Posted Monday, November 24, 2014 at 7:33 pm

I recently met with the former head of a major pharmaceutical company to talk about improving the lives of kids. This well regarded executive was excited to invest a significant amount of his time, post his tenure at the helm of a Fortune 500 company, in leading an effort to address youth substance abuse.* I was energized by his passion and fervor for the issue, and his true desire to use his platform and experience as a leader to make a difference. But I was troubled by his orientation towards the leadership he felt was required. “We need to drive these evidenced based practices through the system, from top to bottom," he said. “We need to force communities to understand what’s good for them.” Uh, oh, I thought to myself. Yet one more well-intentioned, influential individual in society who wants to bring about change for the better, but just doesn’t get how social change happens.

The minute you use words like “drive” and “force” to describe your intended process for bringing about change I head the other direction quickly. We can’t force change in a system – or if we do, it’s likely to result in only temporary change. If we want to exercise leadership in bringing about social change, our leadership task is really about facilitating the conditions within which others can make progress towards the goal. The leadership work of social change requires an ability to catalyze collective leadership in others - a form of leadership that Peter Senge, Hal Hamilton and I refer to in a new Stanford Social Innovation Review article as “System Leadership.”

Nowhere in the social sector is the need for system leadership more apparent than in the collective impact efforts many of us are involved in supporting. By its very nature, collective impact eschews top-down hierarchical forms of leadership that may work in the corporate world (though, even in the corporate world, most executives are finding that top-down leadership works less and less as an effective means for bringing about organizational change). Indeed, one of the qualities of collective impact that enables communities to achieve progress at scale – across, in some cases, hundreds of organizations – is the fact that no one individual or organization is in charge of a collective impact effort.  

So what type of leadership does foster success in collaborative efforts such as collective impact? In our new Stanford Social Innovation Review article, The Dawn of System Leadership we describe three necessary capabilities:

  • The first is helping people see the larger system of which they are a part. Creating this understanding helps collaborating organizations in a collective effort jointly develop solutions not evident to any one individual organization, and to work together for the health of the whole system rather than just pursue symptomatic fixes to individual pieces.
  • The second involves fostering reflection and more generative conversations than can result in truly innovative solutions. Deep, shared reflection is a critical step to enabling groups of organizations to truly hear one another’s perspectives. It is an essential doorway for building trust where distrust has prevailed, and for fostering collective creativity.
  • The third capability centers on shifting the collective focus from reactive problem solving to co-creating the future. Change often starts with conditions that are undesirable, but artful system leaders help people move beyond just reacting to these problems to build­ing positive visions for the future.

A few years ago author and New York Times columnist David Bornstein described the need for collective impact this way:

When it comes to solving social problems, society often behaves like a drowning man whose arms and legs thrash about wildly in the water. We expend a great deal of energy, but because we don’t work together efficiently, we don’t necessarily move forward. (March 7, 2011, New York Times)

For collective impact efforts to move from thrashing to thriving requires leadership less like that described by the eager-for-change pharma exec, and more like the leadership examples Senge, Hamilton and I share in The Dawn of System Leadership. Folks like Molly Baldwin at Roca, a community based organization who is transforming how communities think about supporting troubled young men and boys, and Darcy Winslow formerly of Nike, who helped catalyze an industry sustainability revolution in the design of athletic shoes and apparel.

Baldwin and Winslow truly get how social change happens. What they also get – because they’ve lived it – is the developmental journey that system leaders must go through to become effective at this work. This is a journey that’s both inward and outward facing – and one that typically advances through the use of powerful tools that support system leadership.

As a collective impact practitioner, what other capabilities and tools have you found to be helpful in catalyzing collective leadership? What advice do you have for other Collective Impact Forum members who are attempting to develop their system leadership skills? We look forward to hearing from you.

Read The Dawn of System Leadership on

*Note: while this story is true, I’ve changed the issue here to protect anonymity. 


Alanna Hendren

backbone organization

The difference between 'collective impact' or non-profit organizations and corporate organizations is not only in the mission but also in the fact that corporations focus intensely on results (shareholder value, items sold, overall profits, etc) and collective impact organizations have notoriously focuessed too much on process.  The problem with many corporate investors in collective impact is that they often have 'visionary leadership' in the form of one person but wherever there is public accountability, this responsibility is distributed amongst many leaders.  This can be very exciting if they are all on the same page or an exercise in futility if they are not.  Human beings being human beings, we need to ensure that incentives for change are built into all processes so that people are reinforced for progressive behaviors and not expected to become sacrificial lambs for a cause.  In collective impact we are also usually trying to change a mind set amongst the general population, which takes time, hard work, partnerships and marketting efforts, unlike the corporate relationship which is generally between companies and their customers.  I am always amused at corporate executives who run fabulous profit-oriented businesses but get frustrated and confused when dealing with collective impact and the multitude of stakeholders.

Submitted by Alanna Hendren on Wed, 2014-12-03 19:21

John Kania

backbone organization, funder community of practice, funder of initiatives, technical assistance provider / consultant

Alanna, to be sure, leadership in collective impact efforts is a mindset shift for many, not just corporate folks.  As for incentives, I truly think the most compelling "incentive" is helping people see the full system and how their work is impacted, and impacts others.  This line of sight is rare in the social sector and effective CI efforts strive to acheive it. It's not the pure and only antidote.  But it's more powerful than most people will give it credit for.  And helping folks get this line of sight is one of key jobs of a system leader.

Submitted by John Kania on Thu, 2014-12-11 16:51

Arthur T. Himmelman

technical assistance provider / consultant

I am completely confident, even though Senge, Hamilton and you are waking to the "dawn of system leadership," that it does not mean you have slept through the creation of the massive amount of scholarly and professional leadership literature that has been created over many decades. I am also sure that you know well that this literature is based on countless studies and assessments of diverse and innovative leadership practices. Therefore, I do not understand why you think it is necessary to create new terminology that appears to be just another way of describing what this literature calls collaborative leadership. Collaborative leadership clearly requires the kinds of capabilities and characteristics that you associate with system leadership. For over 25 years, I have used my own definition of collaborative leadership in my community and systems change collaboration consulting practice, namely, "facilitating mutual enhancement among those working for a common purpose." Collaborative leaders know that mutuality is essential in all aspects of community and systems change, including mutual respect, learning, and accountability. And, if there is a common capacity required in all forms of leadership, I believe it is the ability to change a mass of critics into a critical mass :-)

Submitted by Arthur T. Himmelman on Thu, 2014-12-04 01:19

John Kania

backbone organization, funder community of practice, funder of initiatives, technical assistance provider / consultant

Arthur, thanks for pointing to the commonalities between collaborative leadership and system leadership.  Indeed both collaborative leadership and system leadership are about developing the skills to address complex problems by helping to create the conditions in which others - those who have or affect the problem - can learn, collectively, to make progress against the problem.  And I like your personal focus on mutuality.

In putting forth the concept of system leadership, we focus more deeply on the connections between individual development and collective leadership in the contexst of systemic change challenges.  Our hope is that this lens will assist those who wish to lead systems change in diagnosing the areas against which they should focus attention, as well as providing them with the tools to do so.

Submitted by John Kania on Thu, 2014-12-11 17:01

This informative article about a system leadership can itself be placed in an emerging system of collective social change that features a new’ kid on the block’: an interlocutor.  Looking across the international collaborative landscape points towards the emergence of a new species of relation-forming actor that ‘AGEs ‘collective efforts.  That is, they are optimized to Assemble multiple actors and technologies while Guiding interactions over time to ensure that gains from collective efforts become Embedded in institutions and societies’ way of life.

Globally, and aften involving ‘hosting’ by United Nations agencies, the collaboration terrain already contains large scale set-ups to ensure, for example, adequate nutrition for the world’s population; to anticipate and prevent the migration of diseases, such as malaria, as the world warms up; and to replenish the worlds’ natural capital, the oceans fish stock.  Alongside these arrangements, the landscape of assemblers and guides to multi-stakeholder collaboration is expanding to include entities with different names but a shared intention to better address the challenges of collective problem solving at multiple levels of social change.  Industry facilitators (IFs) are a diverse grouping of entities helping corporations and economic sectors scale up their ‘inclusive’ effects.  South Africa is experimenting with and learning from the work of Collective Impact Organisations (CIOs) to enhance sustainability through changes to the peri-urban ecology of Cape Town.  Backbone Organisations (BOs) are being established in the USA as a new scaffolding in the civic infrastructure tailored to the formation of Collective Impact Partnerships (CIPs) at many scales and directions of social action, for example to systemically tackle youth unemployment.  International Nongovernmental Organisations are active embryos for establishing social partnerships across borders that involve more than pairing with a particular business.  The World Wildlife Fund’s leadership of a consortium to rehabilitate and protect coral reefs in the Pacific is one example.  In addition, the world of philanthropy is set to expand with ‘meshwork’ foundations, skilled at interconnecting multiple types of organisations that combine many geographic locations, operating at numerous levels of change that are, for example, required to reduce CO2 emissions.

What appears to make them distinctive is two-fold.  First is being implicated, rather than being ‘neutral’ in the way collaboration is formed and proceeds – that is a process of interlocution.  The interlocutor role is to co-produce and live a co-responsibility for how collective action works and bears fruit.  In Peter Senge’s sense, they have Presence in social change.

Second is in the combination of attributes that they draw on and bring to bear.  Seven seem to reoccur.  (1) Coerced collaboration across multiple types of actors is seldom cost-effective.  Social partnerships arise because people want them to, not because they are legislated to exist.  Voluntarism matters.  (2) Consequently, interlocutors bring a ‘servant’ quality of leadership that exerts influence without formal authority, while treating conflict between parties as a given that needs to be made productive.  Differences in interests are assumed to be in play.  A harmony model of change is not relied on.  (3) Another is an ability to gain the trust of stakeholders on the one hand while engendering trust between them on the other.  (4) Awareness and analysis of the ‘problem system’ they are involved with and the need for scale to make change meaningful and not piecemeal.  (5) Perhaps more critical is sensitivity to the distribution of power and authority that will make or break partnership efforts, allied to realistic strategies which constructively deal with this political fact of life.  (6)  Another vital competence is the attribute of a polyglot, able to understand, translate and communicate across partners with their different jargons and vocabulary which may involve limited literacy and access to modern communication technologies.  (7) Finally, how their governance is constructed and works must provide adequate ‘sovereignty’ in the way that decisions are taken and implemented.  This attribute is typically co-determined by the conditions attached to the finance and other resources that partnering relies on.  Which raises the question of whether or not public resourcing and the politics involved can satisfy the conditions needed for these attributes to come about and endure.  Is private, philanthropic finance that lies behind many of the examples given, a necessity?


A general point from this view and experience is that unless the attributes of system leadership include explicitly include issues of power and politics they will be less well equipped to deal with the real life of bringing about change in society that is more ecologically sustainable and socially equitable.

Submitted by Alan Fowler on Thu, 2014-12-04 04:34

John Kania

backbone organization, funder community of practice, funder of initiatives, technical assistance provider / consultant

Alan, thanks for your perspectives.  I agree with your perspective that diagnosing and engaging with power and politics is critical to system leadership. 

Submitted by John Kania on Thu, 2014-12-11 17:03

Marci Ronik

technical assistance provider / consultant

John, thank you for your insights as always. We have been working in the area of large scale system change for quite a long time and have embraced Collective Impact and the necessary conditions as our "go to" framework. We attended the NextGen Eval conference last year and came away with innovative ideas, techniques, and new colleagues.

Question: We are embarking on a new system initiative that involves a group of diverse actors from local government, the judiciary, legislators, the business community, school personnel, juvenile justice, nonprofit providers, funders and more. We will be facilitating their kick-off retreat soon and are wondering if you could recommend some activities to get them started-there is a common agenda however there are many egos and "I know what's best" in the room. 

Thanks again for the great blog and article. Looking forward to my continuous learning process!! 

Best regards,

Marci Ronik (The Ronik-Radlauer Group, Inc.)

Submitted by Marci Ronik on Tue, 2014-12-09 08:36

John Kania

backbone organization, funder community of practice, funder of initiatives, technical assistance provider / consultant

Marci, glad you're finding the insights to be helpful. 

Regarding your "getting started" question, I find a really powerful tool (weilded effectively) is Appreciative Inquiry - a strengths based way of envisioning the future (at least that's the orientation I would put on AI for your purposes).  In the "Dawn of System Leadership" we describe how AI was used to kick off a contentious collective impact effort to reform the New York state juvenile justice system.  With great results.

There are many free on-line resources you can access on AI. 

Submitted by John Kania on Thu, 2014-12-11 17:09

Marci Ronik

technical assistance provider / consultant


Thank you for the reply. AI is an area that I am very familiar with and have numerous handbooks and books on the topic-we were looking at some AI tools (dream, destiny, etc.). Will let you know how it goes. An exercise that we have used in the past as an icebreaker is called "Getting To Know You, Getting To Know All About You." It is a judgment and reality exercise and the end result (or moral of the story) is that we really don't know all there is to know about each other if we don't take the time to ask the right questions. Oftentimes, we make judgments based on assumptions, stereotypes, prejudices, etc. When leading a CI initiative that is complex and contentious, we find that setting the stage in this way helps to break down some barriers and also offers some levity to the meeting. Thanks again for the reply!


Submitted by Marci Ronik on Thu, 2014-12-11 17:20

Arthur T. Himmelman

technical assistance provider / consultant

John -

Thanks very much for your thoughtful response to my question about how collaborative leadership and system leadership are different. I made my comment to emphasize the value of the work of others on collaborative leadership, academics, professionals, and countless numbers of people in communities, upon which your insights about system leadership may be based at least to some degree.

I agree with you that, in order to see effective demonstrations of innovative solutions to problems taken to scale, we have to find ways of changing large, complex systems. I think we may agree that the leadership to collaboratively move multiple and diverse organizations and stakeholders toward a common systems change purpose and goals requires the creation of a Senge "learning organization" partnership with reciprocal and reinforcing actions and mutually measurable progress and accountability for results.

I also appreciate Alan Fowler's reminder that governance, power, and authority are key elements in the collaborative change process. In my view, a central purpose of collaborative efforts, created for community and systems change, should be the transformation of existing power relations. This is because, in almost every case, exsisting power relations preserve the dominance of people who are not the intended beneficiaries, nor those most affected by, a partnership's decisions and actions.

To help address these issues, I have identified some characteristics of collaborative efforts that maintain (collaborative betterment) and those that transform (collaborative empowerment) existing power relations. I believe this kind of empowerment not only increases the ownership and sustainability of change in communities, but also contributes to our larger capacity to govern our society democratically by meaningfully sharing decision-making power.

Submitted by Arthur T. Himmelman on Thu, 2014-12-11 22:55

Hello John,

I was lookng for emergence in the Collect Impact model and found the last mention of it about 2 years ago.  That seemed dated.   You mentioned your article in the Embracing Emergence webinar, and I would like to find that article.  I have taken a number of MOOCs from the Santa Fe Institute and have been thinking abut from the perspective of education.  I had small non-profit delivering cross age tutoring with great results, but because the local foundations are focused on the major players, truely emergent programs need not apply. 

I will check the Aspen Institute to see if emergence is present in the forums. 

From the Common Pricipals of Comples Sustems, from the SF Inst. course, emergence was occuring in part because of the absence of central control.  That is different from how I see colllective impact being implemented.  This is fine for post-emergence, but it stiffles emergence.

Also, is it still possible to get a copy of the slides from the Embracing Emergence webinar?

Thank you for your time,

Edward C. Krug, Ph.D.


Submitted by Edward Krug on Thu, 2016-01-07 21:08