Executive Summary

The Learning Collaborative was designed as an experimental model for supporting learning practices of civil society organizations working on transparency, accountability and participation in the global South, and to contribute to knowledge about their effectiveness. Recognizing that learning is essential to effective implementation of transparency and accountability initiatives, the Learning Collaborative tested a novel collaborative model with the following features: leadership by southern-based practitioner organizations and support from academic resource organizations; a focus on peer-based and networked learning initiatives; a horizontal governance structure powered by a facilitator function; and dedicated resources for practitioner organizations.

The membership of the Learning Collaborative consisted of four practitioner organizations (Centro de Estudios para la Equidad y Gobernanza en los Sistemas de Salud [CEGSS]; Center for Law, Justice and Society [Dejusticia]; Global Integrity; and Twaweza East Africa) that were supported by three resource organizations (the Accountability Research Center at American University [ARC]; Massachusetts Institute of Technology Governance Lab [MIT GOV/LAB]; and the Transparency and Accountability Initiative [TAI]) that promote practitioner-based generation of evidence and learning. The Learning Collaborative was launched in 2018 with support from the William and Flora Hewlett Foundation and the Ford Foundation and concluded in 2019 at the end of its two-year pilot phase.

In autumn 2019, the members conducted a reflection and assessment process. We experienced several challenges in implementing the Learning Collaborative, and ultimately this effort resulted in a series of concrete lessons about how (and how not) to support practice-based learning to enhance complex governance interventions. This report explains how and why the Learning Collaborative was initiated and designed; assesses its governance and practical functioning; outlines key results and lessons learned from the experiment; and offers recommendations for future collaborative learning models.

In summary, we assess that the Learning Collaborative achieved significant results at the hub (practitioner organization) level, partial results at the cluster (pre-existing networks) level, and very limited results at the collaborative level. On the other hand, our joint reflection generated relevant insights about the design features of the Learning Collaborative, useful to future generations of learning mechanisms.

Key takeaways from the Learning Collaborative are:

• As a result of the Learning Collaborative inputs, practitioner organizations substantively improved their own learning practices. In particular, they involved a broader range of staff roles in organizational learning and strategic planning, and redesigned their monitoring, evaluation and learning structures. These changes contributed to improved programming and strategy development among these organizations.
• Practitioner organizations introduced new learning approaches within their pre-existing partner networks (clusters). Different types of learning strategies were applied depending on best fit, including joint experimentation, horizontal peer-to-peer exchanges, dissemination of lessons, and capacity building. In many cases, these first-time learning spaces and methods improved the networks’ strategy through tools and procedures for reflection.
• At the collaborative level, we implemented innovative mechanisms to support joint learning, such as exchange visits and peerbased organizational learning assessments. Yet we struggled to make progress at this level – for example, joint learning proposals were developed between members but not implemented. The main reason for this was likely the Collaborative’s emphasis on learning without a substantive focus. In retrospect, this core characteristic made it overly ambitious to both design learning processes and create new knowledge within and across the hubs. Other contributing factors to these failures include uncertainty about continuation of funding, a lack of clarity of the role of the resource organizations, and failure to align new activities with organizational annual work plans.

In our view, the most significant contributions of the Collaborative are insights about the successes and failures of its key design features and how these functioned in practice. Although the Collaborative struggled with some of the higher-level goals, it has underscored that practitioner organizations are more effective if they approach implementation through a learning angle. We also know from experience that joint learning can be more rewarding through shared experimentation and reflection. But it is not easy. As we experienced in the Learning Collaborative, it is a challenge to focus on our own organizational practices, support learning in our networks, and enact joint learning plans to contribute to wider knowledge on effective governance initiatives. Our experiment offers important lessons about how to support collaborative learning going forward.

Full report on the project website.