“The acquisition of knowledge always involves the revelation of ignorance — almost is the revelation of ignorance.” - Wendell Berry, writer and Kentucky Farmer quoted in Donella Meadows, Systems Thinking: A Primer
This quote blows my mind. It spoke to me when we were looking back at the build of a global network of 91 social innovation labs inside the United Nations Development Programme. When this (admittedly bonkers) idea was designed, we had the intention not just to build and run LOTS of social innovation labs, but to create a learning network that surfaces, shares, and even creates knowledge.
With the state of the world as it is, we felt we needed to do our part to unleash learning and experimentation to match the complexity and pace of the challenges in front of us. This quote from Berry (like many in Meadow’s systems thinking classic) makes un-knowing okay, even essential. In fact, Epistemic humility — our ability to remain humble about what we think we know and how we know it — has opened doors for us as we prototype ways to tap into distributed knowledge that comes from practice on the front line.
Now that we’ve got a network of almost 100 labs operating, sharing, and learning, we are asking ourselves:
- How do we follow the emergence and dynamism of individual labs in a network to understand if and where learning is being accelerated?
- How do we maintain subsidiarity through distributed, independent operations and still create space for convergence in knowledge creation across a network like this?
- What is this network learning?
- How do we fill in its blind spots with what our (current and future) partners know?
This blog is a download of how we are trying to see and amplify the learning that comes from the network of UNDP’s Accelerator Labs. We’re trying to figure out how to capture some of the gossamer that results from many teams in hundreds of countries, working on their own problems in their own ways, while constantly interacting and evolving in their own national ecosystems. It’s a lot of moving parts! Capturing a sliver of the knowledge lab teams create is a contribution to our larger intent of finding some needed breakthroughs to make the world we live in a more equitable, sustainable, and safer place.
What knowledge is being created in the network of labs?
Our network of social innovation labs seems to be creating a range of forms of knowledge. For the purpose of creating clarity, I’ll frame the four types we are focused on:
- We generate experiential knowledge created on how to deploy a range of innovation methods: what is the most effective way to convene a hackathon? What are the benefits and limits of behavioral insights experiments? What does a data powered approach to positive deviance look like? This is an essential kind of knowledge as we test out and codify new ways of working within global development.
- Our solutions mappers and their teams also generate forms of ethnographic evidence that unpack the practices, knowledge, and inventions of women and men who are innovating around problems they face in their communities. Inspired by the Grassroots Innovation Augmentation Network, we try to learn from frugal innovation to inform the design and delivery of development programmes.
- As a sensory platform, we see and cultivate weak signals of change to understand the real time state of sustainable development. How is the pandemic unfolding and affecting people’s livelihoods? What does the future of work look like? Weak signals of change are a form of intelligence that this lab network creates.
- And when we run experiments, we probe systems to see what effect our interventions make. Here we create and share experimental knowledge. This is where we are trying to learn how to document tests and probes so that practitioners across the world can pick up where another lab team left off, modifying to suit the contexts that make or break the utility of knowledge transfer.
The Network Learning Prototype
With these forms of knowledge on our radar, and knowing that what we are trying to tap into is constantly changing and evolving, we set out a roadmap. It goes like this, although not necessarily in linear fashion:
- The labs reflect on their work in free form. We’re tapping into what the labs write as a start, though there are multiple forms of reflection and audio is probably next on deck.
- Machines help us find the signposts. Given the massively distributed way of working across the network, we need some help from our [machine] friends to know where to look to find patterns in all the written reflections.
- Humans read for patterns. We read the reflections the labs create and try to draw out insights, nuggets of learning, and patterns across the 91 lab teams.
- We set an R&D priority and learning questions. Our first priority is to learn more about informal economic activity — we’re currently seeing that it doesn’t go away with GDP growth and needs some new approaches.
- We convene ‘learning circles’ to explore learning questions.
- We blend the experiential knowledge coming from the labs with the body of knowledge in research circles.
Seeing the synapses created by a network
In the early days of 2019, when our Lead Learning Designer Bas Leurs was onboarding new teams, we brought them together in bootcamps. I’ll never forget the message the Mexico lab team stuck on the wall of the bootcamp room on their way out the door after being the first to undergo our onboarding. As they were about to become the first ever Heads of Exploration, Experimentation, and Solutions Mapping in their country’s UNDP office, they wisely conveyed that what they needed more than advice from us at HQ was a network of others in the same position. They shared their phone numbers and set up Whatsapp groups for explorers, experimenters, and mappers.
The WhatsApp channels grew organically on top of the networking tools we had planned: weekly scrums and Microsoft Teams channels, plus a mandate to work out loud through public blogs and other media. We are now sitting on top of a network that chatters. It creates about 400 blogs a year and interacts with up to almost 300 messages a day. (Clearly, one needs to manage one’s notifications accordingly!)
In 2020 our Lead Data Scientist Jeremy Boy began to build tools for analyzing the text reflections of the labs, and he is now building a data governance compact to ensure safe and equitable use of the data, as well as language models and evolving taxonomies to structure the chatter of the network. Inspired by the work of Edgeryders, we try to use these tools to point to signposts of where the nodes of the network are at any given moment. As the labs decide what they work on and how, these tools point us to where to look for opportunities for convergence in knowledge creation. For the video inclined, check out how Jeremy and I presented this back at FWD50 back in 2020.
Machines point to patterns. Then we explore learning questions
So the synapses of the global lab network are firing, but what are the patterns? This is where the humans get back into the game. We use the advanced search tools of lots of unstructured text reflections (blogs, reflections, Mural boards, etc.) to see patterns.
For example,we started to see that across many areas of lab work, informality kept popping up. Our Kenya lab was working on youth employment, and saw informal waste management as a growth area for jobs. The Ghana team was working on circular economies, and again saw informal waste picking as an invisible part of how value is created from waste. The Paraguay team worked with the Ministry of Labor to set up an informality lab to understand how this dynamic shows up in construction, retail, and other areas.
Many roads were pointing to informal sectors as the great unknown in development. When we looked at available policy offers, we saw gaps: most assume that those in the informal sector need protection, but they don’t see the ingenuity, innovation, and creativity in the sector.
We dug into lab experiences and several other signals across the lab network and used an adapted Johari window to define the edge of knowledge when it comes to informal sectors. Thanks to our Learning and Experimentation Specialist, Eduardo Gustale for driving this process together with our Communications experts, Erika Antoine and Angelica Gustilo Ong.
We then took what we (think we) know about informality and went to the edge of what is known. This is where we set out research questions including:
- What perceptions and motivations drive informality?
- Is there an unseen contribution to environmental sustainability that comes from the informal sector?
- What does the rapid digitization brought on by the pandemic mean for informal entrepreneurship?
- Is there a way to approach informality through hybrid policies that protects people from hazardous labor and taps into informality’s ingenuity at the same time?
We call them learning questions so that we emphasize that knowledge can come from many sources, but they function like research questions. And while we let the knowledge bubble up, we know that we need to dig deeper than written reflections to really get to the most useful knowledge. So we are experimenting with how we convene learning circles to keep the conversation going and unearth the tacit knowledge. More to come on the what and how of learning circles in a future installment…
What do we produce out of all of this?
Admittedly, we are doing two things at once, which comes with some exasperation at times! We’re testing out a network learning prototype that meets the needs of the 21st century by tapping into distributed and often tacit intelligence. And at the same time we are seeking insights related to our R&D priority: informality as part of sustainable development. In truth, it sometimes feels like we are hacking our way through a jungle, changing tools and directions pretty frequently.
But we’re getting somewhere under the broader calling of collective intelligence. With our network learning prototype, we now have a model in motion to transform centrally driven knowledge management and R&D into a distributed model that acknowledges diversity and continuous change across multiple local contexts. This is a mouthful… the pithier version is “the world’s largest, fastest learning network on sustainable development challenges…” We continue to try to live up to that hype!
This graphic of deliverables is helping us find our way as we let knowledge emerge from a decentralized network, while still trying to capture and spread it:
We’re following uncodified, tacit knowledge generated by practice by using advanced search tools to help knowledge about sustainable development emerge. This happens through unstructured written reflections and exchanges. We then struggle to capture this emerging knowledge in a way that doesn’t ignore the context of the labs who created the knowledge, but also moves towards cross-country heuristics. (This part is tough, and we are open to advice if you’ve cracked it!). Then the hope is to spread and blend that knowledge with what our research partners know. Stay tuned for our first such R&D partnership with the Global Partnership for Informal Transportation starting this fall.
Join us as we fumble and occasionally stumble on clarity as we test out a form of knowledge curation suited to help a 21st century organization know what it knows. There are a lot of learning theories out there and the field moves fast, so if we’re missing references, work previously done or useful theoretical frameworks… reach out. We don’t know it all and we’re cool with admitting it!