The humanitarian sector does not suffer from a lack of knowledge in the same way it once did. Over the past decades, evaluations, reports, research, practice, and learning networks have generated a growing body of evidence about what works, what does not, and where the sector continues to fall short. Yet the existence of all this knowledge does not automatically lead to constructive learning and change.
This tension was at the centre of a conversation hosted by Lars Peter Nissen from the Trumanitarian podcast, with Juliet Parker from ALNAP and Sarah Abdelatif from Propel. Both work in the space of learning, but they come from different angles (here this conversation is reproduced as a condensed blog). ALNAP as a long-standing learning network for the humanitarian sector, and Propel as a platform that helps organisations capture, organise, synthesise, and reuse learning in practice. The conversation raised an important question: if the humanitarian sector already knows so much, why does it still struggle to act on what it has learned?
The learning challenge has changed
ALNAP was created in the mid-1990s, as part of a wider push for better learning and coordination after the Rwanda genocide evaluation. Its purpose, Juliet explained, was rooted in the idea that the humanitarian sector “mustn’t repeat the same mistakes” and has an “obligation and accountability to learn,” particularly collectively.
The learning challenge looks different today. At the time, one of the main challenges was that the humanitarian sector did not have enough shared evidence and collective reflection. As Juliet put it, the fundamental challenge has shifted: when ALNAP was formed, “there just wasn’t enough knowledge and evidence.” Today, however, many humanitarians feel “bamboozled by the sheer volume of evidence that exists,” and struggle to navigate or even access it.
Juliet identified three key barriers that shape the current learning challenge. The first is inequitable access. Not enough people contribute to and access documented knowledge. The spaces where learning is produced and shared are still too narrow, often shaped by where resources, language, and power sit. The second is information overload In many areas, there is now an endless amount of research, evaluation, documentation, and learning material. The sector has produced so much learning that it can be difficult for busy humanitarians to engage with it meaningfully. Instead of enabling learning, the volume of all the material can become another barrier because it has become impossible to process and use the available knowledge. Thirdly, she mentioned the tendency to approach learning through a predict-and-control mindset. Organisations often try to define in advance what needs to be learned, how the knowledge will be captured, and how it will be used. But learning is rarely that linear. As Juliet explained, even when the sector has evidence about what needs to change, the process of creating change within teams, organisations, and the wider system remains complex and not straightforward.
Learning is never neutral
The complexity Juliet describes is therefore not just about how difficult it is to organise, access, or apply evidence. It is also about power. Learning involves choices about whose knowledge is recognised, which insights are prioritised, and who has the authority to turn evidence into action. Sarah emphasised this point: “In the end, someone is making a decision about who we are listening to, what knowledge counts, who has access to what knowledge, and what we learn,” she said. This matters especially in relation to some of the humanitarian sector’s most persistent challenges, such as localisation, accountability to affected populations, and the link between relief and development. Topics that have already been discussed for the past decades. Yet, as Juliet pointed out, “the recommendations in many areas are almost word for word the way they were 20 years ago.”
The problem is not simply that the sector lacks commitment to these issues, or that they are not being discussed. Rather, these issues touch the underlying business model of the humanitarian system. As Juliet explained, “the sector was built for speed and scale, and it’s delivering on speed and scale.” However, many of the changes that are being demanded now require more contextual and relational ways of working. The sector has often been able to learn and improve in more technical areas, such as cash programming or health interventions. But it has struggled to change in areas that challenge existing power structures.
This is why the topic of learning cannot be separated from questions of localisation and accountability. If learning spaces are dominated by the same actors and definitions of expertise, they are unlikely to produce the kind of change the sector needs. A more inclusive learning system requires a broader range of voices and experiences to shape what is known and what is done next.
Funding cuts and the need to protect collective spaces
Inextricably linked to the business model and the questions raised above is the current funding environment. Inevitably, the conversation turned to how humanitarian and development budgets are being cut, while crises continue to grow. This creates a difficult and uncertain moment for the sector. Sarah suggested that this pressure may force organisations to think differently. With fewer resources, they may become more open to experimentation, new ways of working, and tools that help them make better use of the existing knowledge.
Juliet shared a related but more cautious concern: “I’m worried that this increased competition for resources between institutions will cause them to be more focused on their survival and less focused on collective issues.” Competition for funding could make it harder for institutions to step beyond their own mandates and invest in collective learning. At exactly the moment when the sector needs shared spaces for reflection and change, those spaces may come under threat. She adds, “I really think that we have a duty to champion the collective spaces and the collective initiatives and to protect them, because once we lose them, they'll be very difficult to get back. And collective learning is critical within that.”
The sector is changing, but not fast enough
Despite these pressures, both Juliet and Sarah recognised that change is happening. There are shifts in conversations around localisation, adaptive management, and more flexible funding. There are also stronger voices coming from national and local organisations, challenging the sector. But the pace of this change remains slow. Juliet expressed the need for tighter learning loops. “The world is changing rapidly, crises are becoming more complex and frequent, with climate change adding even more pressure. And we just don’t have tight enough learning loops to be able to recognise, adapt, and change in response to the speed of change in the world”, she explained.
Sarah added that local voices are getting louder, and technology has made it easier for people to connect across borders and contribute to wider conversations. She said, “meaningful change has to happen both from the top down and the bottom up, with both sides meeting somewhere in the middle.” At the same time, she acknowledged that this is difficult because powerful actors, including donor countries, have “interests at play that help keep the system in place.” Rather than pretending these interests will disappear, Sarah argued that the sector needs to have honest conversations about how money flows, whose interests shape decisions, and how louder local voices can be brought into the conversation rather than dismissing one or the other.
Towards stronger learning loops and shared learning spaces
If the sector needs tighter learning loops and more inclusive learning spaces, then the question becomes practical as well as political: what kinds of infrastructure make it easier for organisations to access, connect, and act on what they already know?
This is where Propel enters the conversation. Propel was created in response to a problem many organisations face: knowledge getting lost. It gets scattered across reports, emails, meetings, spreadsheets, and individual memories. When asked what concrete solution Propel offers, Sarah explained that the platform puts organisational learning processes into “an online collaborative environment.” She adds, “organisations can plan what they need to learn, capture knowledge across teams, and use AI meaningfully to help categorise, organise, and synthesise insights. So instead of just storing information, organisations can build on the knowledge and experience they already have”. It is a new way to connect the knowledge in the sector and make the memory of the humanitarian sector accessible.
For Sarah, the practical role of Propel is connected to the broader learning challenges Juliet described. If learning spaces are too narrow, how can technology help diversify the voices that are included? If knowledge is trapped in long reports or inaccessible files, how can it become easier for different people across an organisation to engage with it? If learning is scattered, how can teams connect insights across projects, languages, and contexts?
At the same time, Sarah was also clear that a tool cannot replace the deeper thinking required around learning. Propel is not a substitute for organisational change. She explains, “Propel needs to be embedded in wider shifts in how organisations approach learning. The role of the tool is to make that process easier. It reduces practical barriers, such as having to search through hundreds of report pages or thousands of lines in a spreadsheet.” It can help create more accessible and collaborative spaces for knowledge to be captured and reused.
From individual commitment to collective accountability
Both Juliet and Sarah mentioned several times during the conversation that many humanitarians are deeply committed to learning. At an individual level, people want to do better and the value of reflection and adaptation is recognised by many. But individual commitment is not enough if the systems around people do not make learning possible. This is why Juliet believes the priority is to strengthen collective accountability to learn. It should not be acceptable for humanitarian actors to repeatedly fail to act on what they already know. For her, this means “creating the cultures and leadership expectations needed to take unpopular decisions, challenge the status quo, and apply past learning even when doing so is not necessarily in an organisation’s own interest.” The challenge is to do better justice to the wealth of learning expertise and experience that already exists across the sector, and to ensure that this knowledge can meaningfully inform future action.
Sarah emphasised that the process of change also requires patience. She explains, “learning takes time, especially if the goal is to include more diverse voices and move collectively. You can’t expect that to happen overnight.” But that does not mean individuals are powerless. Change also depends on people across the system continuing to put learning, accountability, and power-sharing on the agenda, rather than treating systemic problems as someone else’s responsibility. This is especially important for actors who already hold more access, resources, or influence within the current system.
The future of humanitarian impact does not depend on producing more reports. It may actually depend on building stronger, more inclusive learning loops: spaces where knowledge can be shared, connected, and acted upon. Making that work requires tools. But it also requires culture, leadership, accountability, and a broader understanding of whose knowledge counts. The sector already knows a lot. The challenge now is to make that knowledge more accessible, more collective, and more useful for change.