Systems thinking: How not to miss the forest for the trees

UNDP Accelerator Labs
5 min readMar 2, 2020


By Jennifer Colville, Innovation Team Lead, Arab States, and Roxani Roushas, Innovation Programme Analyst

“None of us see the system. We see our own part based on our own background and history. And we all think we see the most crucial part.” — Peter Senge, Systems Scientist

As part of UNDP’s Accelerator Labs, we are digging deep into the issues that challenge our countries and our collective ability to reach the Sustainable Development Goals (SDGs). In the Arab States, we recently brought together our innovation community, including the 10 labs in our region (Algeria, Iraq, Jordan, Lebanon, Libya, Morocco, Palestine, Tunisia, Somalia, and Sudan), to reflect on how we are engaging with systems innovation.

Here are some learnings that are emerging through practice:

1. You can never understand your issue well enough.

Horizon scanning, sensemaking, collective intelligence: These are just some of the tools that the Accelerator Labs are using to unpack complex sustainable development issues that change in real time. One of our most preferred tools is issue mapping which draws linkages between the issue at hand and the various factors (economic, environmental, legal, social, cultural, political, technological…) that affect it.

Our Accelerator Lab teams from the Arab States region meet in Jordan

This tool has been spreading like wildfire to create a shared understanding of complex issues among a diverse group of people. For example, the UNDP Sudan Accelerator Lab placed women’s political participation at the heart of their issue map, and then gathered stakeholders of various backgrounds to map out the factors that limit women’s participation. This exercise, and many similar ones conducted by Labs across the network, have generated deep insights about the issues at hand, informing theories of change and hypotheses for the Labs to experiment on.

But this is just the start. Context for the problems we are trying to solve is fast-changing. Even if we feel like we have a deep understanding of the factors affecting an issue at a particular point in time, its environment invariably changes — affected by a new policy, an emergent technology, a natural or human-made event — and hence the nature of the issue changes. Look and you see one picture; look back a minute later, and you could see a very different picture.

Issue maps can be works of art

2. Complex problems call for complex understanding.

Following months of mapping, hypothesizing and experimenting, and enriched with a deeper understanding of their challenges, many Labs are now returning to their initial issue maps and finding them to be a bit linear, a bit static. We realise what has been missing is the relationship between and among factors in the map. So yes, women’s political participation may be weakened by legal barriers, but how are these legal barriers in turn fed by lack of economic participation? How might environmental factors like land degradation be reinforcing economic exclusion? Such linkages may seem obvious in hindsight, but it is only once you begin to map the system and see the connections that you realize there’s much more at play than meets the eye.

If we look at sustainable development problems only on a two-dimensional plane, we will most likely miss crucial underlying factors that will block progress down the line. Systems thinking provides us a tool to look beyond two dimensions, to overlay issue maps upon each other, to see linkages, interdependencies and leverage points and to design experiments that will help us learn more about what kinds of solutions work and what kinds don’t.

As Michael Goodman says, “Systems thinking expands the range of choices available for solving a problem by broadening our thinking and helping us articulate problems in new and different ways.”

Let’s take the case of the UNDP Morocco Accelerator Lab, who are trying to nudge young people back into the job market. They quite quickly identified a number of factors affecting employment, such as gender dynamics, cultural differences between urban and rural areas, language barriers… the list goes on. As they dug deeper, they discovered an intricate web — with these factors influencing each other, either positively or negatively — that is affecting the ability of young people to engage with public employment services and strengthen their skills.

As the Labs explore the systems in which their issues are situated, they are framing learning questions and designing portfolios of experiments that will help generate information, insights, intelligence and inspiration. At our recent workshop, we added a fifth “i” to illustrate that learning is often found in the “spaces in between”:



occurring in or being an interval or intervening space or segment

By taking a systems thinking approach, we are much more likely to see these spaces and in turn be able to develop more relevant solutions.

3. Know your feedback loops.

You may have surmised by now that systems are circular, not linear; wholes, not parts. The circularity of systems is partially attributable to feedback loops, both positive and negative. A simple example might be lack of access to skills training among young people, which prevents them from getting jobs. Exclusion from the labor market in turn makes it even harder for them to fill these skills gaps. A vicious cycle.

What does that mean for us? Solutions that disrupt, or reinforce, these loops, can have an exponential impact. Feedback loops may also hold vital information about the evolution of an issue. When looking at your systems map, therefore, think not just what it reveals about the present, but what a change in any of its nodes could signal for the future.

All of these lessons get to the crux of the added value of Accelerator Labs for UNDP. The Labs have essentially been tasked with exploring previously uncharted territory: identifying patterns previously not noticed and synergies previously not seen.

To fulfill this role, we must continually update our understanding of the issues we’re addressing and the systems in which these issues sit.

It’s only this kind of learning and understanding that can lead us at the Accelerator Labs to the most relevant and sustainable solutions with our national partners.

In the coming months we will be designing portfolios that help us engage with complex issues, but in ways that lead to action. If you have ideas on how to do so, get in touch!



UNDP Accelerator Labs

Building the world’s largest learning network around development challenges. 91 Labs in 115 countries.