From Fragmentation to Integration: Strengthening the Science-for-Policy Ecosystem in the European Union

POLICY BRIEF STG Policy Papers Authors: Florence School of Transnational Governance SCHOOL OF TRANSNATIONAL GOVERNANCE Jakov Bojovic, Imogen Bayley ISSUE 2025/21 DECEMBER 2025 FROM FRAGMENTATION TO INTEGRATION: STRENGTHENING THE SCIENCE-FOR-POLICY ECOSYSTEM IN THE EUROPEAN UNION Views expressed in this publication reflect the opinion of individual authors and not those of the European University Institute. SCHOOL OF TRANSNATIONAL GOVERNANCE Strengthening evidence use in policymaking is critical for addressing complex transnational challenges and restoring public trust in government. While the EU has established numerous Science-for-Policy organisations, the ecosystem faces significant challenges: fragmentation across institutions, limited coordination, and insufficient interaction between researchers and policymakers. Despite high citizen demand for evidence-based policy — with 68% of Europeans believing scientists should intervene in political debates — researchers report a lack of regular platforms to engage with policymakers. This brief presents three policy proposals to address these gaps: 1. the creation of an integrated knowledge platform that combines responsive and challenge-led research with interactive public consultation features; 2. the development of structured research-to-policy fellowships that embed researchers within EU institutions; and 3. the adoption of innovative research communication methods, including professionalised knowledge brokering, science-policy labs, and reformed academic incentives that reward policy engagement. These recommendations build on successful models from EU Member States and other countries and align with the European Research Area’s commitment to advancing the Science-for-Policy ecosystem. EXECUTIVE SUMMARY Authors: Jakov Bojovic | Research Fellow, Florence School of Transnational Governance, EUI Imogen Bayley | Max Weber Fellow, Max Weber Programme for Postdoctoral Studies & Florence School of Transnational Governance, EUI 3STG | Policy Papers Issue | 2025/21 | December 2025 1. BACKGROUND Policy issues today are transnational, complex, and interconnected. To address them effectively, policymakers must have a broad and interdisciplinary understanding of their causes, insight into multiple possible and innovative solutions, as well as a deep analysis of their costs, effects, and potential trade-offs. High-quality evidence is critical throughout this process. Such evidence can come from within the government or from outside of it. Still, the translation of evidence into policy is far from automatic, and a higher provision of evidence does not necessarily result in better, more evidence-based policy. Its use will depend on many contextual factors. Where and when the evidence comes from, how it is presented, and how it relates to beliefs, values, financial limitations, or public opinion are just some of the key considerations that transform evidence uptake into better regulation and policymaking. Moreover, the OECD and the European Commission have established that strengthening the use of evidence in policymaking is a key driver of trust in governments. Yet, trust remains critically low: 44% of citizens of OECD countries report a low level of trust in their governments, and only 41% believe their government “relies on the best available evidence, data, and statistics in taking decisions.” Significantly, the OECD Trust Survey demonstrates a clear correlation between citizens’ belief in evidence-based decision-making and their overall trust in government, making the current fragmentation of the EU’s Science-for-Policy ecosystem a matter of democratic legitimacy and not merely technical efficiency Similarly, studies have also suggested that citizens require scientists play a larger role in the policymaking process, which would represent a step toward making policies more evidencebased and thus restoring trust. Cologna et al. surveyed around 72,000 respondents in 68 countries and found that 83% of respondents agree that scientists should communicate more with the general public, and 54% agreed that scientists should work more closely with policymakers. Finally, a Eurobarometer survey has found that 68% of Europeans believe scientists should intervene in political debates to ensure decisions are evidence-based. To systematically address the questions of evidence use in policymaking and the factors that facilitate and hinder it, the European Commission has launched a series of projects and publications on Science-for-Policy ecosystems. In 2022, the European Commission’s Joint Research Centre (JRC) launched a twoyear project on ‘Building Capacity for EvidenceInformed Policymaking in Governance and Public Administration in a Post-Pandemic Europe’. This project conceptualised Sciencefor-Policy as an ecosystem, developed ways of measuring its capacity, and assessed the state of seven national ecosystems of EU Member States. In 2023, the Competitiveness Council Conclusions called “on the Commission to further develop the concept of ‘Science for Policy’ and to promote the role of scientific and evidence-based knowledge and its crosscutting integration in public policies.” Finally, in 2025, the Council endorsed “advancing the European Science for Policy (S4P) ecosystem” as one the key priorities of the European Research Area. According to the Commission, a Sciencefor-Policy ecosystem is an ‘’interlinked set of institutions, structures, mechanisms, and functions that interact at different levels to provide scientific evidence for policy.’’ This concept links well with previously established concepts in both literature and policy practice, such as evidence-informed policymaking (EIPM), Knowledge Transfer (WHO), and science-for-policy interface (UN). In 2020, at the United Nations Highlevel Political Forum, ministers committed to “strengthen[ing] the science-policy interface through evidence-based policymaking,” and in March 2021, the United Nations Department of Economic and Social Affairs published a strategy guidance note on the science-policy interface. What unites these concepts is the understanding that strengthening evidence use in policy can be achieved only by a systematic effort to bridge science and policy. 4STG | Policy Papers Issue | 2025/21 | December 2025 Central to this effort is the amount and quality of interaction between evidence providers and policymakers. These can be direct interactions (face to face), indirect interactions, financial interactions, or staff interactions. In this ecosystem, policymakers benefit from interactions with various types of evidence providers: knowledge generators, knowledge synthesisers, knowledge brokers, and policy evaluators, both inside and outside of policymaking bodies. Despite this high promise, there are many challenges in building up a well-functioning Science-for-Policy ecosystem to deliver this promise. These challenges can occur at four different levels: • presence, distribution, and types of actors; • formal and informal functions and roles; • competences and skills of policymakers and evidence providers; and • institutional and individual incentives. Without going in depth on these four layers, it is important to stress that an evaluation of an ecosystem should start from the mapping of the different actors, including people (such as scientific advisors) and institutions (such as government planning units and think tanks). The Commission presented a full list of relevant actors in a 2024 publication. These actors’ functions and roles can be decorative or substantial, which makes a big difference in their perception and impact. Their competences and skills can be lacking, especially as policymakers and scientists are ‘’two distinct communities with different timeframes and approaches to uncertainty, language, incentives, goals, and professional culture.’’ (Commission SWD 2022). Domain Core Question Common Challenges PROCESS (Interaction & Communication) How do actors connect? Linear transmission model; Ad hoc interactions; Shortnotice requests; Lack of informal channels; Reactive engagement; Difficulty defining needs CAPACITY (Knowledge Brokering & Evidence Utilisation) Who facilitates and who receives? Brokers not recognised; No professional development; No career paths; Weak absorptive capacity; Limited evidence literacy; ‘Dual competence’ gap STRUCTURAL (Institutional Architecture) What structures enable or block action? No coordination mechanism; Fragmentation; Political capture; No emergency response; Academic incentives don’t value impact; Independence tensions NORMATIVE (Cultural & Epistemic) What is taken for granted? No shared terminology; Communities disconnected; Evidence for justification > improvement; Disciplinary hierarchies; Patronage > merit; Low trust Table 1: Domains of Science-for-Policy Ecosystem Challenges (developed by the authors) 5STG | Policy Papers Issue | 2025/21 | December 2025 Incentives at both institutional and individual level are also an important consideration to ensure that the impact of scientific advice and evidence is long-lasting. The mentioned 2022–2024 project evaluated the current state of seven national EU Sciencefor-Policy ecosystems and provided the first ever systematic set of empirically tested multi-country challenges. A summary of these findings is presented in Table 1. At the EU level, the Science-for-Policy ecosystem includes a wide range of organisations supporting policymakers with evidence: • the Joint Research Centre (JRC); • the Scientific Advice Mechanism (SAM); • the European Science Advisors Forum (ESAF); • the EU Agencies Network on Science Advice; • the European Research Council (ERC); • DG for Research and Innovation (DG RTD); • the European Research Executive Agency (REA); • the Union Civil Protection Knowledge Network; • the European Scientific Advisory Board on Climate Change; • the European Parliament Research Service (EPRS); and • the European Parliamentary Technology Assessment (EPTA) network. This list does not include advisors to high-level policymakers, special working groups, think tanks and universities that provide research independently, or submissions of evidencebased questionnaires and opinions through the EU’s public consultations platform (Have your say). The evident fragmentation of the EU Sciencefor-Policy ecosystem is one of its main weaknesses. In a 2024 survey, 70.1% of EU researchers agreed that the EU’s Sciencefor-Policy ecosystem is fragmented and has limited coordination of its actors and activities. The absence of a coherent structure or central coordination also contributes to replication and overlap issues within the ecosystem, as well as confusion among citizens, policymakers, and other stakeholders about where to find reliable, policy-relevant scientific advice. This, in turn, can weaken the overall visibility and impact of science-informed policymaking in the EU context, undermining the EU’s ability to respond efficiently to complex societal challenges. Despite the high volume of evidence being created and disseminated and the wide range of actors involved, the most pressing difficulty within the EU’s Science-for-Policy ecosystem remains the lack of sustained and meaningful interaction opportunities between policymakers and researchers. In the same 2024 survey, 71.3% of researchers agreed that they did not have regular or wellsupported platforms to meet and exchange ideas with policymakers. Recent research from the London School of Hygiene and Tropical Medicine and Kings College London has also shown the importance of relationship building and direct interaction for a well-functioning Science-for-Policy ecosystem. The problem is particularly acute for newcomers and earlycareer researchers, who often lack the networks, institutional support, and visibility needed to engage effectively with policy circles. This policy brief presents EU policymakers with actionable recommendations for addressing the most acute issues of the EU Science-for-Policy ecosystem, as described above. The window of opportunity is open, as the advancement of the ecosystem has been confirmed as a priority action of the European Research Area, with specific goals including improving the crosscutting integration of knowledge in public policies and promoting collaboration among S4P actors. 2. POLICY PROPOSALS 2.1 Integrate cross-sectoral knowledge into a single interactive platform Policymakers should develop a dedicated platform for knowledge exchange, combining 6STG | Policy Papers Issue | 2025/21 | December 2025 responsive and challenge-led research and building on existing intiatives. Many institutions maintain separate platforms where policy briefs and longer reports are hard to find; moreover, non-commissioned, challenge-led research is often excluded, limiting external access. The Council of the EU curates a monthly publicly accessible ‘Think Tank Review,’ which collects relevant policy studies on EU issues. The most notable and developed joint platform currently available is the European Commission’s Knowledge4Policy (K4P), which collects research from 19 teams of European Commission scientists. Overall, however, the absence of an integrated crossinstitutional platform with ease of access reduces transparency, increases overlap, and prevents the creation of a common institutional evidence base. Such a platform should be integrated with the public consultation Calls for Evidence (CfE) on the European Commission’s ‘Have your say’ portal. The evidence and stakeholder opinions gathered there through template questionnaires are what is known as ‘reactive evidence,’ provided in the context of the Commission’s policy proposal at a relatively early stage. The platform is time-bound and offers limited interaction: submissions are provided during a fixed window and later summarised. Participants, however, seek ongoing engagement — for example, over 90% of contributors in Finland wish to remain involved, while only 33% of citizens in OECD countries believe that consultation input will actually be adopted. In a welcome development, the Knowledge4Policy platform mentions that an online community where scientists and policymakers collaborate to translate scientific knowledge into public policy is under construction There are already examples within the EU of more collaborative platforms for policy consultation. For example, in Estonia, a platform was developed in which researchers and other stakeholders co-work on legislative texts with policymakers. In Cyprus, the government launched an interactive platform in 2023 in which comments on legislative texts are made visible, as well as how they were addressed. Openness to challenge-led research is another way in which this platform can ensure that all relevant research, including research that can bring new issues to the agenda rather than being reactive, reaches policymakers. A good example of how a platform can be used to connect challenge-led research to policy needs is the United Kingdom’s ‘Areas of Research Interest (ARIs)’ platform. Policymakers are not only inviting comment, as in a public consultation, but are also proactively publishing questions that they want research to address. This allows the research community to identify where the evidence gaps are and align their work accordingly. For example, the ARI database shows questions such as “How can regulation encourage healthy and/or more sustainable food?”, “What digital technology can be used to improve the experiences of families going through the family courts?”, and “How can schools foster belonging and reduce absences?” The platform should serve as a long-term database of comparable evidence across sectors. It should include not only longer written materials and datasets, but also summary documents (such as policy briefs) that are more targeted to busy audiences. Artificial intelligence could be employed to make tagging, searching, and comparing documents more user-friendly and future-proof. Finally, openness to other types of evidence, including experiential evidence and citizen science, would further enhance the openness of the platform. As a recent LERU position paper argues, “In open, non-linear and networked systems, academic knowledge should be seen as a dynamic part of a wider process of knowledge production in which stakeholders bring in their own expertise, knowledge and insight. Societal impact is thus the outcome of the creative encounter of stakeholders and their contributions to a common goal” (Akker & Spaapen 2017). 2.2 Develop more meaningful ways of collaboration through research-policy fellowships Policymakers should offer more meeting and relationship-building opportunities to researchers as a way of bridging the two 7STG | Policy Papers Issue | 2025/21 | December 2025 communities. Research consistently recognises that the main factor promoting the use of evidence in policymaking is the establishment and maintenance of relationships between those who produce evidence and those who use it. Such relationships make both communities aware not only of present research needs and availabilities, but also of the objectives and limitations that each side faces. Beyond advisorships and working group appointments of researchers to policymaking bodies, more in-depth relationships are forged during funded fellowship programmes that allow actors to immerse themselves in the other community. Such fellowship opportunities exist for policymakers, including the Policy Leader Fellowship at the EUI’s Florence School of Transnational Governance. At the same time, policymakers should develop and fund a structured fellowship opportunity for researchers to join the EU’s policymaking institutions for a set amount of time, allowing them to engage with dedicated research projects and current policy needs. Such research-to-policy fellowships have already been developed in other contexts. In the United Kingdom, the UKRI Policy Fellowships are designed for earlyto mid-career academics and provide funding for an 18-month fellowship embedded within a UK or devolved government organisation. The programme supports collaborative research addressing pressing national and global challenges and seeks to foster strong, enduring partnerships between researchers and policymakers. Ireland launched a similar scheme called the ‘SFI Public Service Fellowship Programme,’ which enables researchers to undertake oneto twoyear secondments within public administration bodies. Still, the realisation of such fellowships has not been without challenges. In Ireland, there is a lack of interest among researchers to join the programme, despite a high demand from ministries. The results of the mentioned policy fellowships in the UK were evaluated by researchers during the pilot phase. Their analysis revealed several barriers to effective collaboration, including unclear expectations between fellows and host organisations, competing demands between academic and policy work, and difficulties navigating the cultural and procedural differences between research and government settings. Fellows often experienced tension between producing academically recognised outputs and delivering timely, policy-relevant evidence. Some also reported challenges of ‘administrative absorption,’ where they became lost in bureaucratic structures, and ‘academic absorption,’ where academic commitments pulled them away from policy engagement. These findings underscore that successful research–policy fellowships require clear role definitions, mutual understanding of goals, and sustained support on both sides to bridge institutional and cultural divides. 2.3 Embrace innovative ways of communicating research Policymakers should identify and incentivise best practices in communicating research in policy spaces. This comes with the recognition that communicating about public policy as such is challenging. Issues that evidence providers face when communicating research are similar to issues that governments face when communicating policy to stakeholders, as evidenced by the OECD Report on Public Communication. In the UK, for example, the UK Government Communication Service supports the communication aspect of policy with over 7,000 professional communicators, going as far as declaring communication as one of the four main levers of government (alongside legislation, regulation, and taxation). Policy briefs have become the standard of communicating research to policy; however, they often do not meet the needs of the policymaking audience. Our recent book presented five building blocks of effective policy briefs: context, writing, structure, visuals, and evidence (along with tips on all five). The European Research Executive Agency’s recent guide on policy briefs is also a step in the right direction, recognising the importance of standard setting. In order to strengthen the communication angle, policymakers should support specialised communication skills training and strengthen the role of knowledge brokers (actors who translate, contextualise, 8STG | Policy Papers Issue | 2025/21 | December 2025 and facilitate the flow of knowledge between researchers and policymakers). The European Commission has already recognised the importance of knowledge brokering as part of its ‘Science for Policy’ agenda and is exploring ways to professionalise and expand these roles within EU institutions, recognising that such intermediaries are essential for sustained, two-way exchange and mutual understanding between science and policy communities. Communication between researchers and policymakers can also be strengthened through dedicated science–policy labs, which create structured spaces for dialogue, experimentation, and co-production of knowledge. Science-Policy Innovation Labs (SPILs), for example, bring together scientists, policymakers, and stakeholders from across sectors to jointly assess policy processes and outcomes. They operate as interactive forums where participants review existing policies, explore alternative solutions, and evaluate the effectiveness of interventions from multiple perspectives. A similar approach can be seen in Finland’s Science Sparring model, where researchers and policymakers engage in structured, early-stage dialogues to critically examine draft policy proposals, identify blind spots, and refine knowledge needs. Both SPILs and Science Sparring demonstrate that regular, facilitated exchanges between research and policy communities can foster trust, enhance policy relevance, and lead to more adaptive, evidence-informed governance. Finally, communication of research to policy should also be properly incentivised. The current academic system puts emphasis on peer-reviewed journal articles as the primary route to academic advancement. Diverting time from such outputs toward policy briefs and other innovative channels for engaging with policymakers can therefore be perceived by academics as counter-productive from a career development perspective. Policymakers should address this by exploring ways of recognising and rewarding academic contributions to policymaking processes. International initiatives such as the San Francisco Declaration on Research Assessment (DORA) have already called for a broader set of evaluation criteria, urging institutions and funders to consider qualitative indicators of research impact, including influence on policy and practice. Similarly, the UK Research Excellence Framework (REF) has integrated impact as a key assessment dimension, accounting for 25% of the overall score in its evaluations universities and research units. These examples show that alternative evaluation systems can effectively incentivise engagement beyond scholarly publications. Building on such models, the European Commission and Member States could develop complementary recognition mechanisms to ensure that meaningful policy engagement and evidence translation are valued as core components of academic excellence. 9STG | Policy Papers Issue | 2025/21 | December 2025 Bojovic, J., & Bayley, I. (2025). Policy Communications: How to Write an Effective Policy Brief. London: John Harper Publishing. https://doi.org/10.5281/zenodo.17482531 European Commission, Directorate-General for Research & Innovation. (2024). Science advice to policymakers – Roles, enabling conditions and incentives: Mutual learning exercise on bridging the gap between science and policy: Second thematic report. Publications Office of the European Union. https://doi.org/10.2777/1130472 European Commission, Joint Research Centre. (2024). Science-for-policy ecosystems through the eyes of professionals: A multi-country survey analysis (EUR 40091 EN; JRC139213). Publications Office of the European Union. https://doi.org/10.2760/7490500 European Commission, Joint Research Centre. (2025). Evidence-informed policymaking: A pathway to increasing trust in democratic institutions and boosting competitiveness (JRC 141543). Publications Office of the European Union. https://doi.org/10.2760/3905455 European Commission, Joint Research Centre. (2025). Trust in science for policy nexus: Workshop report “Trust in Science for Policy”, 12–13 September 2024, Ispra, Italy (JRC 141424). Publications Office of the European Union. https://doi.org/10.2760/6212198 Oliver, K. (2022). Assessing national institutional capacity for evidence-informed policymaking: The role of a science-for-policy system (K. Krieger & L. Melchor (Eds.), Expert Report Series: Developing an Evaluation Framework for Science-for-Policy Ecosystems). Publications Office of the European Union. https://doi.org/10.2760/951556 Organisation for Economic Co-operation and Development. (2020). Building capacity for evidence-informed policy-making: Lessons from country experiences (OECD Public Governance Reviews). OECD Publishing. https://doi.org/10.1787/86331250-en Organisation for Economic Co-operation and Development. (2025). Better regulation practices across the European Union 2025. OECD Publishing. https://doi.org/10.1787/6f007516-en Pedersen, D. B. (2023). An evaluation framework for institutional capacity of science-forpolicy ecosystems in EU Member States (K. Krieger & L. Melchor Fernandez, Eds.). European Commission, Publications Office of the European Union. https://doi.org/10.2760/609597 Scientific Advice Mechanism. (2025). Building Bridges: Shaping Europe’s science-for-policy landscape – Conference report (26–27 May 2025, Vienna). https://scientificadvice.eu/conference/ report/ REFERENCES