Content uploaded by Sean Geobey
Author content
All content in this area was uploaded by Sean Geobey on Jul 24, 2016
Content may be subject to copyright.
http://jab.sagepub.com/
Behavioral Science
The Journal of Applied
http://jab.sagepub.com/content/early/2014/04/28/0021886314532945
The online version of this article can be found at:
DOI: 10.1177/0021886314532945
published online 29 April 2014Journal of Applied Behavioral Science
Frances Westley, Nino Antadze, Darcy J. Riddell, Kirsten Robinson and Sean Geobey
Nonprofit Organizations From Canada
Five Configurations for Scaling Up Social Innovation: Case Examples of
Published by:
http://www.sagepublications.com
On behalf of:
NTL Institute
can be found at:The Journal of Applied Behavioral ScienceAdditional services and information for
http://jab.sagepub.com/cgi/alertsEmail Alerts:
http://jab.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://jab.sagepub.com/content/early/2014/04/28/0021886314532945.refs.htmlCitations:
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
The Journal of Applied Behavioral Science
1 –27
© The Author(s) 2014
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0021886314532945
jabs.sagepub.com
Article
Five Configurations for
Scaling Up Social Innovation:
Case Examples of Nonprofit
Organizations From Canada
Frances Westley1, Nino Antadze1, Darcy J. Riddell1,
Kirsten Robinson1, and Sean Geobey1
Abstract
Why do so many social innovations fail to have a broad impact? Successful social
entrepreneurs and nonprofit organizations often “scale out” innovative solutions to
local problems in order to affect more communities or numbers of individuals. When
faced with institutional barriers, they are motivated to “scale up” their efforts to
challenge the broader institutional rules that created the problem. In doing so, they
must reorient their own and their organizations’ strategies, becoming institutional
entrepreneurs in the process. This article proposes a contextual model of pathways
for system change consisting of five different configurations of key variables and
informed by qualitative interview data from selected nonprofit organizations.
The authors argue that the journey from social to institutional entrepreneurship
takes different configurations depending on the initial conditions of the innovative
initiatives. Despite an expressed desire to engage in system change, efforts are often
handicapped by the variables encountered during implementation.
Keywords
social innovation, pathways to system change, complexity, social entrepreneur,
institutional entrepreneur, nonprofit organization
The evident seriousness of today’s most pressing social problems adds momentum to
discussions around the concept of social innovation. Definitions of social innovation
vary, but most include both the creation of a product, process, or idea, and its diffusion.
1University of Waterloo, Waterloo, Ontario, Canada
Corresponding Author:
Nino Antadze, School of Environment, Enterprise and Development, University of Waterloo, 200
University Avenue West, Environment 3 (EV3), Waterloo, Ontario, N2L 3G1, Canada.
Email: nantadze@uwaterloo.ca
532945JABXXX10.1177/0021886314532945The Journal of Applied Behavioral ScienceWestley et al.
research-article2014
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
2 The Journal of Applied Behavioral Science
Much like technical innovation, social innovation arises when an individual or group
of individuals identifies a societal need and responds creatively with a novel solution.
Hence, Mulgan, Ali, Halkett, and Sanders (2007, p. 9) define social innovation as “the
development and implementation of new ideas (products, services and models) to
meet social needs,” and Mumford (2002, p. 253) claims that social innovation “refers
to the generation and implementation of new ideas about how people should organize
interpersonal activities, or social interactions, to meet one or more common goals.”
Implicit in these definitions is the idea that “going to scale” is simply a matter of dif-
fusion, that is, of more people hearing about and adopting the idea. Success is only
limited by demand.
Others, however, maintain that dynamics more complex than supply and demand
are at work when a social innovation moves into the mainstream. Their definitions of
social innovation stress the second aspect: the ways in which inventions, once created,
spread to affect the broader problem domain. Caulier-Grice, Mugan, and Vale (2008,
p. 45), for example, differentiate innovation from creativity and invention by explain-
ing that innovation is “more than improvement (which implies only incremental
change) and differs from creativity and invention (which are vital to innovation but
miss out the hard work of implementation and diffusion that makes promising ideas
useful).” Such “catalytic innovations” are able to address social issues with a funda-
mentally new approach, thus “creating scalable, sustainable, system-changing solu-
tions” (Christensen, Baumann, Ruggles, & Sadtler, 2006, p. 96).
In line with Christensen et al. (2006), we define social innovation as
a complex process of introducing new products, processes or programs that profoundly
change the basic routines, resource and authority flows, or beliefs of the social system in
which the innovation occurs. Such successful social innovations have durability and
broad impact. (Westley & Antadze, 2010, p. 2)
This definition clearly differentiates social innovation from social enterprise and
social entrepreneurship. A social enterprise is a profit-oriented, privately owned entity
that blends business interests with social ends (Westley & Antadze, 2010). Social
entrepreneurship is a human-centered concept that focuses on the qualities and skills
of the person who starts up a new organization or enterprise. It is a necessary but not a
sufficient ingredient of social innovation. Social entrepreneurs provide the new ideas
that set the process in motion, and may even be responsible for the spread of ideas
throughout multiple communities. For example, if funded by government or a founda-
tion, a program to distribute sleeping bags to the homeless may spread easily from one
city to the next, but it does not address the system dynamics that create homelessness
in the first place. Over time, as funds disappear, fewer sleeping bags will be distributed
but the homeless will remain. Institutional change is required if ideas addressing the
system dynamics that create social problems, such as homelessness, are to become
mainstream. And for this to occur and be durable—that is, for innovations to move
across scales and transform legal, economic, and policy regimes—it is essential to
understand the complex dynamics involved in system change, particularly because
failure is more common than success in these endeavors.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 3
Our analysis of pathways to system change draws on a complexity perspective,
which, we argue, may contribute to an understanding of social entrepreneurship. A
complexity perspective contrasts with “linear, formulaic, and mechanical models of
the world” (Patton, 2011, p. 123) and offers a paradigm characterized by notions of
emergence, self-organization, nonlinearity, uncertainty, adaptation, and multiple
scales. As the name indicates, a complexity perspective rests on the premise that the
phenomena and processes we encounter and study are complex (Nunn, 2007). As
such, they cannot be explored productively by isolating their parts, as the final out-
come is not produced by linear cause-and-effect relations of different components, but
rather by emergent and therefore unpredictable dynamics (Moore, Westley, & Nicholls,
2012; Pundir, Ganapathy, & Sambandam, 2007). The emergent behavior leads a sys-
tem toward self-organization, that is, creation of a certain order in response to altera-
tions in the environment (Pundir et al., 2007). Consequently, self-organizing systems
have the ability to adapt to their changing landscape (Nunn, 2007). Our understanding
of social innovation stems from complex systems theory, and therefore we view it as
an emergent, disruptive, and largely unpredictable process.
As Nunn (2007) notes, “a large part of complexity theory can be stated in only four
words: sensitivity to initial conditions. This is a compact way of saying that complex
systems are nonlinear, inherently unpredictable, and dependent on history” (p. 99).
The sensitivity to initial conditions is of particular importance to this article, as our
studies reveal the central role of initial conditions in determining the scaling-up strate-
gies of organizations. Yet another notion from complexity theory that is pivotal for our
discussion is the notion of scale and the role of cross-scale dynamics. As an alternative
to the more broadly accepted linear and equilibrium-based models of social entrepre-
neurship (Goldstein & Hazy, 2008), a complexity perspective offers frameworks for
understanding the patterns that emerge at higher levels of scale, which are of crucial
significance to the social entrepreneur, and not easy to predict (Goldstein, Hazy, &
Silberstang, 2008).
The Role of Cross-Scale Interactions in Successful Social
Innovation
Recent work concerning the nature and dynamics of social innovation addresses the
issue of scaling up new ideas and approaches in order to make a durable and profound
change (Moore & Westley, 2011; Westley & Antadze, 2010). Central to this discussion
is the premise that high-impact change “demands innovation across multiple scales”
(Westley et al., 2011, p. 767). At the micro scale, the invention or idea is initiated by
individuals or groups; at the meso scale, the innovation or novelty is incorporated into
a problem domain; at the macro scale, large institutions are transformed (Westley et
al., 2011). Cross-scale processes may explain the occurrence of sudden transforma-
tions and change.
Considerable work is being done on understanding the dynamics of cross-scale
interaction in complex systems and how this relates to successful innovation. A group
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
4 The Journal of Applied Behavioral Science
of scientists working in the Netherlands has produced a rich body of research focused
on the capacity of that country to develop innovative solutions to climate change.
Called multilevel perspective, it prompts researchers to distinguish and analyze three
conceptual levels: niche innovations, sociotechnical regimes, and sociotechnical land-
scapes (Geels & Schot, 2007). Viewed using multilevel perspective, transitions are the
result of interactions among these three levels, with change occurring through niche
innovations, through pressure on the regime from changes initiated at the landscape
level, or from windows of opportunity for niche innovations presented by destabiliza-
tion at the regime level (Geels & Schot, 2007; Nill & Kemp, 2009).
Resilience theory is another branch of research concerned with complex social–
ecological change. Gunderson and Holling (2002) describe cross-scale dynamics in
ecological systems and name this dynamic panarchy. The panarchy model can be
applied to social systems to explain how novelty at a lower level can result in “revolt”
at higher levels, and how restrictions of novelty at a higher level may lead to the pro-
cess of “remembrance” at lower levels (Westley & Antadze, 2010). With a few notable
exceptions, however, neither resilience theorists nor the multilevel perspective allows
for agency by individuals, organizations, or groups who act to stimulate or support
cross-scale interactions. Although a broad system perspective helps us understand how
change can occur without revolution or even a broad social movement, it does not
provide guidance on how to design successful strategies for change.
As we define it, “scaling up” refers to identifying opportunities and barriers at
broad institutional scales, with the goal of changing the system that created the social
problem in the first place. Most of the relevant literature uses this term to refer to an
organization’s efforts to replicate and disseminate its programs, products, ideas, or
innovative approaches (Dees, Anderson, & Wei-Skillern, 2004; Mulgan et al., 2007;
Wei-Skillern & Anderson, 2003). We label this kind of replication “scaling out,”
defined as the organization attempting to affect more people and cover a larger geo-
graphic area; this allows us to reserve the term scaling up for situations where an
organization aims to affect everybody who is in need of the social innovation they
offer, or to address the larger institutional roots of a problem. This conception of scal-
ing up is related to our definition of social innovation, mentioned above (Westley &
Antadze, 2010).
Not all social innovations are intended to be scaled either out or up. Different initia-
tives may take different trajectories, some thriving on a local scale without any imper-
ative to spread further. Innovative initiatives of this kind strengthen the existing system
by making it more resilient; however, they do not challenge it. For example, the
Working Centre (www.theworkingcentre.org) in Kitchener, Canada offers a number of
services to homeless or vulnerable people living in the city. They remain quite suc-
cessful on the local level by providing new products and services to their target popu-
lation in a chosen locality. Another example is Santropol Roulant (http://
santropolroulant.org/), a Montreal-based nonprofit organization that has built an inter-
generational food security program from a successful “meals on wheels” initiative, but
has declined to expand to other communities despite winning wide media attention in
Canada (Westley & Antadze, 2010). Other social entrepreneurs, however, seek to meet
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 5
a broader demand by scaling out to other communities, disseminating their innova-
tions to affect more people. Among these a further subset come to realize that without
a deeper system change, their ideas and initiatives will never have the desired impact.
This requires new strategies addressed to changing the system that holds the problem
in place. These strategies are often emergent and path dependent—they are shaped as
well as inhibited by the character of the innovation itself.
Although it may seem obvious that a high impact change requires cross-scale inter-
actions, the practice reveals that most innovators are not able to achieve system-level
transformations. Therefore, a more nuanced study is needed to explain why nonprofits
and social entrepreneurs who see the need to scale up are often not successful in doing
so.
We propose a model of five distinct pathways of scaling up with the goal of stimu-
lating discussion around different strategies that social entrepreneurs and nonprofits
use to deepen their impact. The proposed pathways for scaling up social innovations
are shaped by, among other things, the initial conditions, the opportunities and barriers
encountered, and the motivation behind the decision to scale up in the first place, and
are informed by case studies of five different organizations. By arguing that there are
multiple promising approaches to scaling social innovations, we hope to contribute to
filling the knowledge gap that Bloom and Chatterji (2008, p. 25) describe as a lack of
“conceptual clarity” about why some social enterprises are more successful in scaling
than the others.
Method
To study the selected organizations we used the qualitative comparative analysis
(QCA), using it as a research strategy rather than a research technique. The QCA was
introduced by Charles Ragin (1987) and was largely regarded as a comparative, case-
oriented approach.
The QCA aims to capture the complexity of a case while providing a certain level
of generalization (Rihoux & Lobe, 2009). It enables the researcher to examine the
complex causal relationships within each case, and thus to uncover its underlying pat-
terns or configuration (Young et al., 2006). As Ragin (1990, p. 68) explains, “the logic
of the case study is fundamentally configurational.” In his view, the interconnections
of different parts form a coherent whole within a given context. Transforming cases
into configurations implies viewing them as “a set of conditions leading to a given
outcome”1 (Rihoux & Lobe, 2009, p. 228). As Byrne (2009, p. 109) explains, “multi-
case comparative qualitative work is always configurational when it engages with
causes.”
For our case studies, we chose from among members of the Applied Dissemination
Group, representing 24 nonprofit organizations funded by the J. W. McConnell Family
Foundation on the basis of their initiatives to create social change (see Table 1). Since
2002, this group of Canadian organizations, led by social entrepreneurs, has been
brought together periodically by the Foundation to share experiences and learn from
one another and invited experts. Although many of the organizations initially were
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
6 The Journal of Applied Behavioral Science
interested in local impact, or in expanding the reach of their innovations in areas such
as education, community development and poverty alleviation, each in its own way
had come to realize that to accomplish its goals, certain barriers at the system level
needed to be addressed.
We were interested in learning what triggered the momentum for system change in
these organizations, and how they came to recognize the need for system change rather
than for primarily local solutions. With this focus in mind, we were invited to observe
three meetings of this group and engage in informal conversations, as well as conduct
interviews with the Foundation program officers and review secondary website mate-
rials. As a result of these informal but intense interactions with the group’s members
and program representatives, we converged on five distinct configurations under
which we felt the efforts of the 24 organizations could be grouped. These were the
“volcano” configuration, where the momentum for system change evolved from the
experiences of the organization’s members; the “beanstalk” configuration, where the
momentum for system change came from the frustration of the leader (and founder);
the “umbrella” configuration, where system-level goals were introduced and funded
from the start; the “LEGO” configuration, where the need for system change emerged
from the results of previous initiatives; and finally, the “gemstone” configuration,
where the awareness of the need for system change came from the outside in the form
of an invitation to participate in a larger endeavor. In an effort to understand each of
these configurations in greater detail, we then selected one organization from each
configuration to be explored in depth. Given that many of the 24 organizations had
features of more than one configuration, we used the knowledge we had gained to
choose those which were most representative of the categories under study.
Data collection for each case was carried out through personal, nonstructured inter-
views with the organization’s leader (and in most cases, founder), supplemented by
data from the organization’s website and field data from our meetings with the Applied
Dissemination Group. One or two individuals from each organization were inter-
viewed, and most were interviewed twice, with each interview lasting an average of an
hour and a half. The interviewees were selected based on the depth of their experience
within the organization and their proximity to the changes that led to a scaling-up
strategy. Consequently, the informants were either founders or leaders of
Table 1. Participant Organizations of the Applied Dissemination Group.
Centre for Children Committing
Offences
Community Health and Social Services
Network
L’Abri en Ville
L’Arche Canada Foundation Community Foundations of Canada Sierra Youth Coalition
L’Arche Canada Santropol Roulant YOUCAN–National office
Tamarack–An Institute for
Community Engagement
Engineers Without Borders POWER Camp National–Filles d’Action
PLAN Institute for Citizenship &
Disability
Caledon Institute of Social Policy Framework Foundation
Roots of Empathy Community Experience Initiative The Stop Community Food Centre
Meal Exchange Free the Children–Volunteer Now ArtsSmarts
Junior Undiscovered Math Prodigies Eva’s Initiatives Pine River Institute
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 7
the organization, or both. Although other employees also contributed to the changes,
leaders and founders were the primary drivers, introducing system-level goals and
shifting their organizations into new trajectories. Our intention was to understand the
undercurrents of these processes from the perspectives of those who envisioned and
orchestrated them. The interviewees were asked to share their views about their orga-
nization’s journey from its foundation to the present day (including changes in mis-
sion, structure, funders, etc.). They were asked to describe the conditions at the
moment when the need or possibility of “scaling up” was recognized, how the problem
was redefined, and what barriers and opportunities were encountered. Although these
were the guiding themes of the interviews, the questions were not structured and
largely followed on the previous responses. The interviews were mainly conducted in
the offices of the organizations so that researchers could observe the atmosphere and
the environment in which the interviewees worked.
Following the first round of interviews, the researchers discussed the results and
identified questions aimed at further clarification of the data, or at gaining additional
information about particular issues. This produced a more structured second round of
interviews, focused on topics pertinent to each case study. All the interviews related to
a given case study were conducted by the same person.
The data from the interviews was analyzed following these steps:
Step 1: Open Coding by Each Interviewer
Interviewers recorded and transcribed their own interviews. The interviewer then open
coded their interviews, without any prior discussion with the group about possible
coding categories.
Step 2: Developing General Coding Categories
After all the interviews were coded, the interviewers met to work out general coding
categories. Each interviewer presented the coding categories he or she had developed.
Discussion of these led to more refined coding categories, eventually totaling 19 broad
categories and a number of subcategories (see Table 2).
Step 3: Recoding of Interviews by a Noninterviewer
The developed coding system was used to recode all interviews, with researchers cod-
ing their own interviews as well as one conducted by another researcher, thereby
ensuring interrater reliability. The “line by line” coding helped us to detect the nuances
and to “open up the text and expose the thoughts, ideas, and meanings contained
therein” (Strauss & Corbin, 1998, p. 102). It deepened our understanding of the views
of our interviewees and the ways in which they comprehended their realities, and thus
minimized the personal biases of interviewers or the possible influence of predeter-
mined perceptions (Charmaz, 2000).
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
8
Table 2. Coding Categories and Subcategories.
1. Calling or purpose/origin of
initiative
4. Local focus 7. Applied dissemination group 11. Changing strategies
1.1 AD Intermediary role 4.1. Tension between local and
national focus
7.1 Provides emergent solutions 11.1 In response to resistance in
system
1.2 Visionary 4.1.1 Unified vs. diversified resource
base
7.2 Critical reflections with
peers
11.2 A result of learning by doing
1.3 Learning 4.1.2 Who gets credit for outcomes 7.3 Important for learning 11.2.1 Emergence
1.4 Conclusions made based
on the experience
4.1.3 Trust 7.3.1 Seeking patterns 11.2.2 Working on multiple scales
1.5 Misconception 4.1.4 Tension with national focus 7.3.2 Evaluation 11.3 Driven by leader realizations
2. Relationship to McConnell 4.2 Creation of learning community 7.4 Emotionally significant 12. Positive organizational culture
2.1 Huge early funder 4.3 Whole system engagement 7.5 Disappointment that it is
over/still need it
12.1 Inclusive and participatory =
innovation
2.2 Funding helps relationship
to other partners
4.4 Relationship to local partners 7.6 Failure to reach full potential
to reflect on system/social
change
12.2 Being part of a movement is
motivating
2.3 Trust issues 4.4.1 Disseminating success
X-partners
8. Relationship of organization to
system innovation
13. Dysfunctional organizational
culture
2.4 National focus 4.4.2 Perils of privileging individual
partners
8.1 Adapting and disseminating 13.1 High turnover
3. Resources 4.4.3 Importance of place-based
work
8.2 Feeding/animating networks 13.2 Burnout, personal cost
3.1 Constraints 5. System change 8.3 Reflecting back innovation of
others
14. Charismatic leadership/social
entrepreneurship = high
performance
3.1.1 Drives dependence on
national funder
5.1. Definition 8.4 Replication 15. Understanding/recognizing
complexity
3.1.2 Drive to diversity
funding base
5.2 System impact 8.4.1 Can’t replicate programs 16. Failure
(continued)
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
9
3.1.3 Inability to respond to
opportunities
5.2.1 Beliefs 8.4.2 Replicate principles and
ideas
16.1 Linked to Changing strategies
from programs to principles
3.1.3.1 Frustration/implosion 5.2.2 Framing of issue 8.5 Opening the system 16.1.1 Shifted national focus
3.1.3.2 Failure (to have system
impact)
5.2.3 Engagement practices 9. Changing mission 16.1.2 From one intermediary to
another
3.1.4 Limited human
resources
5.2.3.1 Enhanced collaboration (linked
to Broader Problem or Issue)
9.1 Linked to research 16.2 Opportunity for research and
turnaround
3.2. Funding issues 5.2.4 Policy change 9.2 Funding 16.3 Of system
3.2.3. Limited funding 5.2.5 Way of thinking 9.3 External and internal crisis 16.3.1 Opportunity for innovation
3.2.3.1 Need to be financially
self-sustainable
5.2.6 On different levels 9. 4 Local focus 16.4 Fear of
3.2.3.2 Risk of continuity of
Mission
5.3 System Resistance 9. 5 In tension with continuity of
mission
16.4.1 Personal responsibility for
success
3.3 Availability 5.3.2 Drives need to be self
sustaining
10. Continuity of mission 17. Success
3.3.1 Drives strategy 5.3.3. Political interests or political
obstacles
10.1 Leader control 17.1 Fidelity to correct process
3.3.2 Increased human
resources
5.3.4 Not an open or innovative
system
10.2 Continuity of process 17.3 Changing definitions as
strategies change
5.4 Bottom-up 10.3 Maintain the movement 17.4 Success (on the ground,
often tangible, measured with
numbers)
6. Organizational growth 10.4 Research to document and
evaluate activities, confirm
findings
18. Broader problem or issue
6.1 Tensions 19. Maintaining general principles
6.1.1 Linked to bureaucracy 19.1 While improving practice
6.1.1.1 Impedes capacity to reflect
6.1.2 Responsibility for stability
impedes innovation
Table 2. (continued)
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
10 The Journal of Applied Behavioral Science
Figure 1. Sample of data analysis—connecting coding categories and interpreting the
linkages for the Engineers Without Borders case study.
Step 4: Building Connections Into Configurations
After completing the open coding, we reassembled data to make more accurate con-
nections between categories and subcategories. As Strauss (1987) explains, at this
stage of data analysis “the analyst begins to build up a dense texture of relationships
around the ‘axis’ of the category being focused upon” (p. 64). In accordance with the
grounded theory approach (Corbin & Strauss, 2008), the analysis of the primary data
enabled us to see the emergent patterns in terms of different pathways for scaling up.
At this point, we used QCA as a meta-analysis tool to examine the selected case stud-
ies (Fiss, 2009). Comparisons of configurations among the cases helped us to identify
distinct patterns for scaling up as well as the elements shaping these patterns (e.g.,
approach to change, sources of strength, challenges). Figure 1 presents a sample of
data analysis that illustrates the working process of linking different categories and
interpreting these linkages for one of the case studies (Engineers Without Borders
[EWB]). The connections among the nodes revealed that internal learning and an
inclusive organizational culture are the core characteristics of EWB that not only char-
acterize its present work but also influenced the change in their strategy. More impor-
tantly, the connections between Nodes 12.1 and 11.1 revealed that moving to a more
centralized framework in order to pursue system-level goals may undermine the main
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 11
asset of the organization—its participatory and inclusive organizational culture. This
conclusion contributed to conceiving of the Volcano configuration, and particularly to
defining the risk that the organization may face in undertaking this pathway to scaling
up.
Thus, the delineation of distinct pathways that organizations took to scale up were
“derived from empirical evidence based on distinct but comparable case studies using
rigorous analytic procedures” (Young et al., 2006, p. 4). As the patterns were being
identified, we revisited the interview transcripts and excerpted the quotations that best
described the revealed configurations and their particular elements. Throughout the
research, the data collection and data analysis phases were iterative, each informing
the other.
After completing the preliminary analysis, we shared our findings with representa-
tives of the organizations studied. They easily recognized the presented configurations
and provided feedback and comments. This meeting served to confirm the accuracy of
our analysis and to help us refine our findings. It completed the stage of grounded
theory research, referred to as “theoretical sampling” and explained by Charmaz
(2000, p. 519) as “a pivotal part of the development of formal theory.”
Elements Shaping Configurations
Based on our coding of the interview data, five main elements shaping different con-
figurations emerged, and were refined using complexity perspectives. They can be
described as follows:
a. Approach to change is revealed in the way an organization perceives its goals
for change, and its vision of how institutions and structures could be altered to
respond to particular social needs. Often an organization’s approach to change
has been the basis of previous successful scaling-out strategies.
b. Strength refers to the special advantages of the organization’s chosen change
strategies.
c. Challenge refers to the difficulties inherent in the chosen change strategies that
may hinder a move toward tackling system-level goals.
d. Pathway for scaling up describes openings perceived by the organization for
moving from scaling out to scaling up, conditioned by their earlier strategies
and choices.
e. Risk refers to the inevitable downside associated with any chosen pathway for
scaling up.
These elements are present in each of the five generic configurations (Table 3) that
emerged from the case studies. It should be noted that none of the organizations had
arrived at transforming the system they were focused on. These are not so much exam-
ples of outcomes as descriptions of how and why innovative organizations shift their
focus from scaling out to scaling up, and an appreciation of some of the obstacles and
opportunities for success.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
12 The Journal of Applied Behavioral Science
Configurations and Their Pathways to Scaling Up
Configurations sometimes reveal the context for a scaling-up pathway, especially
where the organization began with a scaling-out strategy. At other times, the configu-
ration is equivalent to the scaling-up pathway itself, as in the cases where the organiza-
tion started out with a system-level change objective. In the following descriptions, we
will use the pathway terminology, as that is our ultimate interest. Selected quotations
from interviewees appear in italics.
The Volcano
The organization pursuing a “Volcano” pathway is full of internal energy; internal
interactions are dynamic and intense, and the learning process is ongoing. The organi-
zation needs to reach a “tipping point” to “erupt” and make a profound and system-
level change. The force behind the drive to scale up comes from experiential learning
that is fed back into the organization. However, the learning of individuals must con-
verge to maintain the momentum of the change effort.
Table 3. Five Configurations for Scaling Up Social Innovation.
Approach to
change Strength Challenge
Pathway for
scaling up Risk
Volcano Occurs from
learning and
experimentation
Inclusive and
participatory
organizational
culture
Defining
strategic focus
Centralization of
the strategy
Lose ability to
generate the
energy and
excitement within
the organization
Beanstalk Initiated by
a visionary
and implies
implementation
of their strong
vision
Consistency and
drive
Scarcity of
resources to
respond to
opportunities
Finding a patron
or venture
social capital
Leave behind the
original design
and some of the
energy around
the movement
Umbrella The initiating
organization
stimulates
emergence
through funding
Introduces
system-level
goal at an early
stage
Lack of
ownership,
poor
integration,
absence of a
visionary
Challenge the
concept of
partnership
and “think like
a movement”
Push partners
beyond their
comfort level
LEGO System change
starts with
community
change
The emergence
of new local
networks and
partnerships,
building on
existing assets
Connecting
place-based
strategy to
broader policy/
economic
change
Creation of
strategic
conversations
to consolidate
elements at a
higher level
Hinder active
dissemination of
principles and
ideas
Polishing
gemstones
Refining and
selling more of
a good product
(controlled
replication)
Gives credibility,
legitimacy, and
reputation
to the
organization
Short-term
managerial
thinking in
a complex
problem
domain
Potential
partnership
with a system-
focused
movement or
organization
Lead to a loss of
quality control
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 13
Constant internal learning was the key imperative for EWB, the organization cho-
sen as emblematic of this configuration. EWB was founded in 2000 by engineering
graduates George Roter and Parker Mitchell from the University of Waterloo. They
felt that the engineering profession could do much more to help tackle one of the larg-
est global problems—the extreme poverty in developing countries. With the energy
and enthusiasm of idealistic youth, Roter and Mitchell started to recruit volunteers.
They were ambitious about the size and scope of their organization—they wanted to
“send more volunteers to more villages in order to help more and more people.” By
2010, EWB had grown to 25 full-time staff members, 35,000 members, more than
2,500 volunteers in Canada, and more than 300 volunteers who have worked overseas
(EWB, 2010). By 2011, EWB had 37 individual chapters across the country (G. Roter,
personal communication, February 17, 2011).
The major strength of the organization has been its ability to build on the excite-
ment of its young volunteers. From the start, Mitchell and Roter encouraged EWB
members to solve problems and to bring the solutions back to the organization.
Everybody had a voice and was able to bring their questions and suggestions to the
table, creating constant internal learning:
Everybody is an owner, everybody is a decision-maker, everybody has a stake. Every piece
of information that you bring to the table is seen as very valuable. Everybody feels that.
The strength of such an organization is the ability to build on emergence: to generate
energy and excitement through an inclusive and participatory organizational culture
that is responsive to internal and external factors. EWB’s approach to change—con-
tinuous learning and experimentation—is a direct reflection of its organizational
culture.
This approach to change bore fruit locally in the intense engagement of young,
busy engineering students. The opportunity EWB offered was not only for meaningful
work and the chance to travel overseas but also the chance to influence a rapidly grow-
ing movement. With an already impressive number of volunteers and chapters, EWB
was present in many countries and sectors.
The sheer diversity of views and information led to internal discussions about sys-
tem change, as EWB members started to realize that their project-by-project approach
would not be sufficient to tackle large underlying problems, such as poverty. Projects
felt meaningless without understanding their contexts, or addressing the broader sys-
tem that would allow project benefits to be maintained over time. The EWB team
realized that in order to succeed they needed to tackle problems at much higher lev-
els—within organizations, governments, and international aid agencies. Gradually
they adopted a systems lens and were able to detect issues and problems that had not
been visible to them before:
There was a massive evolution. We incorporated this new understanding at every stage
and changed the organization, rather than saying that we have this thing that works well,
so let’s keep it.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
14 The Journal of Applied Behavioral Science
However, a key challenge remained: Even with so many volunteers, there were not
enough resources to act on all the ideas in the system:
Our ambitions are much larger than our resources. . . . We came to the hypothesis that
other organizations have defined the problem too narrowly from having not seen the
broader picture. We think all these pieces are important, why would we ignore them?
Well, we would ignore them because it deflects resources.
By 2010, it was obvious there was a need to build strategic focus at the system level
and to devote resources to that. Over the course of the year, EWB shifted to concen-
trate on five core areas (water sector in Malawi, Ministry of Food and Agriculture in
Ghana, Rural Planning Offices in Ghana, farmers unions in Burkina Faso, agricultural
sector between Malawi and Zambia) in order to channel their resources toward influ-
encing system change:
We need to narrow down to five projects, where we can reach the tipping point in terms
of the people and resources that we put against it. . . . It’s very complex.
The risk remains, however, that in choosing this pathway, EWB leaders may compro-
mise the key competence and resource of the organization—its excited and engaged
volunteers:
So there will be a change in the culture of the organization that will come with these
changes. If we want to have an impact, this is probably what we need to do, but that
undermines some of the traditional values of the organization where everything was
created at the bottom and came up to the top and then got sent back again.
EWB will need to find ways to allow significant input as more centralized frameworks
emerge.
The Beanstalk
We have labeled this pathway “Beanstalk” as it is about “climbing” up to the system
level without compromising the initially chosen vision and priorities. This pathway is
adopted by organizations with a history of persisting in their efforts despite all the dif-
ficulties faced along the way. Consistency and drive are strengths of this pathway. A
strong visionary sets the priorities and the direction, and continues to lead the organi-
zation throughout its journey. Therefore, a leader is a central figure in this
configuration.
The organization Junior Undiscovered Math Prodigies (JUMP Math) was chosen as
representative of this configuration. A nonprofit group dedicated to helping children
excel in math, JUMP Math was founded by Canadian mathematician, author, and play-
wright, John Mighton, whose own struggles with math at school made him wonder if
his ability was simply limited. In university, Mighton came across the story of Sylvia
Plath, who taught herself to write poetry. Inspired by her example, he came to believe
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 15
that anyone can learn math, and realized this belief himself by earning a PhD in math-
ematics. Over the years, Mighton volunteered in math programs to help students, with
very positive results, confirming his belief that kids have much larger potential than
they are given credit for. John Mighton’s commitment to this strong personal vision
was at the core of JUMP Math’s approach to change:
I actually believe that the root cause of many of our problems is in education. . . . I believe
that if we fixed that, it would change many of our problems with the environment, with
poverty, and so on.
As an organization, JUMP Math’s major strength was consistency and drive toward
attaining its goals and vision:
The general principles have remained relatively consistent.
As part of putting into action his belief in the importance and need to help students
excel in math, John decided to train some of his friends to be tutors. Soon JUMP Math
moved from offering a tutoring program to being invited into the classrooms. The
evidence poured in: Children learned better when they were together in a noncompeti-
tive and supportive environment. JUMP Math began to focus its energy on teaching
teachers and providing resources for them. Gradually a network of teachers who could
support, inform, and mentor other teachers was created. Teachers became the primary
base for the dissemination of JUMP Math’s ideas, actively volunteering to make a
contribution. The teachers’ network served as a forum for discussions and the exchange
of information and experience. Teachers were seen as major agents for change who
were able to reach out to the students and to other teachers. By 2010, at least 50,000
schoolchildren were served by the JUMP Math program, with about 50% growth per
year (J. Mighton, personal communication, January 13, 2010).
The outcomes of the JUMP Math program are also very positive. For example, the
results from Lambeth School in the United Kingdom show that for the group of stu-
dents who used the JUMP Math approach for 2 years, 60% performed at or above their
grade level, whereas before JUMP Math instruction began only 12% of the group
performed at or above their grade level (JUMP Math, 2009).
Initially, the idea was to help those children who were marginalized and often strug-
gled at school. Over time, Mighton and his colleagues realized that by reaching out to
more and more schoolchildren, they could raise the average standard in math. However,
it was not only about math. John believed that the academic success of the kids and
their future contributions to society were linked. By being better educated and aware
of their own potential, they were more likely to become active citizens able to make
informed decisions. He believed that academic success in math could “spill over
everywhere in their lives.”
However, in order to realize this system-level intention, JUMP Math had to tackle
significant challenges. Its dependence on funders for resources made the organization
vulnerable, and eventually it had insufficient financial and human resources to respond
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
16 The Journal of Applied Behavioral Science
to new opportunities. This was often named as a source of frustration in the
organization:
. . . the main barrier is money, because we are always struggling to get resources.
Finding a patron or a source of social venture capital may be a response to this chal-
lenge, but risks leaving behind the original design and some of the energy around the
movement:
They [teachers] don’t feel it is the same old business as usual. I would hate to lose that.
Also we gain a lot by being a movement and not just a business.
The Umbrella
In organizations that follow the “Umbrella” pathway, the “initiating” organization stim-
ulates emergence of an innovation by providing overarching funding. The strength of
the Umbrella pathway is that a system-level goal is introduced at an early stage, and
coordinated local work emerges from that (i.e., the initiative operates as an experiment
in system change). The early funding creates a protected space in which the initiative or
organization can grow and develop, allowing for the introduction of novel approaches
and the development of significant challenges to existing systems. However, as the
“umbrella” is pulled away and the relationships with local partners come to dominate,
the organization may have to reinvent itself (perhaps even shrinking and reformulating
its purpose) to ensure that system-change goals can be maintained and advanced.
ArtsSmarts was founded by the J. W. McConnell Family Foundation in 1998 with
the initial idea of “arts becom[ing] part of the curriculum and the system,” and with the
goal of achieving system change in education. The approach to change for ArtsSmarts
was the stimulation of emergence through funding by the initiating organization:
I think it [approach to change] is driven by high-level goals that are then implemented
differently in different places and by different partners.
Over a period of 10 years, through its local projects, ArtsSmarts reached more than
350,000 young people in 2,500 schools, involving 5,000 artists, 14,000 educators, and
thousands of community volunteers (ArtsSmarts, 2010).
The major strength of the organization was that it introduced system-level goals for
change at a very early stage. Equipped with these goals, Artsmarts drew insights about
its work from numerous local partner organizations, most of which engaged students
in arts-based work inside and outside the classroom. ArtsSmarts’ decentralized pro-
gram delivery model meant that they did not have direct control over the work of
partners, but also that they could learn from the many experiments in different con-
texts. Over time what they learned pointed to “certain key aspects that should be
implemented for all programs across the country.” As they developed a clearer concept
about what was successful, they developed metrics for evaluation:
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 17
If the arts became part of the curriculum and the education system, it would be taken on
and absorbed. . . . If all school systems were to adopt it . . . that would determine a success
at the provincial level.
However, as ArtsSmarts—the founding organization—began to withdraw from its
coordinating role to let the initiative stand on its own, significant challenges emerged.
Among these were the lack of ownership of the initiatives, poor integration, and the
absence of a visionary leader to synthesize and drive the overall strategy:
. . . so somebody else took ownership of this because it wasn’t really anyone’s. Nobody
owned it. The partners owned their own thing, but they certainly didn’t feel any ownership
of the national piece.
A possible alternative pathway for addressing the challenges of scaling up would be to
build on the strength of the organization, not through the concept of partnership, but
by exploring the power of “thinking like a movement”:
Local organizations involved from the beginning [needed] to change their concept of
their role from funding recipient to “community . . . developer.” My role needed to
change from (intermediary) funder to being a catalyst for change. Those are pretty key,
those are key relationship changing concepts that had to be put into place.
However, the organization may run the risk of pushing partners beyond their comfort
levels, as some partners do not have the capacity to move from funder to program
developer:
. . . from being a recipient to being a program developer, a lot of them are not developing
their own programs, they’ve actually just taken over our role as a funder. And they’re still
just a funder as opposed to a program developer, they don’t do the detail work, they just
fund it.
The LEGO
We named this pathway “LEGO” as it focuses on the bottom-up emergence of local
networks, partnerships, and collaborations to build on existing assets. A LEGO path-
way is inspired by the belief that system change starts with community change; there-
fore, connecting the different “pieces” at the community level is crucial for creating
the momentum for system change.
Communities are the major focus for Tamarack, An Institute for Community
Engagement, in Waterloo, Ontario. Tamarack works toward building vibrant and
engaged communities to solve major community challenges such as poverty.
Tamarack’s approach to change builds on community change to create broader system
change:
We started at the community level, addressing issues and problems.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
18 The Journal of Applied Behavioral Science
The major strength of the organization is that it facilitates the emergence of new local
networks and partnerships, building on existing community assets:
. . . we had an effect at a place-based level by affirming those who were already doing it
and encourage those who weren’t to do a comprehensive approach.
Currently Tamarack engages 100 communities in pan-Canadian learning communities
(Vibrant Communities Canada, 2014). Compared with Tamarack’s initial goal of mov-
ing 5,000 people out of poverty, by 2010 number of people whose lives have been
improved has reached 147,000 (P. Born, personal communication, January 7, 2010). In
this process, Tamarack’s president, Paul Born, plays a central role as an inspiring
speaker, visionary, and leader.
Although being successful at the community level, Tamarack experiences the chal-
lenge of connecting place-based strategies to their aspirations for broader policy and
economic change. The organization admits that it had not been able to succeed in mak-
ing changes at the national policy level. In addition, some areas of the country have
benefited more from Tamarack’s ideas and efforts than others:
It was in our minds that to be successful we wanted to change the world—and this went
beyond our contribution locally to how can what we do work elsewhere and what could
have bigger impact.
In an effort to overcome this challenge and realize the potential of the LEGO pathway
to scaling up, Tamarack may consider creating strategic conversations at a higher level
in order to consolidate and bring together the necessary elements to influence policy.
However, this runs the risk of diluting Tamarack’s active dissemination of its core
principles and ideas at the community level:
We decided that our model was to disseminate ideas, concepts and core principles rather
than programs. There are organizations that replicated by packaging and disseminating
their programs, but we did not do this. I call it “maple syruping” work—finding the
essence, the sweetness of this, and that is what you disseminate.
Polishing Gemstones
We labeled this pathway “Polishing Gemstones” as it emphasizes the refinement of an
innovative program or product focused on scaling out—replicating the program in dif-
ferent contexts. In order to do this well, great care is taken to ensure that the program
is systematized and can be replicated successfully. Quality is a primary focus for this
pathway, and sustained effort is directed at turning the program into a product that can
be marketed, supported by efficient business systems.
The Centre for Children Committing Offences (CCCO) was founded in 2001 in
Toronto to replicate and disseminate a program called Stop Now and Plan (SNAP).
SNAP was designed to help child offenders younger than 12 years stay in school, and
to change the way communities engage with high risk and behaviorally disruptive
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 19
children. The program was developed and refined over a decade, gaining international
attention for its evidence-based approach and positive impact. The approach to change
for CCCO was to broaden this impact by refining and selling more of this good prod-
uct (controlled replication):
In ten years it went from development, to licensing and then creating training modules.
. . . We started to look at what was our product. It was about getting the language, talking
about products, talking about dollars. . . . We needed to think long term about being
sustainable.
CCCO collaborated with communities, schools, and mental health agencies, focusing
on teaching self-control and problem solving to young offenders. As they worked to
build this into a successful enterprise model, CCCO focused on finding new markets
and building business systems, and at the same time refining the model for reliable
replication in different contexts and communities. The development of a strong busi-
ness model enabled the product to reach more communities, and the program impact
was ensured through attention to faithful implementation and quality control:
Product development is a huge challenge. I think how we were able to scale up though
. . . having a product that was scalable had to do with the fact that we were heavily
engaged in research.
SNAP’s effectiveness has been widely recognized, and to date, SNAP® licenses have
been issued to children’s mental health agencies, educational facilities, and other com-
munity and social service organizations across Canada, United States, and Europe
(Child Development Institute, 2010). Consequently, the major strength of the organi-
zation is that the demonstrated success and rate of adoption of its product gives CCCO
credibility, legitimacy, and reputation.
As its enterprise approach met with success, CCCO members began to reflect on
how to extend their innovation beyond simply controlled dissemination of a positive
product. At this point the organization faced the dilemma of having emphasized short-
term managerial thinking in a complex problem domain. Emphasis on the product and
on its replicability and scalability made it hard to imagine scaling-up possibilities:
We were so busy just making sure that this program was right, and working and not
causing more harm than good. We wanted to ensure we were developing it in the right
way. So we focused our attention on seeing that it is replicable, scaleable, however there
is this whole other world at that other level.
In order to overcome this challenge and find a pathway for scaling up, CCCO may
now need to partner with more system-focused movements or organizations:
I know that if I am going to impact this higher world here, I have to pull away. . . . I would
love to do that, but I don’t know if I have the expertise to do that. I have the passion for
it. I think I could help. But that is not my area of expertise, my language, my world. For
me, to be able to do that, I would almost need someone to help me.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
20 The Journal of Applied Behavioral Science
Navigating the divide between an enterprise and a system innovation requires new
skills (political, mobilization of partnerships and resources). A related risk for CCCO
in adopting systemic change goals is the possibility of losing focus on the quality con-
trol of its product.
Discussion
To achieve larger impact in a complex environment, the organizations described here
create new pathways through combining different elements that are influenced by the
initial conditions. In doing so, they shift the boundaries of what Stuart Kauffman
(2008) refers to as “the adjacent possible.” In other words, it can be argued that social
entrepreneurs and nonprofit organizations use different pathways to diversify future
possibilities, by undertaking particular actions and making certain choices. Therefore,
the pathways they choose to achieve system change vary, as a particular pathway or
combination of elements may be more effective to shift the boundaries of the adjacent
possible for a given organization and their context.
It is helpful to look at an example of an organization that successfully made the
transition from scaling out to scaling up. Planned Lifetime Advocacy Network (PLAN)
was founded in Vancouver two decades ago by the parents of children with disabilities.
Under the inspired leadership of Al Etmanski and Vickie Cammack, PLAN worked to
develop a different concept of disability, focusing on the gifts that people with disabili-
ties have rather than on their deficits. The group’s initial goal was a secure future for
their own children, both financially and socially. This was achieved by creating a life-
long social network around each person with a disability. The results were very posi-
tive and as the demand for PLAN services grew, Al and Vickie worked to disseminate
the model to communities across Canada. But even as their success grew, so did their
dissatisfaction.
Even though PLAN’s concept of creating a network of friends around individuals
with a disability proved very popular, Al and Vickie decided to step back from this
publicly lauded replication initiative to focus on altering the larger system that contrib-
uted to excluding those with disability from mainstream society. They recognized that
being safe and secure was not enough. Individuals with disabilities and their families
wanted a good life, one that involved contribution and participation. To have this
would require breaking out of the straitjacket of conventional conceptions of the dis-
abled, on one hand, and of the financially restrictive approach to disability pensions on
the other. In 1999, they decided to start a new organization, Philia, devoted to creating
a national dialogue between leading thinkers and individuals with disabilities. They
also developed and actively advocated for nation-wide changes that would mean long-
term financial security for people with disabilities. This strategy resulted in a break-
through: the establishment of the first Registered Disability Savings Plan, which
makes it possible for people with disabilities to accumulate savings without losing
their disability payments. Thus PLAN, beyond serving individuals and families
through networks of support, was able to change the life conditions for all Canadians
with disabilities.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 21
As the PLAN case reveals, succeeding in moving from scaling out to scaling up
demands reframing of the problem to focus on system change, and developing a tai-
lored strategy to achieve it. To scale up, organizations need more than a good idea,
adequate resources, and leadership capacity and drive; they must also be able to rec-
ognize and seize an opportunity without the ability to control it directly (Westley,
Patton, & Zimmerman, 2006). Critical to this process are the institutional entrepre-
neurs—individuals or networks of individuals committed to and skilled in changing
broader systems and helping social innovations scale up (Dorado, 2005). As institu-
tional entrepreneurs, Al Etmanski and Vickie Cammack shifted their focus to the larger
system and were able to address the very core of the problem. With the aim of influ-
encing the broader cultural context, their new organization, Philia, provided a venue
for discussions about how to include the marginalized people in our society by appre-
ciating their gifts and diversity, and how to create a greater societal capacity to care.
Unlike social entrepreneurs who create a new idea or product to satisfy unmet
needs (Leadbeater, 1997), institutional entrepreneurs not only introduce an innovation,
but also manage the broader context “in such a way that the innovation has a chance
to flourish, widening the circle of its impact” (Moore & Westley, 2011, p. 4). In seek-
ing broader institutional change, institutional entrepreneurs aspire to cross scales and
move social innovation from one level to another, in contrast to social entrepreneurs
whose efforts are mainly contained within one scale (e.g., transform neighborhood or
community). To do this effectively, institutional entrepreneurs require a broad range of
capabilities such as cultural and social skills (cognitive, knowledge management,
sense making, convening), political skills (networking, advocacy, lobbying, coalition
building), and resource mobilization skills (financial, social, intellectual, cultural, and
political capital (Moore & Westley, 2011; Westley & Antadze, 2010). Cultural and
social skills enable institutional entrepreneurs to recognize emerging patterns and
sense the moment when change is possible, as well as to discern which innovations
have the potential for institutional change. Political skills help institutional entrepre-
neurs to recognize and mobilize relationships that could help advance social innova-
tion to the upper scales. These relationships are strategically built in order to
communicate social innovation in an accessible and engaging manner and be ready to
shift it to a higher scale when an opportunity emerges. Last, resource mobilization
skills enable institutional entrepreneurs to seek and leverage needed resources (Moore
& Westley, 2011).
The case studies described above highlight the skills needed by entrepreneurs who
aim to scale up their social innovations. However, different configurations will be best
served by different subsets of these skills. For the LEGO and Polishing Gemstones
configurations, political skills are key to building strategic relationships and to con-
vening conversations with high-level policy makers about the social innovation.
Organizations pursuing the Umbrella pathway may need to use cultural and social
skills as they reconsider their role and competence to address system-level challenges,
and the adequacy of their current relationships with their partners. In contrast, organi-
zations adopting the Volcano and Beanstalk configurations may find that resource
mobilization skills are central. Institutional entrepreneurs in these organizations will
be challenged to mobilize and leverage the resources needed to drive innovation to
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
22 The Journal of Applied Behavioral Science
higher scales, and may need to build strategic relationships to assure needed resources
for the future (Moore & Westley, 2011).
Conclusions
An overview of the case studies profiled here shows that the notions of scaling out and
scaling up are often linked (see Figure 2). Most of the organizations studied started
their diffusion efforts on a local scale (Initial Conditions). Gradually they pursued
scaling-out strategies by replicating and disseminating their innovation (Scaling Out).
At this stage, they expanded their activities by creating networks and building up
knowledge, experience, and reputation. If they are successful, however, organizations
sometimes reach a “glass ceiling” of diminishing returns. As one social innovator put
it, “I realized that no matter how many local organizations I began, the root problem
remained the same.” This realization becomes a threshold of decision. The organiza-
tions that were content with their existing activities and results saw the threshold as a
“ceiling”; those interested in pursuing system change saw their existing capacities,
experience, and activities as a “platform” from which to launch into a larger sphere of
activity (System Change). These organizations started to develop strategies to influ-
ence the systems or institutional practices that were generating problems in the first
Figure 2. Platforms and ceilings in the process of scaling up.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 23
place, and began using a scaling up pathway to extend their impact. However, at the
time of writing, all these pathways encountered barriers. Among these was the capac-
ity to switch roles from social entrepreneur to institutional entrepreneur focused on
changing the broader social system to enable a social innovation to flourish.
The cases studies presented in this article illustrate that prior to moving into the
domain of system change, organizations need to build a certain “platform” through
successful dissemination of their ideas or products. Without this platform of experi-
ence, in-depth knowledge of the field, and established reputation, it would be practi-
cally impossible to make a difference on a larger scale. In addition, being successful in
scaling out enables an organization to discern problems and issues that were not visi-
ble before, and therefore, to identify new ways and approaches to changing the
system.
Each organization found its own pathway for scaling up. The choice of the pathway
was determined by a number of factors, such as initial starting conditions, existing
competencies and resources, obstacles and opportunities faced by the organization.
The most important barrier was the internal one: All of them realized that pursuing a
scaling up pathway might mean having to leave behind something that was very inte-
gral to their organization. For example, EWB pursuing a Volcano pathway risked
diluting the energy and excitement within the organization; Tamarack, characterized
by a LEGO pathway, feared that the active dissemination of its principles and ideas
would be undermined.
No general conclusions emerge which would suggest which pathway to recom-
mend; however, being able to distinguish the elements of a particular configuration
allows us to dissect the impetus for and success of efforts to scale up. Our cases also
offer some insight into why so many organizations fail to “scale up” and why, there-
fore, there is so little successful social innovation of the kind that changes the institu-
tional landscape. Similar to technical innovation, it would appear that social innovation
in complex domains is path dependent (Arthur, 2009), and that the starting conditions
are therefore both constraining and enabling. The desire to shift gears, to move from
being social entrepreneurs to institutional entrepreneurs, is not trivial. It involves
reframing the problem, adopting a mind-set of system change, and reevaluating the
organization’s role in addressing the identified social problem. Insights gained in this
process can lead to a reorientation of the organization’s strategy and to mobilizing the
resources needed to pursue a scaling-up pathway. Finally, the institutional entrepre-
neur’s new long-term vision must inform the operation and day-to-day activities of the
organization. Undoubtedly, such profound organizational changes are difficult to
undertake. In addition, as the skills of social and institutional entrepreneurs are quite
different, a leadership transition may be required, even though such a transition could
mean the loss of the original momentum grounded in the charisma of a founder.
The model of pathways to system change presented in this article is informed by
selected case studies. Therefore, more research is warranted to test the model on a
larger set of cases, including in organizations that did succeed in scaling up their
efforts. Further research may also help clarify the similarities and differences in the
skills that characterize social and institutional entrepreneurs. It might also test the
analytic power of these configurations for explaining the limiting force of initial
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
24 The Journal of Applied Behavioral Science
conditions. Conversely, it could be illuminating to use the approach presented here to
reanalyze historical cases of social innovation. The interplay between individual
agency and moments of opportunity, and the skills required to connect the two, as well
as the element of serendipity, could be fruitful territory for both research and practice
(Westley et al., 2006). Finally, we expect that some elements may turn out to be more
important than others in particular configurations—we could refer to them as “core”
and “peripheral” elements (Fiss, 2009).
We hope that the discussion of configurations and their elements may be useful to
practitioners as they seek their own, unique pathway. Box 1 presents some implica-
tions for practitioners drawn from the above cases and conceptual frameworks.
Pursuing a scaling up strategy is a challenging and demanding task. In fact, the
number of organizations that have successfully scaled up their social innovations is
quite rare, which may simply confirm the need for new ways to understand the pre-
requisites and strategic pathways for achieving system change. Even though simi-
larities may be found among the challenges that organizations face and the ways in
which they address these difficulties, it is important to remember, that each organi-
zation is unique and therefore, must determine its unique pathway to achieving sys-
tem change.
Box 1. Some Implications for Practitioners.
1. Scaling up is a difficult and time-consuming process. However, organizations do
not need to embark on it right away. Before attempting to scale up, organizations
need to build a certain “platform” of experience, reputation, and in-depth knowl-
edge of the field.
2. Scaling up implies viewing problems and their solutions through a systems
perspective.
3. As organizations try to pursue a scaling up strategy, they realize the need for new
resources.
4. While pursuing a scaling up pathway, organizations may need to let go of some-
thing that was very integral to their organization.
5. Organizations need to find their unique pathway for scaling up.
6. A complex set of skills is required to undertake a scaling up strategy, including
a. cultural/social skills (cognitive, knowledge management, sense making, con-
vening)
b. political skills (networking, advocacy, lobbying, coalition formation)
c. resource mobilization skills (financial, social, intellectual, cultural, and political
capital)
7. The choice of pathway to scale up may be determined by the following factors:
a. initial starting conditions
b. existing competencies and resources
c. obstacles and opportunities that the organization faces
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 25
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship,
and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this
article.
Note
1. Using complexity theory terminology, Byrne (2009, p. 102) calls outcome an “attractor
state.”
References
Arthur, W. B. (2009). The nature of technology: What it is and how it evolves. New York, NY:
Free Press.
ArtsSmarts. (2010). History. Retrieved from http://www.artssmarts.ca/en/about-us/history.aspx
Bloom, P. N., & Chatterji, A. K. (2008, December). Scaling social entrepreneurial impact.
Durham, NC: Fuqua School of Business, Duke University. Retrieved from http://faculty.
fuqua.duke.edu/~ronnie/bio/BloomChatterji_090108.pdf
Byrne, D. (2009). Complex realist and configurational approaches to cases: A radical synthesis.
In D. Byrne & C. C. Ragin (Eds.), Case-based methods (pp. 101-111). Los Angeles, CA:
Sage.
Caulier-Grice, J., Mugan, G., & Vale, D. (2008). Discovery, argument & action: How civil
society responds to changing needs. London, England: The Young Foundation. Retrieved
from http://youngfoundation.org/wp-content/uploads/2013/02/Discovery-argument-action-
how-civil-society-responds-to-changing-needs-March-2008.pdf
Charmaz, K. (2000). Grounded theory: Objectivist and constructivist method. In N. K. Denzin
& Y. S. Lincoln (Eds.), The SAGE handbook of qualitative research (2nd ed., pp. 509-535).
Thousand Oaks, CA: Sage.
Child Development Institute. (2010). What is SNAP? Retrieved from http://www.stopnowand-
plan.com/whatis.php
Christensen, C. M., Baumann, H., Ruggles, R., & Sadtler, T. M. (2006). Disruptive innovation
for social change. Harvard Business Review, December, 94-101.
Corbin, J., & Strauss, A. (Eds.). (2008). Basics of qualitative research: Techniques and proce-
dures for developing grounded theory. Thousand Oaks, CA: Sage.
Dees, J. G., Anderson, B. B., & Wei-Skillern, J. (2004). Scaling social impact: Strategies for
spreading social innovations. Stanford Social Innovation Review, 1(4), 24-32.
Dorado, S. (2005). Institutional entrepreneurship, partaking, and convening. Organizational
Studies, 26, 385-414.
Engineers Without Borders. (2010). Our history. Retrieved from http://www.ewb.ca/about-
us#history
Fiss, P. C. (2009). Case studies and the configurational analysis of organizational phenomena.
In D. Byrne & C. C. Ragin (Eds.), Case-based methods (pp. 424-440). Thousand Oaks,
CA: Sage.
Geels, F. W., & Schot, J. (2007). Typology of sociotechnical transition pathways. Research
Policy, 36, 399-417.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
26 The Journal of Applied Behavioral Science
Goldstein, J. A., & Hazy, J. K. (2008). Complexity and the generation of social value [Editorial].
Emergence: Complexity and Organization, 10(3), vi-x.
Goldstein, J. A., Hazy, J. K., & Silberstang, J. (2008). Complexity and social entrepreneurship:
A fortuitous meeting. Emergence: Complexity and Organization, 10(3), 6-24.
Gunderson, L., & Holling, C. S. (Eds.). (2002). Panarchy: Understanding transformations in
systems of humans and nature. Washington, DC: Island Press.
JUMP Math. (2009). The impact of JUMP Math in England–Lambeth (Research reports).
Retrieved from http://jumpmath1.org/research_reports
Kauffman, S. A. (2008). Reinventing the sacred: A new view of science, reason and religion.
New York, NY: Basic Books.
Leadbeater, C. (1997). The rise of the social entrepreneur. London, England: Demos.
Moore, M.-L., & Westley, F. R. (2011). Surmountable chasms: Networks and social innova-
tion for resilient systems. Ecology and Society, 16(1), article 5. Retrieved from http://www.
ecologyandsociety.org/vol16/iss1/art5
Moore, M.-L., Westley, F. R., & Nicholls, A. (2012). The social finance and social innovation
nexus [Editorial overview]. Journal of Social Entrepreneurship, 3, 115-132.
Mulgan, G., Ali, R., Halkett, R., & Sanders, B. (2007, September). In and out of sync: The
challenge of growing social innovations (Research report). London, England: National
Endowment for Science, Technology and the Arts. Retrieved from http://www.socialin-
novationexchange.org/node/238
Mumford, M. D. (2002). Social innovation: Ten cases from Benjamin Franklin. Creativity
Research Journal, 14, 253-266.
Nill, J., & Kemp, R. (2009). Evolutionary approaches for sustainable innovation policies: From
niche to paradigm? Research Policy, 38, 668-680.
Nunn, R. J. (2007). Complexity theory applied to itself. E:CO, 9(1-2), 93-106.
Patton, M. Q. (2011). Developmental evaluation. Applying complexity concepts to enhance
innovation and use. New York, NY: Guilford Press.
Pundir, A. K., Ganapathy, L., & Sambandam, N. (2007). Towards a complexity framework for
managing projects. E:CO, 9(1-2), 17-25.
Ragin, C. C. (1987). The comparative method: Moving beyond qualitative and quantitative
strategies. Berkley: University of California Press.
Ragin, C. C. (1990). Fussy-set social science. Chicago, IL: University of Chicago Press.
Rihoux, B., & Lobe, B. (2009). The case for Qualitative Comparative Analysis (QCA): Adding
leverage for thick cross-case comparison. In D. Byrne & C. C. Ragin (Eds.), Case-based
methods (pp. 222-242). Thousand Oaks, CA: Sage.
Strauss, A. (1987). Qualitative analysis for social scientists. Cambridge, England: University
of Cambridge Press.
Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for
developing grounded theory (2nd ed.). Thousand Oaks, CA: Sage.
Vibrant Communities Canada. (2014). Cities Reducing Poverty Charter. Retrieved from http://
vibrantcanada.ca/charter
Wei-Skillern, J., & Anderson, B. B. (2003, September). Nonprofit geographic expansion:
Branches, affiliates, or both? (CASE Working Paper Series No.4). Durham, NC: Center for
the Advancement of Social Entrepreneurship, Fuqua School of Business, Duke University.
Retrieved from http://www.caseatduke.org/documents/workingpaper4.pdf
Westley, F., & Antadze, N. (2010). Making a difference: Strategies for scaling social innova-
tion for greater impact. Innovation Journal: The Public Sector Innovation Journal, 15(2),
article 2.
at UNIVERSITY OF WATERLOO on April 30, 2014jab.sagepub.comDownloaded from
Westley et al. 27
Westley, F., Olsson, P., Folke, C., Homer-Dixon, T., Vredenburg, H., Loorbach, D., . . . Van der
Leeuw, S. (2011). Tipping towards sustainability: Emerging pathways of transformation.
AMBIO: A Journal of the Human Environment, 40, 762-780.
Westley, F., Patton, M. Q., & Zimmerman, B. (2006). Getting to maybe: How the world is
changed. Toronto, Ontario, Canada: Random House Canada.
Young, O. R., Lambin, E. F., Alcock, F., Haberl, H., Karlsson, S. I., McConnell, W. J., . . .
Verburg, P. H. (2006). A portfolio approach to analyzing complex human-environment
interactions: Institutions and land change. Ecology and Society, 11(2), article 31. Retrieved
from http://www.ecologyandsociety.org/vol11/iss2/art31/
