Success factors for communities of practice - evidence from the World Bank

Communities of Practice are a promising intervention in Aid and Development, but what do they need to make them work?

Give a person a fish, you feed them for a day. Teach them to fish and you feed them for life. Set up a Fishing Community of Practice and you have the core of a thriving and sustainable industry that will last for generations.

That is the appeal of CoPs in the development sector. The development agency becomes not the provider of resources, nor the provider of knowledge, but the means by which the developing industries can manage their own knowledge and co-develop their industries.

But how well do these communities work, and what do they need to succeed? This is the topic of a 2005 World Bank study by Johnson and Khalidi, which looks at successes and failures in 3 communities of practice in the Middle East. The lessons they draw from this study as listed below, and even through the study has a limited sample, the conclusions match what we already know from in-house communities. I reproduce the lessons below, some of the commentary from the report, and some commentary of my own in italics.

1. The most important issue determining a CoP’s success is leadership. A committed, energetic leadership is vital. The CoPs that specifically allocated funds to community leadership witnessed more dynamic activity over the course of the two years, whereas the activities of the third CoP (which did not apply the seed funding to staff, but rather covered these costs through the support of their organizational host) have all but stopped.  We know that leadership is crucial (this was number 7 out of the 10 success factors in our survey) and CoP leaders with training and enough time to do the role is success factors 3 and 4 from the Warwick business school survey. See also this post. Community leadership is crucial.
2.  An organic need for networking is another critical success factor. Two of the pilot CoPs had a clearly identified demand from their members, who had requested a formalization of interactions. The third CoP did not tap into an existing network, but rather sought to create a new network. This more supply-driven approach had mixed results. Again survey results show that a demand-based community is likely to be more successful than a supply-based community, and this applies to CoP events as well as to online forums. The best communities are those which answer the questions and solve the problems of their members.
3. A solid issue/knowledge base is needed before a community will coalesce. Moreover, interaction should be based on questions that lead to something concrete, like publications or face-to-face meetings, to help the community gel. Again we see the use of questions, in this case to create content. I would disagree that you need content before a community can coalesce, but I do agree that you need one or more issues. 
4. Technology may not play a large role in networking.  This is a large conclusion from a small dataset - two of the communities attempted virtual interaction using e-discussions and both decided not to attempt another e-discussion. However the reason for the failure of the discussions is unknown. Each community held successful workshops; one held a large workshop but had no follow-up, while the others  had more modest workshops with many side discussions. Each had a static website with minimal new content added (perhaps as a result of the failure of online discussion), one of which has been completely stagnant since its creation. I would say that all these World Bank communities succeeded through workshops, but failed to back them up through technology. The conclusion that "technology may not pay a large role" is therefore untested, and our survey showed that "a mechanism for interacting online" was the number one success factor for in-house communities of practice. The world bank study shows you can have some success without technology, but has not tested the power of technology to, for example, support community interaction between the workshops. 
5. There is a strong cost-benefit argument for supporting CoPs for knowledge exchange and learning. The US$20,000 invested in each CoP has had a far greater impact than had it been spent on a time-bound learning event. CoPs have the advantage of being active and adapting over time. The value proposition for a CoP is obvious, also the means of value delivery, and this blog contains many examples of quantified benefit from KM, many of which were delivered by CoPs. However this is still a mechanism which the Aid sector seems to be struggling with, and evidence of value such as shown in this article is very welcome. 
6. A limited understanding of what a CoP entails can significantly affect the relevance and quality of CoP activities. CoPs can be easily mistaken for short term activities, meaning CoPs may get started but that they will not last. Donors who are thinking of supporting CoP activities should be aware of this, and adjust their expectations accordingly. Communities of practice are not a short term solution. They grow, mature and evolve over years and decades, and it is in their later life that the greatest value is delivered.