The Learning Trap: Investing in Social Sector Capacity Building

By Oluwagbemibori Obasola, Founder & CEO, Boundless Gem Advisory

Every year, funders and philanthropic intermediaries invest millions of dollars in capacity building for social impact organizations within their portfolio. Other organizations build the capability in-house, the way our clients do when they invest in their own teams directly. Either way, workshops are designed, facilitators are hired, and participants travel, sometimes across borders, to absorb information that is supposed to transform how their organizations operate. The facilitator is skilled, the content is relevant, and the room is engaged.

The problem is what happens after the room disperses.

Research on training transfer is unforgiving on this point. Retrieval and application, not passive exposure, are what make learning stick. Passive absorption creates the feeling of learning. It does not create the practice that makes learning durable.

In many cases, the knowledge was real, and the resources provided were thorough. However, the architecture of the intervention was mismatched to the goal of producing lasting organizational change, and that architecture is what ultimately determines whether learning transfers into practice.

This article draws on adult learning research, behavior change literature, and evidence from capacity-building programs in low- and middle-income country contexts to examine why the dominant model of capacity building in this sector may underperform, whether funded externally or built internally, and what a stronger learning architecture looks like.

What Workshops Were Designed to Do

The foundational problem with workshops as a capacity-building vehicle is a design mismatch. They were built to deliver information. Funders, and organizations investing in their own teams, expect them to change behavior.

A workshop that leaves participants informed, even inspired, has done its job as designed. Whether participants do anything differently back at their organization is a separate question, one most workshop designs never answer. Traditional training stops at recognition and recall, but organizational performance depends on application.

Transfer-of-training research has known this for decades. Baldwin and Ford’s (1988) foundational model identified three factors that determine whether training sticks: the trainee, the training design, and the work environment the trainee returns to. Of the three, the environment carries the most weight, because it acts directly on whether the new behavior is maintained, while the training itself only shapes what was learned in the room. If the old behavior is still the default, and the new one takes more effort, the training will not transfer.

For social impact organizations in resource-constrained settings, this problem compounds. Lean teams, fragmented schedules, and no internal learning infrastructure to reinforce what was covered. One team member attends a workshop, comes back energized, and is immediately pulled into the operational urgency of running programs on unpredictable budgets. Without reinforcement built into the design, the learning evaporates.

What Learning Science Tells Us

Roediger and Karpicke’s landmark research in Psychological Science (2006) demonstrated that active retrieval produces substantially greater long-term retention than passive re-reading, a finding further quantified in subsequent work showing retrieval-based study produced approximately 50% better comprehension outcomes than elaborative study techniques (Karpicke and Blunt, 2011). Kolb’s experiential learning cycle, published in 1984 and reissued in 2014, articulates the same finding in a different framework which states that learning happens in the loop between concrete experience, reflective observation, abstract conceptualization, and active experimentation.

The implications for capacity building design are significant. Learning with practice produces way better retention and transfer than passive absorption of information. Practice is not a nice-to-have in addition to learning; it is where the learning actually happens. This means spaced practice built into regular routines, feedback that corrects drift before it compounds, and peer accountability structures that sustain momentum beyond the training event itself.

Why This Matters More in Frontier Contexts

The failure of workshops to produce lasting organizational change is a global pattern, but its consequences are disproportionately severe in low- and middle-income country (LMIC) contexts. A 2024 study by Social Lens, examining more than 100 social impact organizations across India, found that despite the majority having prior experience in capacity building programs, structural and practical barriers continued to hinder the adoption of new skills and strategies (IDR Online, 2025). In LMIC settings, learning is typically acquired by one or two individuals with no peer learning mechanism, no internal knowledge-sharing infrastructure, and no designated follow-through process. The learning lives in a notebook or a slide deck and rarely becomes organizational practice.

A second dimension specific to these contexts receives almost no attention in the mainstream capacity building literature: the relationship between organizational beliefs and learning transfer. Organizations in frontier markets where funding systems were not designed with their contexts in mind frequently carry what behavioral economists would describe as a scarcity orientation. This is a rational adaptation to consistently inhospitable operating environments, and it shapes how organizations engage with new learning in ways that are directly relevant to capacity building design.

The COM-B behavior change model, developed by Michie, van Stralen, and West (2011), identifies three domains that must be present for behavior change: Capability, Opportunity, and Motivation. When the barrier is a motivation or opportunity gap rather than a capability gap, training consumes resources without moving the behavioral needle (Aim for Behavior, 2026). For many organizations in LMIC contexts, the primary obstacle is the belief about what is possible for organizations like theirs, and effective capacity building must address that layer before or alongside the method.

What Strong Learning Architecture Requires

Overall, research on adult learning, behavior change, and capacity building in LMIC contexts points to these five design principles for programs that produce lasting organizational change:

Diagnostic work must precede prescriptive work.
Programs that begin with a structured pre-work diagnostic, mapping where each organization actually is rather than assuming a standard starting point, produce more relevant and transferable learning. The diagnostic primes active engagement from participants before the program begins.

Mindset must be addressed before method.
The most durable capacity interventions in LMIC contexts examine organizational belief systems before introducing new skills. This is consistent with Mezirow’s transformative learning theory (1997), which holds that lasting learning requires a shift in the frames of reference through which learners interpret their experience.

Cohort learning outperforms solitary attendance.
When participants learn alongside peers facing similar challenges, social proof reinforces transfer in ways solitary attendance cannot replicate. Lightweight peer accountability structures produce meaningfully better outcomes than programs that end without follow-through commitments.

Application must begin before the program ends.
Participants should leave with work already in progress. For lean-team organizations especially, the transition back to operational demands is abrupt, and the window for applying new learning closes fast.

Structured reinforcement is not optional.
The strongest reinforcement patterns are structured but not elaborate. E.g., a micro-practice plan, a paired accountability partner, and a post-learning debrief.

The Cost of Getting This Wrong

The consequences of poor learning architecture in this sector are usually invisible. Social sector organizations do not dramatically collapse after a workshop. They keep operating as they did before, slightly more informed, but behaviorally unchanged. The funder’s investment is spent but the problem the funder hoped to solve stays exactly where it was.

Malipatil and Brainard’s work at Overdeck Family Foundation, published in SSIR (2023), points at this directly. Their most effective capacity-building support did not stop at a training session. It stayed engaged with portfolio organizations as they moved from insight to a changed strategy.

Ford Foundation’s BUILD program also shows this at scale offering multi-year, unrestricted funding paired with sustained institutional strengthening. NIRAS’s 2025 evaluation found that 83% of Ford Foundation’s grantee partners said BUILD contributed to their mission impact to a large extent.

Organizations doing the hardest work in the world’s most under-resourced contexts deserve capacity building held to the same evidentiary standard as the programs they deliver. That means:

• Designing for behavior change, not information transfer.
• Addressing belief systems, not just skill sets.
• Building reinforcement into the architecture instead of hoping motivation carries what design didn’t.

The organizations doing this work already care about getting it right but the real question is for funders, intermediaries, and capacity builders. Is the learning architecture we’re offering built to make that possible?

About the Author

Oluwagbemibori Obasola is the Founder & CEO of Boundless Gem Advisory, a strategy and institutional strengthening firm working with social impact organizations and the funders who support them across Africa, Asia, Latin America, and beyond. 

She has lived on both sides of the funding relationship, as a grant-maker and as social sector leader, before transitioning into an advisor to organizations across more than 40 countries. Her proprietary RAISE framework integrates money psychology, strengths-based diagnostics, and evidence-driven storytelling into a structured learning architecture for the social sector.

Learn more about her work via: www.boundlessgemadvisory.com