to share in that interview. Would the other two items relate to backbone function. I’m intrigued that you could do this without forming a 501c3. Could you elaborate on that? In the Stanford Social Innovation Review Collective Impact by John Kania & Mark Kramer Winter 2011 it claimed you functioned as the backbone with a $1.5 M annual budget increasing effectiveness of other organizations with combined budgets of $7B. Does this blogs references to a waning backbone model apply to STRIVE or other backbones you have analyzed?


Jan Garbett February 3, 2014 at 3:51 am

In your MSNBC interview it was observed that a collective impact model could be applied to other social issues. What you propose is very interesting and could be so full of promise for strapped non-profits. We are exploring how we could be a backbone for an array of businesses and organization that are concerned about protecting children from pornography. On the MSNBC interview you encouraged the use of data in your own back yard but didn’t have time to share two other items you claimed essential. Do either or both of the other two items relate to backbone function?

I’m intrigued that you suggest forming a backbone without making it a 501c3. Please elaborate on what that looks like?

Finally, in the Stanford Social Innovation Review Collective Impact by John Kania & Mark Kramer Winter 2011 it claimed you effectively functioned as a backbone with a $1.5 M annual budget increasing effectiveness of other organizations with combined budgets of $7B. Is the reference to a single backbone model being flawed apply to STRIVE or other backbones you have analyzed?


Bill Barberg May 12, 2014 at 8:29 pm

As most of the comments suggest, there is no one-size-fits-all approach to providing the Backbone Support for Collective Impact. I very much agree with Jeff’s primary point/suggestion that communities should think about “backbone support” for Collective Impact rather than thinking about “the backbone organization.” It may be that there is a new or existing organization that can be a primary provider of that backbone support, but it seems wise to embrace the concept of distributing the backbone function to multiple organizations to minimize the power struggles that can otherwise hinder effective collaboration.
It is unfortunate if a community has multiple, competing backbone organizations—especially if those organizations are using different techniques, languages, and approaches. Even if organizations are leading on different issues—such as educational success, chronic disease prevention, teen pregnancy, or crime—there are big benefits if they can use a shared language, approach, relationships and infrastructure, since these issues are very inter-related and often involve many people in the population and network of community organizations.
In our experience, the main reason most organizations jockey for position to be the backbone organization, resist embracing the backbone leadership of other organizations or create computing backbone organizations is that they are concerned about funding or power. Emerging Collective Impact practices can greatly reduce those concerns and simplify the journey to success.
If one of several competing non-profit becomes the backbone organization and they rely on typical approaches to coordinating the work of community organizations, then they will typically use funding for backbone support to hire staff, expand offices and build relationships that make them stronger. This can give them a relative advantage in getting future funding. The other organizations may be weakened with regard to their perceived power and ability to get future funding. And, the greater the success of the Collective Impact effort, the more the partnering organizations can feel they are losing ground to the backbone organization. No wonder there is resistance to collaboration!
One way to minimize this situation is to rely on “digital backbone” technologies to play a significant role in managing the information and coordination to achieve all the other conditions of Collective Impact. When the investment in backbone support is in shared on-line infrastructure and training on a shared approach to community collaboration, then it doesn’t really matter who gets the funding—everyone benefits and experiences improvements in their impact and ability to get additional funding. It is much easier to have distributed backbone support and information sharing when whoever is providing backbone support is working on with the same tools and framework. And, instead of multiple, competing sets of language and techniques, it is much simpler for everyone to use a shared framework and toolset—regardless of the issue. To learn more about digital backbone technology, take a look at this blog article or this 4-minute video:


Leave a Comment

{ 1 trackback }

Previous post:

Next post: