Knowledge Management

In creating the SELCO Foundation, Harish Hande’s objective was to grapple with the complex array of issues that support or impede how DRE could aid in development work. Hande had seen how other organizations had tried to implement DRE solutions in challenging contexts, focusing on the product side of the equation, and ultimately not meeting the needs of the end-user of solar energy. His experience in solar energy had showed Hande that an approach that centered on the user and built an ecosystem around an intervention would be more likely to be successful and scalable.

But from a knowledge management perspective, the ecosystem approach was difficult to administer. SF staffers faced challenges coordinating their efforts - for any given intervention, numerous partnerships with a wide array of actors had to be developed simultaneously. Knitting together these efforts, especially across rough terrain that limited communication, created difficulties.

If internal communications were hard, SF’s goal of diffusing its approach to the wider development world was even more challenging. Hande saw SF as an “action tank,” carrying out various interventions not just to aid groups of recipients, but also to provide data and insights for those interested applying in this approach in other settings. But the Foundation’s bespoke method to developing its interventions meant that every project had unique elements. Nonetheless, Hande believed that the projects could produce relevant guidelines and insights that could be utilized in other contexts – yielding “Lego Blocks” of information and practices that could be redeployed, as appropriate, in other interventions.

Defining information blocks emerging from SF’s livelihood promotion vertical presented a particular problem, since the means of procuring a livelihood ranged by area and culture. For the education and healthcare verticals, successful innovations were more readily transferred – both verticals were intent on building site-specific facilities (schools and clinics) that would have similar goals no matter their location.

But livelihoods varied greatly from region to region. Were there lessons that SF could impart from working with turmeric processors in Assam that would be applicable to, say, blacksmiths in Patna, let alone cocoa processors in the Ivory Coast? What could be transferable to other contexts? The lessons could vary in terms of their specificity. There was a broad methodology of user-centered design, ecosystem mapping, or rolling out interventions that could be imparted. But could other, more specific lessons be harnessed as well? What about technical innovations, narrative strategies, organizational structures, and business models? How could these Lego Blocks of knowledge be organized and made available to organizations that might be interested in taking a similar approach?

Another issue that prevented diffusion was the dearth of metrics. One reason for the popularity of product-oriented interventions among development NGOs was that metrics of success were easy to track – the number of solar panels installed, or the number of megawatts produced. This gave funders confidence that targets could be set and then met in a reasonable period of time.  The SF approach was more nuanced and required patience for learning by trial and error, for developing the needed ecosystem partners, etc. Building relationships and incubating technologies was an uneven process that could take many funding cycles to get to a point where scalable interventions could be successfully launched. SF was backed by patient funders willing to trust the organization to find its way, but few organizations were as fortunate. Could SF provide a roadmap with appropriate metrics and milestones? Or was the process not amenable to this type of quantification?

SF had worked to make its process as transparent as possible. The Foundation’s website featured documents derived from its work – such as examples of problem statements and ecosystem reviews. SF had also created a website for the various livelihood innovations that its partners had created, providing a catalogue of devices that would work with electricity from solar panels. In a more far-reaching project, SF, in 2023, began experimenting with artificial intelligence, allowing the program to not only digest the foundation’s documents, but also allow it to record and transcribe meetings for its database. The AI program was showing promise in helping internal knowledge management. Could it also provide a useful service to those outside the foundation? What other mechanisms could be used to capture SF’s diffuse learnings and support successful adoption of a user-centric, ecosystem approach?