# Systemic Strategy: Systemic Design Methods for Complex Systems Change
A presentation from Peter Jones and I at RSD9, virtually in Ahmedabad, India.
A presentation from Peter Jones and I at RSD9, virtually in Ahmedabad, India.
a16z is a podcast from Andreesen Horowitz, one of the leading venture capital firms in the US. Their portfolio has included Airbnb, BuzzFeed, Facebook, Twitter, and many many other big companies—they know what they’re doing!
Two recent episodes on the podcast feature Andreesen Horowitz general partners, Andrew Chen and Jeff Jordan (some of the people who make investment decisions). They are interviewed by host Sonal Chokshi about how to think about user (or customer) acquisition and user engagement and retention. (You should also be able to get these episodes using any given podcasting app, searching for a16z, and grabbing the two recent episodes titled The Basics of Growth.)
I’ve listened to these episodes a couple of times since they came out a week ago. There’s a lot of terminology and concepts to parse—they explain it all well, but it took me a couple of tries to wrap my head around all of it. Given what I’ve come to learn so far, though, is that the two episodes seem like a (free!) one-oh-one masterclass in growing or scaling any given initiative. Despite being less than an hour long in total, the conversation goes way beyond “number of users”. Andrew, Jeff, and Sonal dive deep into the kind of patterns we should hope to see (and find ways to encourage) in order to garner genuine growth, engagement, and retention patterns. It is super sophisticated while still being kinda simple.
That said, they’re talking about startups, customers, and technology-driven businesses. Applying those lessons to initiatives in public post-secondary—co-curricular leadership development programs, in my case—is not straightforward.
So, that’s the challenge: how do we generalize their insights into the world of post-secondary student services? Here I explore that question with more questions. There’s probably no simple answer for these, but thinking about these concepts could deepen how we grow (and evaluate the growth of) post-secondary services.
Question 1: Lifetime Value vs. Lifetime Impact? The first episode focuses on user acquisition. The analogy for us is when students first use our services, tools, or go to their first event or program. Jeff and Andrew describe organic growth—word of mouth-type acquisition—and inorganic growth (paid ads and other campaigns). They talk a lot about the tension between the “cost” of acquiring users and the lifetime value (LTV) of a given user.
How should we think about these ideas in student services? “Student Acquisition Costs” have a direct parallel. Is there a useful analogy for LTV? (I.e., “Degree/Career Impact”?) The implication is that acquiring certain students might cost more, or at least demand different strategies, but have a greater impact.
Question 2: Semesterly Active Users? The podcasters reference a variety of “active user” metrics: Daily Active Users (DAU), Monthly Active Users (MAU), last-seven (L7) and last-twenty eight (L28)… What is a useful and realistic “active user” metric for post-secondary student services? We don’t actually want our students to use all of our services daily. As the podcasters put it, there’s a certain ideal cadence to the way they should use our services. But what is that cadence, and how do we ensure we’re meeting it?
Question 3: Network effects? Jeff, Andrew, and Sonal talk about the “network effect” in both episodes. This is the concept that a service becomes more valuable/powerful the more users it has. Is there an analogy to the services we’re already running, and are we taking advantage of it? Or, are we missing an opportunity to build in a network effect or similar positive acquisition, engagement, or retention feedback loops somehow?
To be clear, I don’t think it’s a good idea to translate directly from the high-growth profit-driven startup world into post-secondary services work and life. The organizations have different motives and needs, incentives and costs, and—most important—different stakeholders, with different obligations to those stakeholders. Still, there may be powerful, creative opportunities hidden in the lessons we borrow from one another.
A thought-terminating cliché limits conversation by capturing a complex (but potentially debatable) subject within a reductive term or phrase. Merlin Mann references this idea in episode 164 of Back to Work when discussing curiosity and buzzwords.
Thought-terminating clichés can be used to avoid discourse on a subject: by never unpacking the components of an idea that are debatable, those components go unexplored. They can also be exploited to veil ignorance or illogic—the speaker can state the complex term and allow the implication to have impact without contextualizing/explaining it while the intimidated audience shies away from critique or questioning.
This explanation makes the phenomena seem villainous, but many of us are prone to committing these crimes—through buzzwords! Buzzwords are terms that catch on because they represent something exciting to a discourse. Then, because they’re popular, they get used frequently, by many people. Because they are somewhat novel, these different uses attach slightly different meanings to the same word. Eventually the buzzword’s overused (reducing the novelty, and therefore the impact of its meaning) and/or overloaded with meaning.
Most of our buzzwords were real things at one point (and sometimes they still are). When buzzwords are used effectively they allow a good conversation to move faster between speakers who have the same mental models about the buzzwords.1
Sometimes, however, buzzwords are said to represent concepts that aren’t fully understood by everyone in the conversation. When my meaning of the word “design thinking” differs from yours, but we both refer to design thinking in conversation nonetheless, we can run into trouble.
In these situations, buzzwords obfuscate the ideas we’re actually talking about. In my experience, we also know when we’re using buzzwords. We can guess at when others are using them, too. As a result, the conversation loses meaning, and we lose trust in the conversation.
So what can we do?
Well, the easy thing to do is to clarify. When you use a phrase with many potential interpretations, try to clarify how you’re using the phrase. When others use words that may have multiple meanings, ask specific questions about what they actually mean. This clarification might seem like extra work, but it only needs to happen when terms are first invoked—and it’ll prevent lost time and energy due to the consequences of thought-termination later on.
More importantly, though, we should try to avoid thought-terminating clichés altogether. Take time to break down the concepts you’re talking about in concrete terms. Explain them in ways you haven’t heard before to avoid relying on trite metaphors and anecdotes. If you can really get at what you mean, your language will be minimally re-interpretable: that is, it should be near-impossible to understand your explanations differently from how you intended.
As a result, your communication will become more impactful. The conversations you participate in will have more novelty, too, making it more exciting to discuss the ideas you’re sharing. This may result in more buzzwords emerging, but that’s okay—use the same approach to break those down, too.
See the excellent Notes on the Role of Leadership and Language in Regenerating Organizations for more on this. ↩︎
Updated Nov 14, 2019
Below I’m collecting a set of resources dealing with a variety of topics for shareable reference—a sort of living, semi-public annotated bibliography. Many of these readings should be accessible via a library or a good Internet search, but if you find something here that shouldn’t be this easy to download, feel free to contact me and I’ll remove it. Or, if you come across any broken links, please let me know!
Kudoz’ Where Are You Scaling? Placemat (2017) Kudoz’ Where Are You Scaling? “placemat” is an explorable tool to help think through what scaling really means in a given venture. It builds upon Riddell & Moore’s (2015) Scaling Out, Scaling Up, and Scaling Deep in expanding our notion of scaling. When you hear the word “scale” in a conventional conversation, the meaning is likely synonymous with scaling out: reaching more people (stakeholders or customers) with the product, service, program, or policy at the heart of the conversation. Yet there are other underappreciated aspects of scaling: some are required in order to successfully scale out, while others are different ways of achieving impact—maybe scaling out isn’t the best route to take after all! Scaling up, for instance, refers to reaching higher in power hierarchies in order to implement higher-level solutions to the problem you’re addressing. Scaling deep means to reach the same people you’re already reaching more deeply, or with greater impact. Kudoz introduces two more concepts: scaling infrastructure, or building out more tools and supports to make the work more effective and sustainable; and scaling scree, or lending support and legitimacy to other, different actors and initiatives addressing the same issues you are.
Mulgan & Leadbeater’s Systemic Innovation (2013) While the above concepts help make concrete the different ways an intervention might change a system, Mulgan and Leadbeater’s (2013) discussion paper on systemic innovation dives a bit deeper into theory. They link the practice of systems thinking to approaches to innovation, providing a framework for “joining up” a set of innovations in order to achieve systemic change.
Mulgan’s Social Innovation: What It Is, Why It Matters, And How It Can Be Accelerated (2007) Geoff Mulgan offers a comprehensive yet down-to-Earth guide to social innovation. The reading really helps communicate what the differences are between social innovation and conventional innovation with practical examples.
Phills Et Al.‘S Rediscovering Social Innovation (2008) The Stanford Social Innovation Review has been one of the preeminent publishers on social innovation since its inception in the early 2000’s. In this review Phills et al. summarize a brief history on social innovation itself while paving a way forward in research and practice.
What is “design”, really? It’s easy to think that design is the layout of graphics or illustrations—perhaps even making decisions about layouts, as in architecture. These disciplines—graphic design and architecture—certainly leverage design, and many people employed in these industries are designers. Yet these disciplines do not fully capture what design is, thus we should not limit the understanding and application of design to these industries.
The term design shares latin roots with designation. To design is therefore to mark; to frame. Framing means to view a given problem or system from different perspectives in order to deeply understand it and, in turn, to develop novel ways of doing something about it or with it. To become a better designer, you must get better at both analyzing and creating frames.
Systems thinking is most useful when applied to complex, chronic issues: “wicked problems” that resist taming through conventional approaches. David Stroh, who wrote Systems Thinking for Social Change, argues that effective systems initiatives begin with a focusing question: a question that bounds the project (so that you don’t spend forever debating about where the problem system really ends) and anchors your systems-understanding work in a framing that you can return to if you get lost. This is a question of the form “Why, despite our best efforts and intentions, does [x] persist?” or “…have we not reached our goal of [y]?”—where [x] and [y] are phenomena that remain unassailable through conventional initiatives.
Goodman, Kemeny & Roberts’ The Language Of Systems Thinking (2000) The Language of Systems Thinking introduces the reader to reinforcing and balancing loops using simple examples.
Braun’s The Systems Archetypes (2002) Systems archetypes extend the links and loops of basic systems thinking into more complicated structures – and in doing so, begin to share with the reader deeper insights into the counterintuitive behaviour of the systems we work with.
Donella Meadows’ Leverage Points: Places To Intervene In A System (1997) With robust systemic models, we can begin to identify leverage points: key phenomena that have great potential for change throughout the system.
Jim Dator’s What Futures Studies Is, And Is Not (1995) What Futures Studies Is, and Is Not is a wonderful introductory and reader-friendly summary of futures thinking.
Choo’s The Art Of Scanning The Environment (1999) Chun Wei Choo’s The Art of Scanning the Environment offers a walkthrough of scanning practices.
Bishop, Hines & Collins’ The Current State Of Scenario Development (2007) The Current State of Scenario Development provides a comprehensive review of scenario development approaches and techniques.
Dubberly Et Al.‘S Notes On The Role Of Leadership & Language In Regenerating Organizations (2002) Hugh Dubberly and a suite of colleagues provide an elegant—poetic, even—walkthrough of how change happens to organizations through their language. They reframe an organization as a continuous set of conversations between people, in turn showing the power of changing the language those people use in order to change the nature of the organization itself.
Weick & Westley’s Organizational Learning: Affirming An Oxymoron (1999) Karl Weick and Frances Westley attack the very idea of organizational learning by showing how “organizing” and “learning” are, in fact, opposing processes. To organize is to order; to learn is to disrupt. Success in organizational change is therefore rooted in a leader’s ability to handle the tension between organizing and learning. Too much organizing and the organization will miss opportunity in change—but too much learning and the organization might not be stable enough to survive.
Gawde’s Learning From Systems Thinking (2017) How can systems thinking help with changemaking in an organization? Writing for People Matters, Sanjay Gawde provides a succinct argument and explanation linking systems approaches to organizational change.
Based on the Applied Systems Thinking workshop, I’ve collected a variety of resources to help you map complex problems below. The buttons link directly to files to save you some trouble. Be mindful that most of these files are published documents or books, so if you really like them, you should buy a copy and support the authors!
If you have questions or if you’re looking for more resources, never hesitate to reach out to me via firstname.lastname@example.org.
What are systems? Interconnected sets of elements whose interactions lead to emergent, “purposeful” behaviour. (A system’s purpose is not necessarily what someone intends of it, though, nor can it be derived from rhetoric about the system. A system’s purpose can only be defined by examining its actual behaviour.)
Systems—and the actors within them—do exactly what they are “designed” to do. That is, systems act perfectly in tune with the structures and incentives that they have. Only by understanding these complex structures and incentives can we begin to make real progress on the challenges we’re addressing.
However, “How might we..?” questions aim to provide solutions. Before we can “solve” systemic issues, however, we must understand them. And unfortunately for us, these are usually wicked problems.
To understand a problem, we must begin to explore causality. Systemic designers seek to understand complexity by searching for the patterns that cause our problems—and finding the underlying structure of those patterns that enable their persistence.
To that end, David Stroh suggests developing a “focusing question” in systems work. The purpose of systems mapping, he says, is not to map a system; it is to answer the focusing question.
A focusing question has the form “Why does this problem persist?” or “Why, despite our best efforts, intentions, and resources, have we been unable to achieve a certain goal or solve a particular problem?”
Rich pictures come from Checkland’s Soft Systems Methodology (SSM). Akin to a sketchy infographic, these maps are illustrated and make heavy use of labels and symbols to help the mapper or the reader understand a messy situation.
Actor maps graph the relationships in a social system. Who (or what organizations) influence who? Who funds who? How is the vision of the system determined? By identifying different stakeholders, including who are the most important beneficiaries and victims of a system, systemic designers might catch gaps, missed connections, or other issues.
Also known as influence diagrams and effect maps, Causal Loop Diagrams graph the phenomena of a system. What are we trying to stop from happening (or what do we want to happen more often)? What encourages or limits those phenomena? Then, what encourages or limits _those_causes or limits? By drawing causal connections between the phenomena of the system, we can recognize the complex interactions that lead to the (frequently counterintuitive) emergent patterns of behaviour normally invisible.
How much of what quantities flow at what rates? Stock and flow diagrams make explicit the system’s stores (e.g., the heat in a cup of coffee) and its rates of change (e.g., how quickly heat escapes from the cup). These diagrams also recognize what controls these rates of change.
Add detail to your systems maps by exploring the different dimensions along which influence might flow: wealth, power, values, knowledge, or beauty. Different types of obstructions—poverty, maldistribution, and insecurity—can cause different types of problems in each of these dimensions.
Leverage points are places within a system with which a little effort yields great reward. Likewise, bottlenecks are places within a system which resistance could cause significant problems.
Leverage points are places within a system with which a little effort yields great reward. Likewise, bottlenecks are places within a system which resistance could cause significant problems.
Systems archetypes are common patterns identified in causal loop diagrams or stock-and-flow diagrams. Archetypes exhibit similar behaviours and can be resolved by similar solutions.
Do you need a particular actor or phenomena to receive resources/power/etc.? Where would that resource come from? What influences how much of the resource gets distributed? How can you increase those influencing forces? The business idea uses a causal loop diagram to map the systemic structure of an organization’s strategic sustainability. It makes explicit the phenomena that generate resources for the system to reinvest—and the strategic competencies that the organization can use to enhance those phenomena.
What is the change strategy you’re adopting? Similar to the business idea, a systemic theory of change plots a theory of change model in systemic form, identifying the goal phenomena to enhance (or limit), the key activities that can support that enhancement (or limitation), and the inputs required to sustain and scale those activities.
Never explain a systems map in a pitch or to a general audience. Instead, follow the iceberg model to distill a systems story.
Mural.co is a collaborative tool for design sprints. As such, it provides features for collaborative whiteboarding and sticky noting, voting via dotmocracy, and a variety of other neat and helpful tools. It is very free-form (and as such has no features specifically made for systemic design) but that open-endedness may be useful.
Nicky Case’s Loopy is a simple tool that allows you to simulate systemic behaviours. It is very unsophisticated (e.g., it is challenging to provide precise system settings) but it is extremely gestural and is therefore fun to use to illustrate and explore ideas.
Plectica is a “visual mapping software”. It allows you to nest and draw connections between cards representing anything. The goal of the app is to provide a simple interface for complex ideas.
Kumu is a web app built specifically for systems mapping. It features extensive features and documentation and a lively support community full of fellow systems mappers who like to help one another. The developers/founders are active participants in that community and regularly provide customer support, too. Develop actor maps, systems maps, and all kinds of other interesting interactive visuals with Kumu.
I’ve collected and described resources on related topics at https://systemic.design/resources. In particular, you may want to check out the items on organizational change (e.g., the “Notes on Leadership and Language in Regenerating Organizations” paper and the article on organizational learning).
A paper presented at RSD7 in Turin, Italy.
I argue that an under-emphasized but crucial variable of success in systemic design is the perspective through which systemic design processes are implemented and executed. While rooted in design (a consciously empathetic discipline, especially in recent years; cf. Kimbell, 2011), it is easy for systemic designers to conduct the research required for their projects in externalized ways. These approaches risk misrepresenting the stakeholders who contribute to projects and, in turn, they are a danger to the potential impact of these misresearched problem systems. I propose to advance a theoretical argument for this danger, the development of an assessment framework to check whether an internalized perspective has been effectively achieved, and provide a proof of concept of this framework through hermeneutic case study analysis.
As I will show, systemic design processes that are not executed with the direct and explicit engagement of stakeholders – to the extent of achieving an emic (or from within) understanding of the system – may be flawed at their foundation. By fostering recognition of the importance of an emic perspective, and by providing a framework of principles, practices, and process to accomplish systemic design with this perspective, I hope to ensure that systemic design processes are as accurate and valid as possible with respect to the stakeholders of the system.
This is not to suggest that systemic design practice is “too etic”. In fact, with roots in design, systemic design is often deliberately emic. Systemic designers make use of designerly tools that help the researcher to build empathy with system stakeholders (e.g., soft systems methodology, critical systems heuristics, appreciative inquiry; Jones, 2014). They often seek to engage stakeholders in the systemic design process and include reflective analysis of what has been learned in order to assess where deeper engagement with the system is required (Ryan, 2014). That said, with the advent of crowdsourcing (the facilitated involvement of the general public in problem solving, usually using online tools; Lukyanenko & Parsons, 2012) and data science (the use of computational tools to analyze and understand large quantities of data; cf. Scepanovic, 2018), it is likely that data-driven methods will increasingly influence systemic design practice. One recent example sought input from hundreds of people to identify opportunities for change in Canadian post-secondary systems through an iterative online survey (cf. Second Muse, Intel, & Vibrant Data, 2016). This data-driven direction is a powerful opportunity, of course, but it underscores the need to develop principles and best practices for assessing and supporting emic understanding as we gain more data from these tools.
This proposal consists of two steps. First, I will look to the principles and theorists of ethnography to develop a framework for assessing the emic/etic perspective of a given research project. Namely, Geertz’ “Thick Description: Toward an Interpretive Theory of Culture” (found in The Interpretation of Cultures, 1973, chapter 1) provides a foundation for the process of emic research, while Creswell and Miller (2000) provide a set of procedural principles for emic validity. Taken together, we generate a critical research framework with which we may assess a given research project’s emic perspective. Second, I will provide a proof-of-concept of this framework (and its theoretical underpinnings) via a casebased assessment of three systemic design projects. Case studies provide an effective venue for learning about the context-dependent manifestations of the phenomena being studied (Flyvbjerg, 2006). One of these case studies is one I have developed through my experience in participating and contributing to the development of the Canadian National Youth Leadership and Innovation Strategy framework, which convened hundreds of youth and youth-serving organizations in order to understand the youth leadership and innovation system in Canada (cf. MaRS Studio Y, 2017). The second and third case studies are those profiled by Ryan and Leung (2014).
In order to interpret and analyze the chosen case studies, I turn to the methodology of phenomenological hermeneutics (Eberle, 2014, p. 196; cf. Wernet, 2014). Phenomenological hermeneutics are appropriate as I have access to the described phenomena of the systemic design projects captured by the chosen cases, but these phenomena are not explicitly captured with reference to emic or etic perspectives – thus some construction of the inherent emic or etic data is necessary in order to make judgments about the perspectives found in the projects.
In each case, I will use identify phenomena representing the practice of emic (or etic) understanding in the research orientation of the work, as acknowledged by the above framework. In each case, I will examine the step-by-step procedure and any associated notes about the experience of the researchers and participants involved. In each step or experience, I will look for evidence of the four steps of emic understanding or the six techniques of emic validation reported above.
Creswell, J. W., & Miller, D. L. (2000). Determining Validity in Qualitative Inquiry. Theory Into Practice, 39(3), 124–130. https://doi.org/10.1207/s15430421tip3903_2
Eberle, T. S. (2014). Phenomenology as a Research Method. In U. Flick (Ed.), The SAGE Handbook of Qualitative Data Analysis (pp. 184–202). Los Angeles, Calif. [u.a.]: Sage. Retrieved from https://www.alexandria.unisg.ch/228374/
Flyvbjerg, B. (2006). Five Misunderstandings About Case-Study Research. Qualitative Inquiry, 12(2), 219245. https://doi.org/10.1177/1077800405284363
Geertz, C. (1973). The interpretation of cultures: Selected essays (Vol. 5019). Basic books.
Jones, P. (2015). Design Research Methods for Systemic Design: Perspectives from Design Education and Practice. Proceedings of the 58th Annual Meeting of the ISSS - 2014 United States, 1(1). Retrieved from http://journals.isss.org/index.php/proceedings58th/article/view/2353
Kimbell, L. (2011). Rethinking Design Thinking: Part I. Design and Culture, 3(3), 285–306. https://doi.org/10.2752/175470811X13071166525216
Lukyanenko, R., & Parsons, J. (2012). Conceptual modeling principles for crowdsourcing (pp. 3–6). ACM. https://doi.org/10.1145/2390034.2390038
MaRS Studio Y. (2017). A strategic framework for youth leadership & innovation in Canada: Insights from the 2016 National Youth Leadership and Innovation Strategy Summit. Toronto, ON. Retrieved from http://www.studioy.marsdd.com/wp-content/uploads/2016/12/MaRS_NYLISstrategic_framework_Final.pdf
Ryan, A. (2014). A Framework for Systemic Design. FORMakademisk–research Journal for Design and Design Education, 7(4). Retrieved from https://journals.hioa.no/index.php/formakademisk/article/view/787
Ryan, A., & Leung, M. (2014). Systemic Design: Two Canadian Case Studies. FormAkademisk - Research Journal of Design and Design Education, 7(3). Retrieved from https://journals.hioa.no/index.php/formakademisk/article/view/794
Scepanovic, S. (2018). Data science for sociotechnical systems - from computational sociolinguistics to the smart grid. Aalto University. Retrieved from https://aaltodoc.aalto.fi:443/handle/123456789/30187
Second Muse, Intel, & Vibrant Data. (2016, May 11). What Your Data Says: Post-Secondary Education Mapping Survey Highlights. RECODE. Retrieved from http://re-code.ca/whats_happening/watch-recodewebinar-what-your-data-says/
Wernet, A. (2014). Hermeneutics and Objective Hermeneutics. In U. Flick, The SAGE Handbook of Qualitative Data Analysis (pp. 234–246). SAGE Publications, Inc. https://doi.org/10.4135/9781446282243.n16
A systems model I helped visualize for the Global Steering Group on Impact Investment in 2018:
Notes, slides, and the Innovation Auditing guide presented at the talk are found below.
The research presented during the talk is discussed on the following pages:
Find a PDF of the slides I presented at the link below. Beware: the animations don’t translate well to print, so some of the pages have graphical issues.
Innovation Auditing is a simple procedure that individuals, organizations, and governments can use to detect the gaps in the innovation process they seek to support.
See the PDF guide by clicking the link below.
What are the futures of art and design schools in Canada?
In 2015, I supported Kinetic Café in developing OCAD University’s latest Vision and Mission statements. As part of that work, I helped scan for signals, trends, and drivers in art and design school futures. Our scan revealed five important drivers of change: reforming education, creative economies and cultures, new geographies, empowering technologies, and conscious collective.
A walkthrough of this work is visualized and embedded below. If it doesn’t display, you can visit it directly here.