As the development pact shifts, can open research help countries to unlock innovation and sustainable growth?

In a time of limited funding, a new development pact that focuses on systems, partnerships and evidence will require a more open research system to ensure impact.

Alice Chadwick El-Ali, Nora Ndege, Jon Harle

A consensus about the future shape of development cooperation is emerging which places genuine systems strengthening at the heart, underpinned by evidence and respectful partnership. There’s a critical infrastructure that these priorities depend on to realise the greatest impact: an open research system.

You can’t build stronger systems, you can’t generate and deploy the best evidence, and you can’t foster the partnerships to do that, when knowledge is locked behind paywalls, data can’t be shared, and partners aren’t working from the same foundations. Each of these three priorities, systems, evidence and partnerships, requires an open research system.

Open research – the missing foundation

The current global research system is not delivering. That paywalls hinder research is well known. But recent work by INASP and APHRC has for the first time mapped this against Africa’s continental development agenda – and the figures are sobering, only 61% of relevant evidence is freely accessible, dropping to 50% or lower in critical sectors like education. The accessibility of research is symptomatic of much deeper problems: Global South researchers are often excluded from participation in global science and face persistent barriers to collaboration and partnerships. Doing research together with policymakers and communities is too often an exception rather than the norm. Science that only targets publication, feeds neither policy nor innovation: it cannot generate the products, services and business that turn knowledge into development.

Open research or open science – represents a fundamental transformation of the research system, making knowledge (publications, protocols, data, code) freely accessible, and, collaborative, and transparent. Open research values indigenous and alternative knowledge and, while it’s not sufficient to shift the way research is done, it creates the conditions for engaging communities, policymakers, practitioners and commercial actors to turn openly available knowledge into public goods, services, and enterprises. Here’s what it could look like.

Access to evidence

The starting point for evidence informed policy and the key to a more equitable system is ensuring evidence is accessible to those who need it.

Example: a district health officer in Malawi is updating clinical guidelines for managing hypertension, the burden of which is rising rapidly across Sub-Saharan Africa. However, most accessible research evidence comes from the Global North and doesn’t reflect drug availability, diet, or health-system constraints in Southern Africa. Regional evidence from Malawian and South African research groups exists, but many articles sit behind journal paywalls that the district office can’t afford. Local data from a nearby teaching hospital remains unpublished as it does not make it past the internal report stage. When the outputs are open: the district health officer can access the evidence from recent African hypertension trials because it is open access, a regional evidence synthesis network has already pooled findings from similar health systems and uploaded their findings to a preprint server, the teaching hospitals reports and datasets are also available in a national health evidence repository. The guidelines the health officer drafts reflect what actually works for patients in comparable settings.

Data to drive decision-making and new technologies

Open data is needed to drive policy, innovation and commercialisation. AI models need to be driven by good quality data and evidence aligned with national and regional languages and priorities.

Example: Rwanda’s Ministry of Finance commissions university researchers to study how to tax digital economy transactions, gathering business income data from five provinces. The findings shape a new policy and get written up in a journal, but then the raw dataset disappears when the project’s funding ends. Three years later, a new tax-compliance unit wants to build an AI tool to flag under-reporting, but there’s no usable local training data. It must either commission the same costly survey again or use training data from other regions, which doesn’t reflect Rwandan digital-economy patterns. When the data is open: the original dataset is deposited in a national data repository with a clear reuse license. The compliance unit builds its AI tool using this data, updating it with newer data as it’s collected. Because the model is trained on data that actually reflects local businesses, it flags anomalies more accurately than a generic tool would, and the country gets far more value from its original research investment.

Co-production of knowledge with society

Open research includes opening up the process of knowledge creation. Working alongside communities from the earliest stages of research design builds trust and increases the relevance and usefulness of evidence.

Example: A research team in Uganda is studying how smallholder coffee farmers can adapt to shifting rainfall and rising pressure from the coffee berry borer, a pest that thrives in warmer, wetter conditions and can wipe out a season’s yield. The research team’s findings, on optimal shade-cover ratios and pest-resistant varieties, are published as a journal article. However, the actual farmers who are the subject and audience of the research never see it and do not shape the research conclusions based on what is feasible within their existing resources. When research is open to society: the research team works with farmer cooperatives from the outset, co-designing trials around techniques farmers can realistically afford, like intercropping with shade trees they already have access to, rather than expensive imported inputs. Because farmers helped shape the research and the results are shared back through cooperative networks in local languages, uptake is immediate rather than dependent on someone eventually reading a journal article years later.

Data and evidence reuse to support innovation

Open research includes permissive licensing of data and evidence to enable others to find it, build upon it and translate it into other applications. This ensures greater value from research investments, enhancing innovation.

Example:a team of biologists document fish species distribution and habitat data across Lake Victoria, shared by Kenya, Uganda, and Tanzania, with fish stocks estimated to have dropped 25% in the last decade under pressure from overfishing and climate change. The dataset they develop stays locked in the original project’s files, leading to a wasted investment once the project ends. When data is open to reuse: the team of biologists share their dataset in an open data repository with clear licensing allowing for reuse. This allows another research team to build on the data and create a tool predicting how fish populations will shift as habitats change. A private fishing-technology company adapts the tool into a commercial risk-planning product for commercial fleets, while the East African Community’s fisheries body uses the same underlying data to set catch quotas and negotiate shared management rules across the three countries. One dataset supports both a viable business and better-informed regional fisheries governance, something that would not have been possible unless the underlying data was open and reuseable.

How Africa is leading on open research

The African Union’s Science Technology and Innovation Strategy (STISA-2034) positions open science as a crucial enabler of the continent’s development goals. A new Vision for Open Research in Africa, endorsed by AUDA-NEPAD and developed by the Science for Africa Foundation (with INASP’s support), is gaining traction in continental processes, with proposals already adopted in East Africa’s AI declaration. The ambition is palpable.

Advancing open research requires more than raising awareness: it is a political and governance challenge, which requires reform of the institutions and incentives shaping how research is funded and conducted – North and South. That must be led by the science councils and funding bodies within countries and economic communities.

But there’s a great opportunity for Northern partners to help advance Global South priorities by supporting the open research systems that they’re building and ensuring they are in the driving seat. This includes investing in open research infrastructure (e.g., publishing systems and data repositories), supporting governance and policy developments (Open Access, and research assessment), and supporting researchers to build skills for open systems. The Vision for Open Research in Africa and the Africa Union STISA-2034 offer a good starting point to guide these plans and investments.

Declaration on use of AI: Claude assisted with developing the Open Research examples shared in this blogpost.

Photo by Brian Kungu on Unsplash, Dunga Beach, Kisumu, Kenya