In a village in central Kenya, Daniel, a coffee farmer, has just completed his fourth training this season. These sessions are part of a growing wave of climate-smart agriculture initiatives (run by NGOs, agritech companies, and government programs) aiming to boost yields and resilience through sustainable agricultural practices (SAPs). For farmers like Daniel the promise of new knowledge is compelling. Extension services are limited, and farmers often attend trainings because there is an incentive (such as input vouchers or seeds) and also because the local extension agent has communicated that it will be beneficial. One workshop introduced composting techniques and hoped Daniel would reduce his reliance on pesticides. Another focused on drought-tolerant seeds and aimed to encourage him to adopt a specific variety promoted by the project. A third demonstrated soil testing with the expectation that Daniel would adjust his fertilizer use accordingly, while the latest one focused on mulching and urged him to include his wife more actively in farm-related decisions. 

All of them, he recalls, were run by different entities, none of whom seemed aware of the others. Some messages overlapped, others conflicted. One encouraged the use of synthetic fertilizer; another discouraged it. By the end, Daniel feels unsure which practices are right for his farm. He’s grateful for the attention, but also overwhelmed. 

‘Too many people come to tell us what to do,’ he thinks. 

‘Then they leave.’

Daniel’s story is not unique. Across East Africa (and many other regions), development actors deliver well-intentioned but fragmented interventions. In areas perceived as hotspots or easy to access, the presence of NGOs, donor projects, and government programs can be dense (Koch & Ruben, 2008). In others, farmers are left behind entirely. And where the state cannot provide consistent extension services, these imbalances deepen. The result is a patchwork of effort: uneven, uncoordinated, and sometimes counterproductive.

The problem isn’t just that efforts are duplicated; it’s that they often speak in different languages. Farmers are left to make sense of conflicting advice from multiple actors (Lamm, et al., 2020). As one observer put it, without coordination, the risk is not just redundancy but confusion, where farmers hear contradictory recommendations that erode trust and discourage adoption of new practices.

Trust and consistency are also fragile. Farmers often describe the fatigue of revolving-door initiatives, with NGOs and companies that arrive for one or two seasons and disappear just as quickly. Such cycles weaken rather than restore the sense of ownership, reinforcing the need for consistent, long-term presence.

This is more than a communications issue. Behavioral science tells us that humans have a limited capacity to absorb and act on complex or conflicting information. Information overload doesn’t empower people, it paralyzes them. When extension services or technology dissemination are poorly tailored to farmers’ context, adoption rates can stall . And in the context of climate adaptation, where action is urgent and adoption of new practices is critical, disengagement is costly. Several studies show that when extension services are poorly tailored to farmers’ circumstances, adoption of promoted practices can stagnate (Mugari et al., 2025; JPAL, 2023).

Farmers, however, are not passive recipients of these interventions. Daniel, like many smallholders across the region, constantly weighs new recommendations against his own experience: his soil, his labor availability, his financial constraints, and the risks he knows well. He adopts what seems workable, modifies practices to fit local realities, and disregards advice that feels misaligned or contradictory. But agency does not automatically translate into well-informed decisions: when the information landscape is fragmented or inconsistent, farmers make the best choices they can within imperfect conditions, which can sometimes lead to practices that cancel each other out or undermine resilience. The challenge is therefore not a lack of agency, but a lack of systems that recognize that agency and support it with coherent, reliable, and aligned guidance.

The consequences of this fragmentation ripple outward, as donors and implementing agencies unknowingly duplicate efforts. Resources are wasted. Communities become fatigued or skeptical. Progress on shared goals slows, despite the appearance of intense activity.

The promise of collaborative governance

This problem is not new, unique to development or the Global South. Recognizing these challenges, many turn to collaborative governance as a solution. The logic is sound: bring organizations together under a shared strategy, coordinate interventions, reduce duplication, and amplify impact. There exists a plethora of frameworks on global governance (Ansell & Gash, 2008; Emerson, Nabatchi & Balogh, 2012), and different variants of it. However, not a lot of them are adapted to East Africa. Amongst the most famous, we can find public-private partnerships, where governments and private sector partners share risks, resources and benefits to deliver infrastructure or services (World Bank, 2025). Another well-known model in the development sector would be consortia – voluntary alliances of organizations that pool resources and expertise to pursue a shared goal while maintaining institutional independence (Oxfam, 2016). While less common in the sector, other models such as the triple helix model – innovation ecosystems where universities, industry, government, civil society, and the environment interact to co-create development and innovation-  have also been tested and implemented (Gatune, 2023). Our previous research has similarly pointed to co-opetition as another model present in practice, though rarely documented in the region, and often subject to the same pitfalls of competing priorities, unequal power relations, and administrative burdens that characterize more formal frameworks (Oberoi et al., 2023). All of those examples highlight that there is a belief in the potential success of such models. 

However, as mentioned by Macharia (Oxfam, 2016), such models often fall short of that promise. Evidence has revealed common pitfalls: slow decision-making, competing priorities, unequal power dynamics, and administrative burdens that weigh down smaller partners. In one study of a consortium strategy in Kenya, local actors reported being sidelined during key decisions, while the reporting demands of the consortium left little room for adaptation or community engagement (Osoo, 2016).

This isn’t a failure of intent. It’s a failure of design. When collaborative governance is formed hastily, often in response to funding opportunities, there’s little time to align visions, anticipate governance challenges, or design mechanisms for inclusive decision-making. The result can be a collective that exists more on paper than in practice: a group of organizations working under the same banner, but still largely in silos.

Rethinking collaborative governance

Coordination is not a logistical challenge. It’s a human one. This is where a shift in mindset could help. What if we stopped thinking about coordination as a structural fix and started thinking about it as a design challenge?

Before forming a consortium or launching a multi-actor initiative, what would happen if we paused to ask: how will this impact the people we’re trying to reach? Will they receive a coherent, manageable set of messages? Will their feedback shape the way our organizations collaborate? Are we designing a system that includes invisible, local voices, or simply replicating the same top-down dynamics under a new logo?

Drawing from human-centered design principles, we might treat coordination not just as alignment among organizations, but as an act of empathy and intentionality toward the communities we serve. That means investing time upfront to map what already exists, understand local information ecosystems, and design joint efforts that are easy for people to navigate and act on.

But shifting to a design mindset is not straightforward. It demands confronting the power dynamics that shape whose knowledge influences decisions. Larger international organizations often have more staff, stronger English proficiency, and greater donor familiarity, which allows them to dominate coordination spaces, even inadvertently. Local partners may lack the administrative capacity to meet reporting requirements, limiting their ability to participate meaningfully. Donor timelines rarely allow for the deep contextual work needed to build trust and shared understanding. These structural and behavioral factors influence not only how organizations collaborate, but what ultimately reaches people like Daniel.

The issue isn’t that coordination doesn’t work. It’s that we too often rush into it with the wrong tools. We treat it as a technical task to be managed, not a behavioral reality to be understood. When multiple actors show up in a village, their good intentions can cancel each other out, unless they speak with clarity, coherence, and care.

If we treat coordination as a design problem, the first step is clarity, not structure. What is the simple purpose we all recognize? Who decides what, by when, and how can a small, local partner raise a concern without friction? Research on collaborative governance shows outcomes hinge less on the banner and more on the quality of shared motivation, joint problem-solving, and visible routines for making and revisiting decisions (Tonelli et al., 2018).

The question is how to make the experience clear for everyone involved. Ask: What keeps power in check? Are there transparent decision notes? Are there clear time limits on approval? Is the leadership regularly transferred? Is there an ‘overlap check’ to avoid duplicating others? What can we stop, standardize or simplify so that partners spend more time delivering than reporting?

Power dynamics are not a side issue; they shape voice, pace, and whose evidence counts, so we should notice and name them early rather than after trust erodes.

Finally, how will we know the system is getting healthier? Pick a few humane signals: time from idea to decision, proportion of decisions made closest to the problem, and whether partners can state the shared purpose in the same words.

This isn’t a call to abandon collaborative models. It is an invitation to rethink how they are built. Effective coordination does not require uniformity, but it does demand shared understanding, humility, and the center of human experience. Continuity is especially critical: when external actors or NGOs eventually withdraw, farmers and local partners should not be left with the sense that the partnership has abandoned them, but rather with capacities and relationships that endure beyond project cycles.
Climate resilience will not be achieved through more activity alone. It will come when the messages we send are not just well-meaning, but well-aligned, when people like Daniel are no longer confused by what they hear, but confident in what they can do.

This article is part of a broader effort to rethink how we understand and respond to climate change. Too often, climate action is discussed strictly in technical terms, while the human realities that shape whether solutions succeed or fail remain in the background. These questions sit at the heart of Climate change is a human problem, Busara’s new publication that brings together insights from researchers and practitioners working across the development sector. The book explores how behavior, social norms, power, and local knowledge shape climate resilience in practice, and why centering lived experience is essential for designing responses that are effective and equitable. If this piece resonated with you, the full volume offers more perspectives that continue this human-centered conversation. 

References

1. International Association for Agricultural Extension Education. (2020). Gendered pathways and barriers in agricultural extension: A qualitative analysis. Journal of International Agricultural and Extension Education, 27(3). https://newprairiepress.org/jiaee/vol27/iss3/5/
2. J-PAL (Abdul Latif Jameel Poverty Action Lab). (n.d.). Data collection for randomized evaluations. https://www.povertyactionlab.org/print/pdf/node/4525
3. Maastricht University. (2023). Triple helix model for rice value chain in Kenya. https://cris.maastrichtuniversity.nl/en/publications/triple-helix-model-for-rice-value-chain-in-kenya
4. Osoo, V. (2017). Challenges of implementing consortium strategy in development projects at ViAgroforestry, Kenya (Master’s thesis). University of Nairobi. https://erepository.uonbi.ac.ke/bitstream/handle/11295/99369/Osoo_Challenges%20of%20implementing%20consortium%20strategy%20in%20development%20projects%20at%20viagroforestry%2c%20Kenya.pdf
5. Oxfam. (2016). Working together in consortia: A practical guide for civil society organisations in Kenya. https://oxfamilibrary.openrepository.com/bitstream/handle/10546/610858/cs-consortia-kenya-310516.pdf
6. PPP Knowledge Lab. (n.d.). Introduction to public-private partnerships. https://ppp.worldbank.org/PPP_Online_Reference_Guide/Introduction
7. Ricciardi, V., Waddington, S., & Giller, K. (2025). Participatory methods in sustainable food systems research: The role of community-led data collection. Frontiers in Sustainable Food Systems, 9, Article 1618938. https://doi.org/10.3389/fsufs.2025.1618938
8. Schut, M., van Paassen, A., & Leeuwis, C. (2008). Beyond the project: Innovation capacity in agri-food systems (PEEG No. 0814). Utrecht University, Department of Human Geography and Planning. http://econ.geo.uu.nl/peeg/peeg0814.pdf
9. Smith, K. B. (2008). Administrative behavior and bias: An institutional perspective. Journal of Public Administration Research and Theory, 18(4), 543–557. https://academic.oup.com/jpart/article-abstract/18/4/543/1090370
10. Weiss, C. H., & Tschirhart, M. (2012). Public administration research: Challenges for the next decade. Journal of Public Administration Research and Theory, 22(1), 1–7. https://academic.oup.com/jpart/article-abstract/22/1/1/944908