by and

Filed under: Development Development 2.0 Social innovation

For the last two years we’ve been experimenting with a variety of methods and approaches to development that were, up until recently, entirely new to us – from gamification to open policymaking, from behavioural science to user-driven innovation, from design thinking to micronarratives and real time monitoring.

In total, we ran around 60 interventions following a portfolio approach, the assumption being that some high risk interventions would fail while others might lead us in unexpected directions.

We are about to embark on an independent evaluation to see whether our initial hypotheses were correct.

While we cannot claim that our early work was totally exempt from pilotitis, we like to think that we set the basis for taking our innovation early work to the next stage – and the organizational learning that goes with it.

And this is where we are encountering an interesting challenge. If we follow the common innovation parlance, it is time for us to focus on “scaling up” (or “scaling out”) – with an implicit (and sometimes explicit) expectation that this means going bigger – investing more money, replicating one success to more countries and cities.

We found this type of linear, manufacturing-style approach of running-one-good-project-and-then-exporting-it-to-someplace-else not in line with some of our own experience in applying complexity science in projects:

As Anna Davies at the Young Foundation put it:

“… the concept of scaling has strong connotations of standardization. It has its origins in manufacturing, where the aim is to achieve economies of scale, by spreading fixed costs across more units of output. But in the messy social field, the potential for standardization is more limited. Here, concepts of reinvention and adaptation will be at least as important, if not more so, than standardization. Social outcomes are not products that can be easily made to formula and packaged. This is especially clear in the context of innovation in public services.”

To make an analogy, Galapagos finches didn’t get a memo one day instructing them to come up (or scale up) to 17 different beaks that will allow for a more efficient food gathering, but the different beaks in 17 different types of finches evolved through time and depending on where different finches lived and what type of food they eat.

In our practice, we are increasingly learning that to be effective, we need to spend far more time ‘listening’ and sensing where the system is and where it is moving to (some would call it monitoring), with constant probing (or prototyping) as a way to understand the issue better and inform our next move.

Our experience seems to chime in with others’. Paina and Papers put it in their analysis of scaling up efforts in the health sector, we need to shift “from the current models around scaling up .. which revolve around linear, predictable processes, to models that embrace uncertainty, non-linear processes, the uniqueness of local context and emergent characteristics.”

Similar considerations emerge, to a name a few from Lant Pritchett’s work on systems thinking and education and institutions and isomorphic mimicry, Owen Barder’s lecture on applying complexity theory to international development and Duncan Green’s reflections on redefining project design.

Where to start if we want to develop something akin to evolutionary principles to re-frame the approach to growing successful interventions?

Our plan is to bring together players from cognitive and complexity science, biology, design and development practitioners to compare notes on how they look at “scale”, but also, importantly to answer some very practical questions, such as:

  • How to set up a project management system for development projects that mirrors evolutionary principles of variation, selection and adaptation?
  • If we were to design a “scaling up” fund for development, what criteria would we use to stay away from the “replicating best practice” curse and adopt evolutionary approaches?
  • How can we design an intervention that helps us move from rhetoric to action on feedback loops?

If you are interested in joining the journey, please get in touch.

  • Alberto Cottica

    Of course we are interested. It’s high time to either deeply reform the concept of scale or abandon it altogether. And yes, biology is a good source of metaphors.

    • Millie

      Albert, will take you up on the offer to band together on this!! 🙂 We’re just realizing the extent of the challenge ahead and are super excited about a possibility to redefine the metaphor all together!

  • Tess Newton Cain

    Very interested. Here in the Pacific, there has been and continues to be a propensity on the part of donors to adopt a cookie-cutter approach to development with insufficient attention to establishing an in-depth knowledge of each environment with its own story of history/politics/culture and so on. ‘The Beak of the Finch’ is very applicable in our part of the world on numerous levels

    • Millie

      Dear Tess, great to have support from your end of the world! Will use this space for updates on where we are with this idea but would love to have you on board. Most critically, i think, it will be important to have a balance between a conceptual discussion on where development can borrow and should borrow from other disciplines and turning it into something very practical. Thanks again for the comment and i hope to continue this discussion!

  • Marcus Jenal

    Yes, very interested. I am doing a lot of work in market systems development and we face the same challenge – markets being social systems after all. Scale is supposedly reached by finding a good (business) model and then others will automatically crowd in on this new idea. Not working too well so far. An evolutionary perspective on change is definitely needed here!

    • Millie

      Marcus, thanks so much for your comment and encouragement. We’re excited to take this step forward and see what we learn and can apply from other sectors. And of course, we’re super happy to have others join with their opinions and examples (good and bad!).

  • Women♥Tech Europe

    Very interesting indeed. As ECWT we work on engendering innovation policies and we constantly face the bottleneck of the “best practices” approach predominant also in gender mainstreaming policies. Contrary to any basic learning theory where crisis and failure are actually the triggering factor for learning, the focus on lessons to be learned seems to better serves institutional marketing than change…

    • Millie

      Thanks for the reply on the post! Yes, exactly, our intention is to move away from ‘best practices’ as simply taking a success from one place and transplanting it somewhere has proven to be futile. So the question for us is how to distill experience from other practices and sectors to development in a very practical, hands on way that can help us speed up evolution so to speak in our work in the field. Stay tuned though, we’re excited that the idea resonates with so many people and we’re excited to learn from experience of others!

  • Would enjoy knowing how I can help support or participate.

  • Graham Leicester

    Very interested in this. Whether we consider the process of transition over time from the dominance of one pattern of activity to the rise of another pattern as either ‘scaling’ of the new pattern or ‘evolution’ of the old, a critical issue is the mechanisms, structures, values etc that guide the ‘selection’ process. Those that fit the selection environment best will survive and thrive, but who determines the rules? This is a central inquiry at International Futures Forum. We are concerned that ‘innovation’ processes relying on or mimicking the market tend to draw public and social innovation backwards towards improving the status quo (see for example this post on ‘crossing the chasm’ in the public sector: ). And in another strand of work, closer to your interest in complexity science etc, we are exploring how we might design in a complex operating environment for what might be called ‘intentional’ emergence, eg set up the initial conditions such that some things are more likely to emerge than others. So – I would love to get involved with others in this dialogue if possible.

    • Millie

      Graham, thanks so much for the comment!! Intentional emergence is exactly what we’d like to translate into some very practical criteria that a project manager working in the field could apply to his/her work and use it as a filter to back up those things that evolve. Would love to catch up with you on this, what’s the best way, email? (im at:

  • Nelson T. Enojo

    So simple. What if fighting poverty & protecting the planet is one and the same?

  • NM Horacio

    Hey, interesting perspective. However, I wonder why does it take so long to realise these kind of things. I mean, from my perspective and experience, people adapt themselves to new situations and conditions and survive. Through their daily practice, with the resources they have at hand, people (by themselves) adapt to the social, economic and environmental conditions, this is evolution. More than 50 years of “development” have passed and we are just realising that “one-size-fits-all” solutions do not work. Why? because -again my view- we do not need to “scaling up”. Still today, We as dev. practitioners, policy makers and scientists believe we are here to give solutions to people, to design processes for them, to find alternatives for them. Of course there are other kind of experiences, but in general the “development business” is what it used to be, with changing discourses depending on what sells more: sustainability, participation, inclusion, etc, etc… Here is where I like this notion of evolution. Instead of scaling up in a lineal and bottom-up way, I think we need to “spread the word” in more horizontal, messy, complex, informal, non-standardised ways. That is exactly how evolution works!
    If we are in search of a project management system that mirrors evolutionary principles, we need to do what biologists do: sit and observe! understand what is happening, see what people actually do, understand what people actually want…
    Taking the three aspects given in the article: variation, selection and adaptation.
    Variation: “one-size-DOES NOT-fit all” because we live in a world that is highly heterogeneous. Even a small community is highly heterogeneous. When looking at the practices, at what people actually do, then we realise that people engage in so many activities for coping with pressing conditions. This heterogeneity can serve as a source of ideas, solutions, alternatives, inspiration, hope. Within this diversity there are people who already found creative solutions, this would be the starting point for “development” design.
    Selection: as in nature, where the strongest survive, in the social wild, although more complex, we could talk about the most creative, innovative, “sustainable”, solution or alternative survives. People do not do what is told to do, and that is fine, we tend to select what we consider is best for us, based on our knowledge, experience, culture, traditions, etc… Again, we should stop doing doing prescriptive, obtrusive and westernised interventions and do more reflexive observation, understanding and facilitation.
    Adaptation: Well, there is not much to say about this. I mean, we just need to see how people is surviving and understand how they managed, by themselves, to do so.
    Anyway, this is taking long.. I will stop here. And yes! I am very interested in joining this journey…

    • Millie

      Horacio, sounds like we’re preaching to the converted 🙂 I really like your comment as it depicts where we are at the moment, and the real question for us is the ability to translate some of these concepts into practical criteria that we can apply in development. In terms of whether ‘scaling up’ is what we need, i am completely with you… one of our allies on this initiative suggested we change the name of a Scaling Up Fund to the Finch Fund 🙂 will keep updating this space with new movements and i look forward to having more discussions on this with you, and support!

    • Marcus Jenal

      Dear Horacio, this is a great comment and I can subscribe to it 100%. The only thing that I would like to add to the discussion – and the thing I have been struggling with myself – is the question about what role our own values play in this process. Because we do have values and we cannot turn them off. In fact, our values are often responsible for getting us into this job in the first place. When reading through your comment (and many similar things I have written for that matter), it always feels like we are only facilitating in a neutral way and we want everything to emerge from within the system, as this is the ideal. Don’t impose solutions from outside, but let them evolve from within. In any case, we always have a bias towards what we want to see emerge. We want to see less poverty, more equality, less violence, more democracy, etc. So in a way it’s not a pure evolution, but it is very much a guided evolution. Guided by our own values. Is this a bad thing? I don’t know. Maybe not. Imagine that we are facilitating an economic development process in a rural area, let’s say in the dairy sector. The work goes great and the local actors start to work much more closely together and more milk is flowing. We will, however, not be happy with this if we see that only men are profiting from this and if we see that women have to carry the whole burden (more work additionally to what they have been doing anyway, no decision about how to spend the increased income, etc.). This does not really comply with our values. Another example would be if we create new economic opportunities for businesses through our facilitation but we find out that poor workers are being exploited. Better to be exploited than ignored? Not sure about that. Not every progress is ‘good’ progress. Not sure if this makes sense and as I said, I’m struggling myself with this a lot.

  • Ian Toal

    One of the problems about using evolution as a metaphor is that the process of evolution is value-neutral – there is no end point, not goal. The assumption that the best comes out on top is not actually correct. The best, the strongest, are not always the successful. Part of the reason we have come to believe this is that humans consider themselves as the ‘top of the pyramid’ when in reality we just have a set of character traits that allows us to be fairly numerous. So do bacteria.
    With that caveat, I can see two ways of letting a project ‘evolve’. One is to transplant it from one place to another, and allow the people there to modify the things they like or don’t like about it. Most successful programs have something that makes them successful, so taking that to a new place (unless it is a completely foreign concept) should at least form a basis for a program in that area. ‘Western’ health care in an indigenous community has a lot to offer, but if the program is not open to the incorporation of traditional healing practices it may not be successful. Flexibility may be the key.
    The second is to allow something to grow organically from the area, or to modify something that is already there. Traditional healing can be supplemented by other health care services. This takes longer, and may or may not be successful.
    Anyway, just some thoughts. Interesting ideas, though.

    • Millie

      Ian, thanks!! Raising few issues we’ve been grappling with internally and that i very much look forward to hearing what people far smarter than i am have to say about. My main question is how to design a process of decision making internally (eg. for us in UNDP for example) that will put us in a better position to ‘invest’ in those initiatives that have that 1 thing that would be worth transplanting as you say- so how do we come up with a criteria to evaluate what that one or two things are? On your 2nd point- that’s a really important one as it doesnt get at HOW to scale up but WHAT to scale up, and for us the main question here is whether we use ‘scaling up’ for initiatives that come out of a particular enterprise (eg. again, UNDP) or whether we set the criteria and a burden on the one who wants to invest in ‘scaling’ to look out for the most advanced and promising solutions outside the enterprise. Ultimately and if we are to have some success in supporting initiatives that tend to be most promising inthe local context, we’ll have to find a way to translate this into some very practical set of rules-of-thumb that we can apply in our decision making.

      • Ian Toal

        Hi Millie: I don’t believe I’m far smarter than anybody, but I’m always willing to put my two cents in.
        Using the evolution analogy again, it is hard to know which parts of an initiative will ‘take root’. Some imported organisms thrive, while others die out, and it is not always possible to predict which will work. Who would have predicted that mobile telephones would become so popular, and important, to people in developing nations?
        Some guidelines are possible, though. If you are going to transplant an initiative, a good place to start is how similar the cultures are. If I’m going to take a rural development program from the US to Canada, the likelihood of it working are good – the cultures are fairly similar. The same program may not do as well in Mexico, and would likely do really badly in say rural India. So cultural similarity would be a good place to start, and depending on the project, physical/geographical similarity too.
        In medicine there are often decision trees that are based on sets of symptoms. If the person has these symptoms, X is most likely wrong with them. But W, Y and Z also have those symptoms, so those are considered but only after the most likely is ruled out. I could see a similar decision tree for a development initiative. The program would have to define in some way the core concepts of the program, what needs to succeed otherwise it becomes worthless, then assessing those concepts against a set of criteria that are likely to create successful outcomes.
        It’s kind of hard to talk about this in the abstract – is there some way that you could give me an example so I can think about it?

      • Ian Toal

        Further to below, the concepts that ensured mobile phones would become so important are human need to communicate, convenience and affordability. Those are pretty basic human traits. Examining the concepts that make a program successful can help with adapting it to fit a different place.

  • Nelson T. Enojo

    Hello Millie,
    We envy your humility. I love your compassion. What I see is the increase in volunteerism and involvement. That’s the end result of education, learning. Evolution… Perspective… May we share a simple drive to step up our compassion: What if protecting the planet and poverty alleviation is one & the same?
    Thank you so much.