The Messy Work of Changing the World

Steve Johnson 643285 Unsplash E1547425109576

This is the third of three posts on the art and strategy of changing systems. The first post focused on two critical dimensions of the strategy of systems change: the kernel of change at a micro level that’s at the core of a larger shift, and how this kernel connects up to the higher levels of a complex system. The second post focused on two dimensions of the journey of systems change: how change unfolds over time, and how a “we” can be built with the cohesion, reach and sustainability required to drive change at scale. This post concludes the series by taking us into the heart of the messy, tactical, often seemingly intractable work necessary for the success of any strategy for systems change.

Identifying and addressing rate-limiting “how to” problems.

Strategy can only be as effective as the tactics addressing the problems that, at any given moment, constitute a strategy’s cutting edge. In our experience, it is relatively rare to see a combination of strategic acumen – breaking big changes into specific pivotal events, identifying leverage points at the right level in a system, defining what must be accomplished in this era of a broader arc of change, determining how to balance cohesion, reach and sustainability for the critical “we” – and disciplined focus on the tactics that matter most at the present moment. Part of the challenge of executing systems change work well is that once one does zoom in on the right tactics, they often don’t look like “systems change” problems, but like problems of an entirely different, operational kind.

Returning to the work on youth employment that I described in the first post, after the launch of the 100,000 Opportunities Coalition, it became clear that retention of opportunity youth was among the most critical issues to the approximately forty employers who had joined the Coalition, but that none of the employers had a scalable solution for the attrition issues so many of their peers were experiencing. Adding to the challenge, most of the employers believed or had the data to demonstrate that manager skills and behaviors were a particularly critical attrition driver – and one that is very difficult to change. The Rockefeller-Incandescent team didn’t have a “ready to roll out” solution, any more than the employers did. But we were crystal clear that we needed to answer three “how to” questions quickly:

(A) How to develop a framework for identifying the specific manager skills/behaviors that would move the needle on retention in a specific company’s context, and embody that in a pilot that would demonstrate compelling results both to that company’s senior management and to other employers in 100,000 Opportunities

(B) How to secure strong sponsorship from one or two companies in 100,000 Opportunities to drive this work and share learnings publicly

(C) How to develop the right channels and relationships to ensure that once (A) and (B) were solved, a solution would be adapted and scaled by others

(B) was essentially a problem of professional services sales, where the “sale” didn’t involve money changing hands but instead exchanged access and commitment to share results for a specific kind of targeted engagement. While it happens that for Incandescent as an actor, this was “right down the center of the plate,” for most actors in the youth employment space this would have been incredibly difficult – they’d been built to do many things, but persuading senior executives that there was a way to improve the effectiveness of their own managers was not in this repertoire. (C) was a very different kind of “how to” problem, which was partially solved through the work that the Schultz Family Foundation, Starbucks, The Rockefeller Foundation and others had done to build the 100,000 Opportunities Coalition. As it happened, the timing of this need coincided with the launch of the White House First Jobs Compact, in the final days of the Obama administration – which provided a major forum to share work that we were driving with HMS Host that they and we believed greatly moved the needle on (A). As is always the case in operations, “how to” problems at one level then translate down into “how to” problems at the next level of granularity. As we identified the role that the First Jobs Compact could play in solving for (C), we needed to figure out how to engage with the leaders driving the event to ensure that what we believed would be a pivotal topic for employers would be treated centrally and that the innovations we were advancing would be given significant focus.

While in this example, the pivotal “how to” questions were broadly in the “zone of competence” of The Rockefeller Foundation and Incandescent as members of a “core we,” that needn’t be the case. Aravind, which started as an eleven-bed hospital in Madurai, India, has become the world’s leading eye-care provider, measured in terms of volume of care (two thirds the volume of the entire UK national health system), phenomenal quality and provision of access to the poor. They’ve had immense impact at the systems level, building capabilities to address preventable causes of blindness in a wide range of countries. In order to achieve this, they’ve had to work from their initial competence in surgery to address “how to” problems that have ranged from how to establish clinics that would funnel patients from rural areas far from Madurai to their central hospital, how to find and train girls from villages to deliver the rigorous standard of nursing core to how Aravind achieves world-class quality at costs approximately 1% of leading institutions in the West, and even how to develop an FDA-approved manufacturing facility for ocular implants that were impossible to source from established manufacturers. Aravind didn’t choose these “how to” problems – their importance flowed from strategic decisions Aravind had made at a higher level. Commitment to the strategy meant the problems chose Aravind. Aravind couldn’t “stick to their knitting” and just deliver standard eye care based on an established model. To achieve their vision, they had to rise to the very specific, incredibly demanding requirements to solve each of these problems, or accept a rate-limiting constraint on achieving the impact to which they’d committed.

There’s no shared recipe for solving “how to” problems that would apply equally to Aravind’s challenge of how to engage elderly people in rural areas who need cataract surgery and to The Rockefeller Foundation’s challenge of how to create a proof point of how manager behaviors could impact opportunity youth retention in a Fortune 500 company. The underlying principle here is that for systems change strategy to be useful, it should, at any given moment, be translated into intense focus on not more than a handful or so of “how to” problems. These should be areas in which it is feasible to move the needle in a bounded period of time, and where moving the needle will drive a significant advance in fulfilling the broader strategy. Once these critical “how to” problems are identified, it is imperative to confront each problem in its own terms and determine what will effect the necessary result — and not to default to what one’s own organization already knows how to do.

Dealing with the inevitability of getting stuck

Driving systems change, by its nature, implies reaching beyond one’s grasp. If entrepreneurship is indeed, in Howard Stevenson’s classic formulation, "pursuit of opportunity without regard to resources currently controlled," then those who seek to change systems are the most extreme entrepreneurs. In such work, one should expect to frequently find oneself stuck in important ways, and shouldn’t expect getting unstuck to be easy. (See my post “On Being Stuck” for a broader exploration of this question.)

Any actor seeking to change a system will sooner or later confront an intractable problem: an impasse that stands in the way of achieving one’s most important goals, which doesn’t yield to one’s initial ideas of how to overcome it. Facing an intractable problem means acknowledging that one doesn’t have a good enough way to achieve the necessary result, one doesn’t know how to find a good enough answer, and simply turning away from the problem means sacrificing one’s ability to achieve an important goal.

These are adaptive challenges. Confronting an intractable problem, finding the path forward hinges on learning—as nothing one knows appears to hold the key to moving forward. One of the most influential strands of thinking in the current literature about systems change begins with this imperative to learn, and derives a paradigm for making change that focuses on emergence. John Kania and Mark Kramer’s 2013 Stanford Social Innovation Review piece “Embracing Emergence: How Collective Impact Addresses Complexity,” like arguments in this spirt, begins with how attempts to apply prescriptive solutions stall in the face of the complexity that characterizes most major social problems. The way out of this bind, as this story goes, is to embrace the paradigm of emergence: to operate in the organic way that natural systems do, rather than in the mechanical way programmatic solutions operate. Kania and Kramer write:

Collective impact works differently. The process and results of collective impact are emergent rather than predetermined, the necessary resources and innovations often already exist but have not yet been recognized, learning is continuous, and adoption happens simultaneously among many different organizations.

In this context, the process becomes the solution. “The rules for interaction from collective impact create an alignment within complex relationships and sets of activities which, when combined with shared intentionality, causes previously invisible solutions and resources to emerge.”


The model of collective impact that Kania and Kramer lay out can have immense power. I have no doubt that it should be more broadly applied in a range of contexts. My own framing of the “we” in the previous post of this series, while most deeply influenced by Chester Barnard’s classic work on organizational behavior in the 1930s, is certainly also indebted to the work on collective impact that Kania, Kramer and their colleagues at FSG have advanced. That said, there is great risk that belief in the powers of learning as an emergent process becomes an article of faith. Emergent learning is only as powerful as the specific paths the learners take to accumulate ideas and experiences, and to shape insight from this material.
Take, for instance, one of Kania and Kramer’s central examples in the SSIR article, the case of Communities that Care in Franklin County, Massachusetts, an initiative formed in 2003 to reduce teen substance abuse by 50 percent. As Kania and Kramer relate, Communities that Care formulated a key goal of improving the attitudes and practices of families, and an intervention to train select parents who would then spread what they learn to others. They measured results after three and six years – and found they’d made no impact. Kania and Kramer describe how the initiative then shifted to a learning approach that engaged a much larger set of actors, tested many messages, and ultimately found that messaging about the power of family dinners gained traction. Reorienting to advance this new opportunity, the initiative in fact was able to drive a measurable increase – 11% -- in the percentage of youth having dinner with their families in Franklin County and first the first time saw a 10 – 20% decline in specific risk factors relating to family conditions.

All this is impressive, all the more so when one considers that CTC faced an existential crisis and was able to build the momentum and credibility through this relatively modest success that positioned the organization to survive and move forward. With the onslaught of the opioid crisis in Northern New England, the value to the community in recent years of having an organization positioned to facilitate collective action has likely been even more critical than first envisioned.

Let’s rewind the clock to 2009. CTC is stuck, confronting not only the failure of their initial program, but the further lack of impact of the refinements they’d made after the first three-year evaluation to make a difference. They face a strategic question. They could place a bet on an emergent process, in the way they actually did, recognizing that results would be uncertain but that the possibilities for progress they’d be exposed to would be vastly greater. They could fold their hand, making an honorable decision that resources could more productively be allocated to addressing other issues in a community with many needs. They could revise assumptions closer in, maintaining their initial thesis and searching for more radical ways to realize this thesis. They could adopt any of a wide range of other approaches, differing on many dimensions: shift the goal itself or hold the goal fixed and simply shift the approach; concentrate bets and “swing for the fences” or diffuse bets broadly; navigate possibilities in an emergent way or through a structured, linear process of evaluation; focus on the community itself and its stakeholders or focus on the spectrum of solutions tried elsewhere; and so on. Where on this game board to play depends on a great many factors, ranging from the expected value of the programmatic solutions previously applied in other contexts, to the probability one estimates that a breakthrough can be achieved through different processes of discovery, to how one weighs progress today versus to building the asset of a cohesive “we” that can drive progress over time.

Looking at the case of impact hiring, when it became clear that many of the 100,000 Opportunities Coalition employers saw the retention of opportunity youth in entry-level jobs as a powerful leverage point, emergent learning didn’t stand out as the best and most direct way to solve that “how to” problem. This learning wasn’t, after all, emerging naturally from dialogues among the coalition members. By focusing on one corner of the system, with HMS Host, and moving quickly to generate a tangible intervention, we could move the needle much faster than an emergent approach would likely have achieved. As HMS Host, Walmart, and a range of others began to see traction with new approaches to equipping and motivating managers to retain opportunity youth effectively, emergent learning became a much more promising approach. Our joint team at The Rockefeller Foundation and Incandescent at that point brought FSG into the mix to facilitate an extended process for several employers to learn from one another in an emergent way, as each employer drove forward a set of experimental initiatives. Unlike in the Franklin County story, here emergent learning wasn’t so much a way of getting unstuck as a tool that became valuable after we’d made a focused bet on understanding how large employers could overcome being stuck on moving the needle on opportunity youth retention.

While there’s no one reliable way to get unstuck, particularly in the face of the kinds of intractable problems that those who venture into the terrain of system change must commit to confront, there are underlying principles to steer by. First, acknowledge being stuck, and get precise about just what kind of stuckness one faces in driving change. Second, form a clear, logical picture of what approach to learning and discovery is most promising to address this particular set of roadblocks. Learning and discovery will always be critical at these junctures; if one already knew the relevant solutions, one wouldn’t be stuck. Acknowledging that this approach to learning and discovery often won’t take the form of a programmatic plan, an emergent path should be pursued with passion and with discipline no less exacting than the discipline of executing a known roadmap. Recognize that whatever early advances might emerge are likely to be insufficient to achieve the “big goal” – and keep alive the tension between what one is learning how to do and what will ultimately be required. Celebrate advances, even small ones, but don’t prematurely declare victory.

Recognizing that one has become stuck and dealing with it well builds the resolution and cohesion among many actors required to drive change in a system over the long arc. Part of what is powerful about the Franklin County story is the sustained commitment of many actors to an effort now late in its second decade. The progress that Kania and Kramer write about in “Embracing Emergence” built both positive momentum and social capital without which the CTC’s work might not have reached the end of its road. The substance use numbers in the original five districts that were the focus of CTC’s work didn’t fall rapidly in the first handful of years after the choice to embrace emergence. Alcohol use and binge drinking among 8th through 12th graders fell a little, with marijuana and smoking by and large flat. The 2014 numbers trended in the wrong direction. The effort might have faltered, with actors impatient at the real but slow progress. It didn’t falter, and measures in 2017 showed that at least the pattern of teen drinking – both alcohol use and binge drinking – had fallen by a little more than the goal of one half set so many years before. Nearly all efforts at systems change get stuck, and nearly all face crises significant enough that they threaten to erode all the progress that’s been made. Some of the successful ones pursue a lateral, emergent path to get unstuck. Others analytically narrow down and then concentrate their bets in the way that the marriage equality campaign did in Massachusetts. All of the successful ones confront being stuck without illusions, and use these moments to learn how to advance their goals in a new way.

In this three-post series, I’ve laid out a way of navigating both the strategy and the art of systems change. The work we’ve begun and envision deepening in the months and years to come is to develop a body of thinking and a body of practice about how to navigate these different aspects of systems change across a wide array of contexts.

We hope to emulate the way that the field of business strategy advanced. BCG, for instance, built a set of analytical methods and a body of knowledge and experience behind the conceptual foundations of the experience curve and the growth-share matrix that made these strategic concepts practical in a wide range of contexts. We don’t talk much about these fundamental ideas now not because they’re no longer relevant, but because they’re so deeply ingrained in how we think that we can take these concepts for granted, and focus our attention elsewhere – much like a commuter could focus her thought on some difficult problem while driving a known route on what feels like auto-pilot.

Just as with business strategy, mastery of concepts and analytical methods isn’t a substitute for the indispensable arts of leadership, of entrepreneurial building, of dealmaking, and so on. The elegance of strategy isn’t a substitute for the messy work of building companies. Rather, just as the artist’s mastery of technique is the foundation on which her creative work can be built, the business leader’s grasp of the concepts and methods of strategy unlocks creative potential and grounds the human work of judgment. This is all the more true in the dramatically more complex and difficult work of changing systems. We show respect for the gravity of the causes that call us to the work of changing systems by approaching this work with the best disciplines we can bring to bear, knowing all the while that they’ll shine a light only a small part of the way to our journey’s end.

Niko Canner Profile Cropped
Niko Canner

Niko Canner founded Incandescent in 2013. His work spans the firm’s three major areas of focus: serving as a thought partner to leaders of large enterprises on strategy, organization and innovation; advising founders on the development of their ventures; and partnering with foundations and non-profits engaged in systems change.

View Niko's profile


The Practical Leader's Guide to Culture


The Art and Strategy of Changing Systems: Part 2