Is the District Transformation Project Living the Idea of Problem-Driven Iterative Adaptation?

Written by Hifza Azeem

Why are countries not developing? Even when governments, non-profits, and policymakers develop policies and programs with a clear intention to solve a real-world problem, many fail to achieve their intended outcomes.

Multiple studies show that even well-designed programs do not translate into expected results. Throwing light on this, Angrist and Dercon (2024) point to a stubborn gap between design and delivery, in which programs lose implementation fidelity. Century et al. (2010) argue that implementation fidelity is critical for understanding whether interventions achieve their intended outcomes. Complementing this, May (2014) notes that implementation failures often emerge from administrative outcomes and governing arrangements that undermine policy commitments. Therefore, the challenge lies in ensuring that the developmental policies and programs are implemented as effectively as possible in the complex system.

Implementation research describes this gap as a result of how programs interact with various governance structures and real-world conditions. According to UNICEF (2024), as policies turn into programs and are delivered through administrative systems, they engage with many actors at different levels. This impacts how the programs are interpreted, adapted, and often reshaped by local contexts. Additionally, Braithwaite et al. (2018, as cited in Hudson et al., 2019) note that implementation is a complex, non-linear process shaped by the constant friction between institutions, incentives, and political realities. This complexity means programs rarely follow the expected path. Hwa (2025) further highlights that in Global South education systems, middle-tier officials, the very actors responsible for implementation, routinely carry overlapping mandates, face ambiguity around their roles, and operate with fewer resources than required, all of which significantly compromise implementation success.

While existing research documents the prevalence of implementation gaps, it often focuses on the symptoms instead of the underlying cause. To understand why systems consistently lack the internal capacity to execute programs over time, Andrews et al. (2013; 2017) argue that the problem arises when states fall into capability traps through isomorphic mimicry, which is the tendency to adopt externally defined ‘best practices’ such as policies, laws, and organisational structures to appear legitimate without considering their appropriateness or compatibility. Eventually, this charade replaces actual performance. This process creates ‘capability traps’ where the organisation or state remain stuck in low performance because they assume adopting best practices will result in effective function, even when system actors lack the capacity to implement these practices effectively.

Once the state or organisation has fallen into this trap, Andrews et al. (2013; 2017) argue that standard reform efforts, such as more training, do not help. To overcome capability traps, they propose Problem-Driven Iterative Adaptation (PDIA) to rebuild state capacity. PDIA is a process-oriented strategy. It is designed to build state capability by addressing ‘wicked hard’ problems that are difficult to solve, and are shaped by local context. It seeks to address these challenges through iterative learning and continuous adaptation.

Andrews et al. (2013; 2017) outline four key principles of PDIA:

1. Local Solutions for Local Problems: Begin by deconstructing the problems identified within the local context. Local actors should define what the problem is, why it matters, and to whom, instead of borrowing best practices.

2. Pushing Problem-Driven Positive Deviance: Build and sustain an environment that allows teams to test new ideas and encourages learning. This includes recognising and supporting effective practices that already exist but are not widely adopted.

3. Try, Learn, Iterate, Adapt: Take small steps to test ideas, learn from experience and improve over time. Ongoing feedback helps adjust approaches and make them more suitable to the context.

4. Scale through Diffusion: Engage with a wide range of stakeholders across sectors to ensure solutions are accepted, practical, and can expand and sustain over time.

Although research on PDIA is still emerging, early evidence suggests that this approach can significantly improve implementation in certain contexts. For instance, in Kenya, the implementation of the Legal Assistance for Economic Reform (LASER) program illustrates how PDIA was applied to improve an underperforming judicial reform project. The practitioners refocused on local problems as a priority, such as dispute delays, consultations, feedback-based action, and restructured the program to be more flexible and responsive to local needs (DFID LASER, 2015). Similarly, evidence from Lawson and Harris (2023) in their assessment of the Public Finance Management (PFM) Capabilities program across six African countries suggests that PDIA can yield short-term benefits, particularly in areas where there is strong support and a dedicated reform team.

While these examples show the potential of PDIA, its effectiveness depends heavily on the context in which it is applied. This makes it important to examine how these principles translate to the complex development or administrative systems present in India. 

India’s governance structure is complicated and multi-layered. While policy decisions are centralised, implementation is decentralised. As Mangla (2015) notes, bureaucratic norms (unofficial rules) are a key factor in how well agencies provide services. In states with legalistic norms that strictly follow hierarchy and procedures, officials struggle to adjust policies to local conditions. Ranjan (2021) highlights that India’s bureaucratic structures create constraints for implementing PDIA. However, the focus on decentralised decision-making and moving away from uniform, top-down solutions reflects a need for approaches that prioritise local problem-solving, such as PDIA. 

Understanding whether this approach can function within existing institutional constraints is important because implementation realities often shape whether policy goals translate to meaningful outcomes. One way to explore this is by examining programs that attempt to work through existing public systems.

The District Transformation Project (DTP), run by Leadership for Equity, is one such program. While DTP was not explicitly designed as a PDIA initiative, its implementation features, such as district-level ownership or context-responsive decision making, align with core PDIA principles. Hence, examining it through a PDIA lens is useful as it operates within the same institutional constraints that make adaptive reform difficult in India.

DTP is implemented across five districts of Maharashtra: Pune, Nashik, Nagpur, Palghar, and Ahilyanagar. Aligning with the NIPUN Bharat Mission, it focuses on improving foundational learning outcomes for students and enhancing teacher competencies. DTP operates across existing governance structures such as SCERT, DIETs, and Zilla Parishads. It functions through three interconnected pillars: 

1. Teacher Professional Development: By analysing national assessment data (such as PARAKH and ASER), and third-party assessments (through organisations such as Convegenius and IDInsight), we identified priority competencies a teacher needs in literacy and numeracy. These competencies were converted into structured training modules and delivered to Teacher Mentors (TMs) such as cluster heads and resource persons using a cascade model.

2. Teacher Mentor Professional Development: We trained TMs to observe classrooms and provide mentorship to teachers using the Teach Primary tool developed by the World Bank. These regular cycles of observation and feedback helped improve everyday classroom practice.

3. Strengthening Governance: We established Project Management Units (PMUs) in districts to strengthen the governance processes. These PMUs enabled officials to identify key instructional gaps and move discussions from administrative reporting to focused problem-solving by using data in District and Block Quality Cell meetings.

Through this approach, DTP aims to show how improving district-level systems can lead to sustained improvements in foundational learning outcomes and attempts to build the capacity of the system itself to function better.

Let’s assess DTP on the four core principles of PDIA:

• Local Solutions for Local Problems: DTP uses student assessment and teacher competency (internal pre-and post tests) data to identify learning gaps and shape training content. The program design and intervention structure are informed by local data, policy priorities, leadership alignment, and existing district systems. The program strengthens ownership by working through existing structures, such as District and Block Quality Meetings, and Shikshan Parishads, rather than creating parallel structures. However, DTP is led by an external organisation, where its local actors, even with alignment, do not fully own the process.

• Pushing Problem-Driven Positive Deviance: DTP uses data and feedback to understand what works, but it focuses on following a pre-defined theory of change with planned interventions, and does not give opportunity to amplify local practices.

• Try, Learn, Iterate, Adapt: Training, mentoring and governance processes are continuously adjusted using data from assessments, observations and evaluations. The program is informed and improved through continuous feedback loops. These feedback mechanisms allow the program to learn and evolve based on evidence.

• Scale through Diffusion: DTP works within government systems, which helps integrate practices and support scale. However, long-term sustainability depends on whether systems can continue performing without external support.

Ideally, DTP does not live the idea of PDIA, but it does not exist in a vacuum either. It falls somewhere in between, as a structured program that has unintentionally adopted the essence of PDIA, without fully embracing its method. However, examining DTP helps us understand how adaptive and problem-driven principles may operate within large public systems. DTP can be understood as a form of ‘partial PDIA’, and in the Indian context, where implementation often operates within administrative and bureaucratic constraints, this may be the most realistic version possible. 

Andrews et al. (2013) acknowledge that PDIA can be utilised across a variety of approaches, and is not a fixed model but a set of principles. Therefore, the tension surrounding DTP is merely a challenge that most reform efforts face due to the complexities of bureaucratic systems. In such conditions, sometimes it is essential for a structure to exist. However, if left unchecked, the structure can limit the adaptations it aimed to support. 

Another concern is that DTP may fall into the same trap that it is trying to solve. If DTP’s feedback loops become regular reporting exercises, if their governance meetings return to compliance discussions, and if their training modules stop evolving, DTP will engage in some form of isomorphic mimicry of PDIA. It will merely mimic PDIA without capturing its essence. Lawson and Harris (2023) found this pattern in their study, where short-term progress did not last because the system had replicated the form of adaptive practice without internalising the purpose of adaptive practice.  

What matters more than how DTP looks against a PDIA checklist is whether districts are developing a genuine capacity to learn. Are officials starting to take ownership of identifying problems? Are governance structures able to learn without prompts? If the answer is yes, then DTP is heading in the right direction. If no, then even its best practices risk becoming the next generation of borrowed solutions, well-intended, evidence-informed, but ultimately unable to survive beyond the timeline.

DTP exemplifies how district systems can be supported to improve, but whether that shift becomes permanent depends on the capability of the system once the program ends. 

References

1. Andrews, M., Pritchett, L., & Woolcock, M. (2013). Escaping capability traps through problem-driven iterative adaptation (PDIA). World Development, 51, 234–244. https://doi.org/10.1016/j.worlddev.2013.05.011

2. Andrews, M., Pritchett, L., Woolcock, M. (2017). Building state capability : evidence, analysis, action. In BiblioBoard Library Catalog (Open Research Library). https://doi.org/10.1093/acprof:oso/9780198747482.001.0001

3. Angrist, N., & Dercon, S. (2024). Https://www.wwhge.org/resources/understanding-Gaps-between-policy-and-practice/. https://doi.org/10.35489/bsg-whatworkshubforglobaleducation-wp_2024/04

4. Century, J., Rudnick, M., & Freeman, C. (2010). A framework for measuring fidelity of implementation: a foundation for shared language and accumulation of knowledge. American Journal of Evaluation, 31(2), 199–218. https://doi.org/10.1177/1098214010366173

5. DFID Legal Assistance for Economic Reform Programme. (2015). Kenya Case Study: Exploring how to use a problem-driven iterative adaptation approach in restructuring pre-existing institutional reform programmes. https://assets.publishing.service.gov.uk/media/591d67cf40f0b63e08000042/q6_laser-kenya-case-study-final-november-2015.pdf

6. Hudson, B., Hunter, D., & Peckham, S. (2019). Policy failure and the policy-implementation gap: can policy support programs help? Policy Design and Practice, 2(1), 1–14. https://doi.org/10.1080/25741292.2018.1540378

7. Hwa, Y. (2025). Entry points for supporting middle-tier officials in foundational learning reforms: Equip, connect, inform, empower. https://doi.org/10.35489/bsg-whatworkshubforglobaleducation-ri_2025/005

8. Lawson, A., & Harris, J. (2023). Is the problem-driven iterative adaptation approach (PDIA) a panacea for public financial management reform? Evidence from six African countries. World Development Perspectives, 31, 100526. https://doi.org/10.1016/j.wdp.2023.100526

9. Mangla, A. (2015). Bureaucratic norms and state capacity in India. Asian Survey, 55(5), 882–908. https://doi.org/10.1525/as.2015.55.5.882

10. May, P. J. (2014). Implementation failures revisited: Policy regime perspectives. Public Policy and Administration, 30(3–4), 277–299. https://doi.org/10.1177/0952076714561505

11. Ranjan, U. (2021). Problem-Driven Iterative Adaptation - Accountability Initiative. Accountability Initiative: Responsive Governance. https://accountabilityindia.in/blog/understanding-problem-driven-iterative-adaptation-and-its-limitations/

12. Samji, S., Andrews, M., Pritchett, L., & Woolcock, M. (Eds.). (2018). PDIAToolkit. https://thepolicypractice.com/sites/default/files/2023-03/PDIA%20Toolkit.pdf

13. UNICEF Innocenti: Global Office of Research and Foresight. (2024). Implement: Bridging the gap between policy and practice – Synthesis of UNICEF Innocenti research on education in Africa. UNICEF Innocenti. https://www.unicef.org/innocenti/media/10286/file/UNICEF-Innocenti-Bridging-Policy-Practice-Gaps-Education-in-Africa-December-2024.pdf