DSC_0012By: Fayyaz Yaseen, Accountability Lab Pakistan Country Representative.


Building accountability in Pakistan is a difficult, long, and non-linear process. It is highly political and requires a deep understanding of the context, relationships, and incentives. With this kind of work, adaptive learning (learning that leads to real change) is essential. Too often organizations fail to internalize lessons from experience to iterate and improve. Last month at Accountability Lab Pakistan we brought together a group of representatives from Pakistani civil society, academia, media, and donor organizations to discuss how organizations can work to integrate adaptive learning to improve their work for accountability. This is what we “learned about learning” in Pakistan:


First, trust between donors and implementers is key. Pakistani civil society organizations still feel that there is little room for them to experiment within the log-frames and matrices that donors tend to demand of them. It is very difficult to substantially change plans based on feedback from citizens; and almost impossible to change budget allocations in meaningful ways based on an iterative approach to building accountability and transparency. The group agreed that both donors and implementers need to begin to understand learning as a key performance indicator- but that this requires much greater trust than currently exists within the system. At the co-design phase of any programming, honest conversations need to be had about how a program will push the envelope in terms of adaptive learning.

Second, the status quo is ineffective
. Organizations working in the accountability and transparency space agree that linear development models and poorly designed interventions often do not lead to desired outcomes. More significantly, without learning from these failures and changing the approach, mistakes are often repeated. The group discussed how Pakistani organizations- funded by international donors- lack incentives to push for changed practices. One participant stated that there are “too many contractors and not enough architects”. Even where lessons are clear, we need to take it a step further and ensure that the system itself is set up to promote adaptive learning.


Third, learning must be systematic or incentivized. Even where organizations are hoping to learn and improve, internal procedures are often not in place to gather lessons in systematic ways. Sporadic, individual learning informs approaches but staff turnover within development organizations is high and learning often “falls through the cracks” because it is not documented. This is at some level about incentives for learning, which are still not clear enough. As one participant pointed out: “I don’t see any change in communities as a result of development efforts but the absence of change does not affect us as donors and implementers enough to make us alter our behaviors”. In order to make adaptive learning the norm in Pakistan, we need to make sure it is a central part of decision-making and performance evaluation.


Fourth, learning is a political process. Internally within organizations, even with top-down support, the learning agenda is often seen by staff to be superficial. It is seen as an “add on” that is useful for marketing purposes but often not central to the organizational approach. Bureaucratic politics remain resistant to change. Externally, organizations that have embraced this agenda have also found that implementing lessons learned is difficult amidst Pakistan’s varied political dynamics. The participants agreed that we each need to get better at not only adaptive learning but also real-time learning—so that changes to approaches can be made within a given context based both on experience and immediately as challenges arise.


Finally, learning needs to be made easy. To ensure adaptive learning becomes part of the way organizations in Pakistan operate, it has to be low-touch. Civil society representatives often complain about the complex, English language reporting processes to which they are beholden, which are difficult and time-consuming. Collecting and synthesizing these reports is also burdensome and expensive for donors. Pakistan is traditionally an oral society, and one donor shared how they have experimented with collective storytelling sessions as a way to gather information from grantees. At the Accountability Lab, we run Quarterly Impact Calls, analogous to quarterly earnings calls for businesses, as a quick, easy way for our partners to speak publicly about the challenges they face and the impact and learning they are generating. The more we can get creative about the process of learning in Pakistan, the easier it will become.