Systems change frameworks
Many individuals and organisations have developed conceptual frameworks to help us better understand and think about systems change. But which one is best?
I recently reviewed 28 systems change frameworks to understand their similarities and differences. Most were from sectors related to human and social services, inclusive economic systems and entrepreneurship. Each framework reflects different perspectives, experiences and ideas about systems change. Some are adaptations of other frameworks. You can also find the framework I developed in 2020 here.
Systems: a set of things working together as parts of a mechanism or an interconnecting network; a complex whole.
Systems change: ‘systems change is both a process and an outcome. Both a verb and a noun’ (FSG).
Most frameworks were developed by management consultants, evaluators, government or philanthropic grant-funding organisations. Interestingly, several different frameworks were developed by or under the auspices of the United States Agency for International Development (USAID). Generally, I focused on frameworks related to the what of systems and excluded frameworks that mostly looked at how to change systems e.g. build capacity for people to engage.
All frameworks consist of a set of elements, which I sorted into four broad categories:
- Parts of a system like institutional structures or relationships between people and organisations in a system.
- Characteristics of systems change e.g. the size of the change (scale) or sustainable or dynamic.
- Related to the intervention to achieve change e.g. its quality.
- The outcomes or results of the systems change.
Some frameworks are based on one category of elements while others use a combination. Each is explained below, with examples.
4 categories of framework elements
- Parts of a system
Frameworks built around system parts or components were most common. The parts included:
- Institutional structures and rules such as policies and regulations.
- Functions or roles played by actors (such as in market systems that include retailers, consumers, wholesalers, customers, importers, employees etc.).
- Relationships and connections between actors (individuals, groups and organisations).
- Resources that actors have or have access to, such as finance.
- Actors’ behaviours.
- Actors’ mental models.
- Power dynamics between actors.
Institutional structures/formal rules; relationships and actors’ behaviours were most often included. Resources, power dynamics and functions were less frequently mentioned.
A few frameworks include all parts while others only touch on a few. This reflects the importance given to them by their authors and the purpose of the framework. For example, Lomax wants to help users assess systems change. His framework focuses on actors’ actions, changes in their behaviours and resources since, he argues, these are the more tangible, visible elements of systems that makes it easier to then evaluate changes. This has the benefit of simplifying what we examine but it makes it harder to understand why actors’ actions, behaviours and resources might have changed. Understanding why is useful for deeper learning and improvement (like double-loop learning).
Unsurprisingly, frameworks use different labels for the same element. For instance, norms might also be called mental models, beliefs, paradigms, biases and informal rules. Or elements were combined. For example, the Rockefeller Philanthropic Advisors’ separated formal and informal rules while others like USAID’s 5 R’s framework grouped these different types of rules together. Combining parts has the benefit of simplifying frameworks but might also risk overlooking important aspects and distinctions.
A few frameworks listed incentives or drivers underlying systems or systems change as a separate element (such as Posthumus et al) while others (e.g. USAID Feed the Future program in Bangladesh; Tamarak Institute) saw that policies or mental models were incentives and drivers that influenced other elements (i.e. actors’ behaviours, relationships, power dynamics, and functions).
Most frameworks are generic even if they were developed with specific sectors in mind. This makes them easier to adopt or see their relevance to different situations. A few were more narrowly focused and targeted towards specific audiences e.g. GiZ’s guide to strengthening entrepreneurship eco-systems.
2. Characteristics of systems change
The second category relates to the type of change and included:
- Scale or scalability
- Directionality
- Speed
- Dynamism
- Sustainability.
Scale was most frequently mentioned (eight out of 28 frameworks), mostly emphasising breadth of change e.g. how many people in the system have changed their behaviour. Depth is mentioned explicitly by some authors, sometimes as part of scale. Another USAID framework incorporates depth into its system parts by distinguishing between ‘agent level’, that is individual actors, and ‘collective level’ or individual actors working together. This separation could signal both depth and breadth. Others (like FSG’s framework) capture the depth of change by distinguishing between the change potential of different parts of the system. For instance, a change in policy or resource flows are structural changes but changes in mental models are transformational changes.
Sustainability and/or resilience emphasised adaptability to changes in context. Basically, do changes revert back to a previous state or change in a negative way or do the changes make the systems stronger (for instance, the Climate Investment Fund talks about withstanding pressures and challenges and ‘being robust’). Two frameworks also referred to self-reliance, with a specific focus on information and using information to continue to make improvements. For others, this might rather be seen as a function of a system.
Few frameworks include the speed or direction of change. In my own experience working on systems change initiatives, there sometimes seems to be an assumption that change processes will follow a transformational pathway because that is what we would really like to see. In reality, change follows many different paths (as as illustrated by Ruedy). This seems to be somewhat of an overlooked element, but two frameworks that consider directionality are the adopt-adapt-expand-respond framework and the Climate Investment Fund.
One framework listed dynamism as a separate element but in other frameworks a similar idea was captured in relation to functions and the level of innovation and diversity of products, services, processes and actors.
3. Interventions
Some frameworks include elements that relate to a programme or intervention that is being implemented to achieve systems change. This makes sense if the framework might be used for evaluating systems change. These elements are not strategies, or the how, but relate to the quality of the intervention and include:
- Relevance of the programme or intervention to achieving the objective, the target group or context (as in Chaplowe and Gupta’s paper).
- The range, character, quality and location of existing programmes and support (for example in the ABLe Change Framework).
- The connections between different programmes that aim to achieve change.
In some frameworks, programmes are seen as external to the system, they are envisaged to be temporary interventions to trigger positive and lasting systems changes. In other frameworks, the programmes are part of the system, e.g. as in health sector, and therefore might also be seen as an actor, with relationships and connections to others in the system.
4. Outcomes / results
Again frameworks took different approaches. Some didn’t address the issue of intended outcomes at all.
A few frameworks included a separate, but broadly stated, element such as the ‘outcomes at a community or population level’ or ‘the extent to which the system was seen as working’. In other frameworks, more detailed outcomes for the target groups were be captured under descriptions of system parts. For example, some frameworks mentioned ‘increase in inclusion’ under relationships; or changes in actors’ behaviours e.g. ‘local founders are mentors’.
Which is the best framework?
There is no one framework that is the best. No framework is ‘right’. The effectiveness of a conceptual framework is the degree to which it helps us think and learn more about systems change.
I wouldn’t recommend taking an approach ‘off-the-shelf’ but rather investing in the time and effort to build your own. Examining existing frameworks, and their parts, can help us learn faster and speed up this building process. Like many tools and processes, a lot of value is created in the process of developing tools and the discussions and debates it generates between different stakeholders who often have different understandings, ways of thinking and assumptions about change and change processes. These discussions can improve communication, relationships and teamwork as well as strengthen strategies and knowledge.
That said, the better frameworks are likely to cover all four categories of elements since this means that different ways of thinking about and looking at systems change has been considered. I might now go back to my own framework and think about how it can be improved.
What systems change frameworks have you found useful? What made them useful, for whom and in what situations?
A list of frameworks can be found at this post here.
