Imagine trying to navigate a big city with dozens of bus and metro lines – but with no map, no signs and no app. Technically, the system is there. But finding the right route for your journey is frustrating, slow and sometimes impossible.
For many teachers, school leaders and policymakers, this is what using education research can feel like. There is more evidence than ever, but it is scattered across journals, databases and websites, often written in technical language and not clearly linked to the questions people have in their day-to-day work.
Over the past few years, several new developments in the “evidence movement” have started to redraw this map. At the OECD, we call our contribution to this work the Evidence Web for Education (EWE) – a long-term programme to make it easier to find and use research in education policy and practice.
What is still missing to put evidence to good use?
To build evidence systems that genuinely support educators and policymakers, we need to strengthen three interconnected areas.
- Matching: Too often, evidence does not address what people are actually asking. Many reviews still focus on “what works”, while offering far less on how something works, under what conditions, at what cost, or how it might be adapted to a different context. This reflects a broader problem: the research community still lack a deep understanding of users’ needs.
- Navigation: Evidence is spread across numerous repositories, each presenting different types of syntheses and using varying terminology and formats. Without shared standards for communicating the strength and nature of evidence, users struggle to compare information across platforms and interpret what they are reading and how it applies to their context.
- Collective learning: Many “knowledge intermediaries” and evidence brokers would benefit from more opportunities to learn from each other and from practical tools to design, implement and assess their strategies to support evidence use. However, initiatives developing infrastructure in this area lack coordination and alignment.
In short, we are still asking a lot from people who want to use research.
What is changing in the world of evidence?
While the “evidence city” is still hard to navigate, several profound shifts are already transforming how evidence is produced and shared.
1. From supply-driven research to user-centred questions
For a long time, research agendas were driven mainly by researchers’ interests. Now, more governments and organisations are asking what teachers and policymakers actually need to know. Some countries are mapping ministries’ and practitioners’ questions, or co-producing research with schools to ensure that studies genuinely respond to real-world challenges.
2. From narrowly focused studies to “living” evidence syntheses
New methods and AI tools are speeding up research synthesis. Instead of waiting years for updates, “living” reviews and evidence maps can be refreshed continuously as new studies come out. That means decision makers can rely on more up-to-date, comprehensive pictures of what we know – and where the gaps are.
At the same time, the methods behind this evidence have matured. After years of debate about whether “what works” trials fit the complex reality of classrooms, research methods now mix quantitative and qualitative approaches and include new designs that can explore how and why programmes work, and under what conditions, even when big experiments aren’t possible. While these methods are still not widely used, the field’s methodological foundations are far stronger than a decade ago.
3. From information “push” to systemic support for use
We have learned that simply publishing research reports is not enough. People need time, skills, incentives and organisational support to use evidence well. This has led to a greater focus on knowledge mobilisation: the structures, people and processes that connect research to practice and policy, from intermediaries and resource banks to professional learning and system-level enablers.
4. From isolated initiatives to global infrastructure
Finally, there is a growing global effort to build shared infrastructure: common databases with unified taxonomies, shared standards for evidence synthesis, and AI tools that can scan, review and summarise research from across contexts. This is supported by networks that bring together funders, evidence intermediaries and experts across sectors.
Taken together, these developments could fundamentally change how easily policymakers and practitioners can access high-quality evidence.
Why this matters
These developments will not only affect how countries make policies and teachers teach; they will also reshape the core of what organisations like the OECD do. As global infrastructure and AI tools become increasingly capable of rapidly synthesising and summarising research, then organisations like the OECD need to sharpen their focus on coordination, standard setting and tailored support to ensure that evidence is trustworthy, relevant and genuinely useful for decision making.
Through the Evidence Web for Education, the OECD is aiming to do exactly that because, ultimately, building an impressive evidence “city” is not the goal. What matters is that teachers, school leaders and policymakers can find the routes they need – quickly, confidently and in time to make a real difference for students.
Further reading:
- OECD Evidence Web for Education programme
- ESIC (2025), Evidence Synthesis Infrastructure Collaborative
- UKRI (2025), AI investment to transform global policy with scientific evidence
- White, H. (2022), The unfinished evidence revolution: riding the five waves, Centre of Excellence for Development Impact and Learning (CEDIL)
- Jacobs Foundation (2025), Synthesis-Ready Evidence Repository