Testing a Lab Model

For the past two years, we have had the honour in working with WISIR in designing and implementing two labs –  the Ontario Tender Fruit Lab and the New Solutions for Youth Employment lab in New York City. Below is my contribution to their recently published Social Innovation Lab Guide, published as the Prologue, reflecting on our lessons-learned the past two years. 

“Translating words into deeds is typically a serious challenge, but it seems you guys have been a real catalyst.” Such feedback after running a social innovation lab with people that were largely strangers before the lab started, is why I believe this approach can be so valuable. However, while the field of innovation labs is growing, this approach is still very much under development. And as a field, we need more rigor to ground and test the approach. That comes with becoming a mature infrastructure for social innovation. That is why I am so thankful we have been able to help test of a new lab model.

Over the past year we had the privilege to work with the Waterloo Institute of Social Innovation and Resilience (WISIR) to prototype a social innovation lab model. Professor Frances Westley and her team developed this model based on extensive research on different kind of labs, but also by going back to some of its theoretical foundations, from the work of Eric Trist on systems thinking in the 1960s to Tim Brown’s design thinking. They wanted to test it on some real cases and approached the MaRS Solutions Lab to work with them on that.

Testing the WISIR model allowed us to explore our own approach to what labs are and develop some our own methods. In a time that was most formative for us, as we started in 2013. MaRS Solutions Lab is a public and social innovation lab that helps to tackle complex social and economic challenge that require systems change. What we recognized in the WISIR model was a combination of design and systems thinking. When convening diverse stakeholders, it is crucial to have them understand the problem from not just different institutional perspectives, but also from the user and system perspective. And that is what this model also is about.

In our view, real change that helps solve complex social challenges can only be achieved when three elements of a system are being innovated. First, when present solutions do not work, we need to develop new solutions. As it is impossible to predict what works, we need to experiment. Through prototyping, we can get evidence what works, even before implementing it in the field. What works can be brought to scale. It must be noted though that a combination of solutions is needed. One magic bullet solution does never exist. The goal of a lab is to develop an adaptive change strategy that tests multiple solutions, which together could solve the challenge. These solutions can best be characterized as interventions that each solve a crucial part of the problem or create the right inventive towards that.

Second, we need to innovate the way the system behaves. This means changing how the system is being governed, funded and/or incentivized. It can require changing public policy, but also organizations changing their strategies. It is about creating the conditions for new solutions to become accepted and replace the old ways. Building the support system around new solutions to make them sustainable and bring them to scale. That starts with framing the challenge in a way that it creates a sense of urgency and a desire for action.

Third, we need to build the capacity of the people and organizations involved. Simply saying they need to change will not work. We need to build a movement, starting with the innovators that pioneer new solutions. They are deeply passionate, committed and willing to take risks. But for systems change, we can not just rely on that small group. We need to also engage the early adopters, who see the need to innovation but require some guidance and a safety net. And beyond that the early majority, the people who will only innovate when we can show evidence and offer support to help implement it.

What WISIR’s social innovation lab model provides, is the start of the process of systems change. It helps to convene stakeholders and help them understand the challenge from user and system perspectives. It helps to create a common change strategy and early prototypes of interventions. It helps to frame the challenge and build momentum for action. It helps to find the innovators and build their capacity to more effectively address the challenge.

All of this is crucial to the success of a social innovation lab., and core to the value that social innovation labs bring. The thoughtful way the model is constructed gives a strong fundament to the work of labs, which is much needed. I advise every lab practitioner to learn from it. And to conclude, let me share two of the many lessons we learned from testing this lab model:

First, when you have completed this model, you have only just begun. The model helps to convene a diverse group of stakeholders and bring them to the starting line. The real innovation and impact will only come after, once interventions are being implemented. And when we learn if they work, or not. Despite the enormous efforts and energy put into this stage, you should realize this is just the beginning. But without a good beginning, nothing will change or you will make the wrong changes.

Second, as every challenge is different, adjust this model to the challenge. Do not just follow it step by step. We had intensive team discussions on all elements of the model. About if, how and when in the process to apply each element, and why. Even in the two tests the model was applied differently. For instance, where in the first test the steps were done consecutively, we did them the second test more in parallel. We also developed our own tools to add to the model, or to translate the model into group exercises. That can be done in different ways. On a higher level the model provides a path from research to workshops that probably remains the same, but make your own translation of the model the more specific you get.