This chapter provides an in-depth overview of the findings from the EIPM project. It is structured around the original project’s analytical framework which underpinned the country analysis: it is split into individual, organisational, and inter-organisational/systems levels. In terms of individual capacity, it addresses enhancing collaboration at the interface of science and policy, increasing capacity for policymakers to work with evidence and for knowledge brokers to carry out their roles effectively, and increasing the incentives to engage in EIPM. In terms of organizational capacity, it includes strengthening institutional structures and processes for EIPM, improving access to data for policy-relevant purposes, and improving intra-organisational communication. At the inter-organisational and systems level, it includes developing EIPM-relevant tools such as impact assessments, evaluation, and strategic foresight, strengthening co-ordination of EIPM across government, and promoting an EIPM-conducive culture. All sections outline the relevance of the issue, how it relates to the participating countries, and measures the countries are taking or have taken to overcome them.
Strengthening National Evidence‑Informed Policymaking Ecosystems
Abstract
This chapter brings together the findings of the national country reports and is therefore structured following the analytical framework underpinning these country reports (see Box 2.1 in previous section) (Koppel et al., 2025[1]) (Dorren et al., 2024[2]) (Ladi et al., 2025[3]) (Nakrošis et al., 2025[4]) (Petr et al., 2025[5]) (OECD, 2024[6]) (OECD, 2025[7]). However, the discussion of the findings may at times go beyond these analytical categories as a result of the broader findings of the project. It is important to note that the project focused on the needs and gaps in the EIPM ecosystem of the countries, not on identifying good practices. As such, some of the examples focus more on the missing elements than on the already established structures.
The analysis conducted in all participating countries demonstrated a great diversity in approaches to EIPM, reflecting various research infrastructures, government setups and cultures. This variation indicates a need for care when analysing EIPM ecosystems. Approaches, best practices and recommendations should be informed by international experience and practices while remaining grounded in local context. On the other hand, several challenges and needs emerged during the project that are shared among the seven countries. The analysis that follows draws on these common challenges to offer an all-encompassing perspective, while being attentive to the particularities present in each country.
At an individual level, three main types of actors exist within country evidence ecosystems. These are: researchers, who produce original scientific results, knowledge brokers (see Box 2.2), who translate scientific input into policy relevant outputs and produce analytical evidence within government, and policymakers, who use these outputs to support decision making, often ministers and their policy advisers, or top levels within the civil service. This chapter focuses on challenges met by these actors and discusses options for building their capacity.
The lines between these three roles are often blurred. For example, researchers can also be located in government units and departments, which have a policymaking function, research institutes may have policy impact units which transform research into policy, and knowledge brokers can reside on both sides of the spectrum. Taking a broader perspective, there is also a need to listen to citizens, as in democracies any decision made on public affairs should be informed by citizens’ perspectives, as a result of public consultations. The analysis in this chapter will not address the role of citizens per se, as this falls outside the scope of the project. However, it will refer to the role of citizen engagement in relation to these three actors, as many of the country reports and analysis refer to consultation and engagement practices.
Facilitating co-operation between policymakers and researchers
Enhancing collaboration
Several EIPM-related competencies are already shared by policymakers and researchers. As indicated by the Competence Frameworks of the JRC on Science for Policy (for scientists) and Innovative Policy Making (for policymakers) presented in Figure 3.1 and Figure 3.2 below, the ‘Engage with Citizens and Stakeholders’,1 ‘Collaborate’ and ‘Communicate’ clusters are common to both scientists and policymakers. These competencies have the following characteristics:
The ‘Engage with Citizens and Stakeholders’ competency cluster requires understanding how to plan, design and carry out stakeholder both consultation processes and citizen engagement opportunities, and advocates for such engagement.
The ‘Collaborate’ competency cluster involves recognising the value of sharing and knowledge exchange, helps strengthen collaborative practices through instilling trust and developing collaborative strategies, and taking an empathetic and open-minded approach to collaboration.
The ‘Communicate’ competency cluster involves recognising the value of communication strategies, being competent in clear writing and storytelling, and being able to combat mis- and disinformation effectively.
These three clusters illustrate a collaborative mentality, an engaging attitude and the capacity to engage with a variety of perspectives and opinions.
Source: (JRC, 2023[8]))
Source: (JRC, 2023[8]))
However, bridging the work of researchers and policymakers and identifying avenues for collaboration comes with many challenges, including different working cultures and objectives, as well as different timelines. In several of the countries, including Estonia, the Netherlands, Latvia, and Belgium, researchers report difficulty understanding the policymaking process and feeling that their research was not adequately understood by policymakers. Regarding the former, these researchers often do not understand when their advice is needed by policymakers, what format it should be in, and how they should go about promoting its value. This is particularly the case for tendering and procurement procedures, where many research institutions feel they are at a disadvantage when competing with larger consulting firms, both due to excessively short timeframes and lack of understanding of the tendering process, even if they had greater knowledge of the topic at hand. Regarding the latter, researchers often worry about policymakers seeing things in excessively black-and-white terms and thus missing important nuances. They can also feel frustrated when their advice is not considered, feeling that the reasoning behind such decisions is not always clear.
Collaboration may also be hindered by operational obstacles such as misalignment in timelines between researchers and policymakers, and dissimilar focuses of work, as well as the absence of concrete contact points and lack of meeting opportunities. A recent survey published by the JRC involving all three actors of the EIPM spectrum shows that the most frequently identified obstacle (71.3%) referred to a lack of meeting opportunities (Scharfbillig et al., 2024[9]).
The first step in improving collaboration is to increase awareness and understanding of EIPM as a concept. There are often significant gaps here: a recent study on science for policy expertise revealed low levels of knowledge both in scientists and policymakers, with 73% of respondents saying that researchers don’t understand how policy works and 80% stating that policymakers lack sufficient understanding of science (Pearson, 2024[10]).
Sustainable collaboration between researchers and policymakers can also be built by using the shared perspective on the value that EIPM brings to society, while ensuring mutual respect for the different roles. Discussions with researchers at the capacity building workshops organised as part of this project have indicated the need for a convincing argument for science for policy focused not only on the benefits for policymaking but also on the benefits for society. This notion is supported by research by the European Commission's Scientific Advice Mechanism (SAPEA) , which found young researchers to be more open to this more co-operative approach, considering the contribution of science to societies as their responsibility (SAPEA, 2024[11]).
A collaborative evidence-policy mindset can also be cultivated by research institutions which encourage a policy relevant approach. Several such institutions exist in the United Kingdom, including the Universities Policy Engagement Network and the European Federation of Academies of Science and Humanities, both of which can serve as good practices.
The significance of a communication mindset appears as a common missing element in researchers’ skillsets. For example, the Greek report refers to the importance of communicating research outputs for Science for Policy activities, while the Lithuanian report notes the lack of efforts to train scientists on Science4Policy. The findings of these reports were also supported by feedback from the capacity building workshops with researchers, where exercises involving communication were often carried out with difficulty. Another common issue is familiarisation with the policymaking process. This was evident, for example, in the Estonian report, which underlines the gaps in understanding of the policymaking process and limited communication skills of scientists. While training could help alleviate some of these issues, training programmes focused on these skillsets were rare.
Communication issues can also be identified from the policymaker’s perspective – frequently, the skills to communicate evidence needs and ask relevant questions to researchers are usually not in place. Ad hoc and personal interactions still dominate science advice processes (i.e. in Latvia), often leading to limited engagement with a narrow range of individuals.
Engagement with stakeholders and citizens is another competency that is required for both actors when they engage in EIPM. EIPM is inherently affected by multiple interests and values, and understanding the plurality of available perspectives helps ensure the EIPM process remains socially relevant. A comprehensive study of a problem must consider the departure point of each stakeholder and distinguish between facts, opinions and interests.
Increasing capacity for working at the interface of science and policy
Even experienced researchers may lack the relevant skills that allow them to be efficient and productive in their interaction with policymakers. A key issue here is the lack of familiarisation researchers often have with the policymaking process and how evidence is used throughout the policy cycle. This can be mitigated through developing understanding of EIPM processes and how evidence can impact policymaking, which can both incentivise researchers to contribute to these processes and aid them in understanding the input needed at each stage.
One way to develop this understanding is through training programmes. In Greece, the Netherlands and Belgium, supply-oriented organisations often have high-quality personnel able to effectively collect, process, synthesise and even interpret data, but do not always have the soft skills necessary to communicate their findings and translate them into policy – for example, the ability to make research output accessible to policymakers via relevant outputs (e.g. policy briefs). As noted for example in the Greek report, supply side organisations need to expand the skills of their staff so that they can produce output that leads to comprehensible and actionable policy proposals. Similarly, in Lithuania the existing Competence Framework for Researchers only implicitly mentions science for policy skills. It is equally important to encourage researchers to engage with experts from other disciplines. This can foster greater inter-topic understanding and build synergies, which can help fill expertise gaps. This was seen in the Czech Republic, where greater collaboration held the potential to increase understanding on the relationship between the different disciplines.
Several country reports also propose pairing schemes and secondments as a way for applying and informing the theoretical knowledge acquired in EIPM-relevant training programmes. These could involve both policymakers spending time in research institutions, and researchers spending time in government. In the latter case, these programmes should be designed in such a way that allows researchers to continue their academic research, and thus reduce the extent to which they pose a risk to career progression. Such an increase in time spent in each other’s camps is likely to significantly increase understanding of their respective working styles and processes, thus reducing the sense that they are speaking completely different languages. The project offered to the seven countries a series of examples of pairing schemes drawn from other European and broader jurisdictions. In Europe, the visiting researchers’ scheme in the JRC is an example of such a practice, and has enabled many researchers to acquire a better knowledge on policymaking procedures.
Less formal ways of engaging with policymaking can also facilitate increase of knowledge. Communities of practice like the JRC’s Training of Trainer community can foster collaboration, exchange of ideas and discussion on potential hurdles within EIPM.
Another option is a policy lab, which generally functions as a space to bring together a wide range of participants in order to connect on research and policy. The most prominent example is the EU Policy Lab which functions as a collaborative space within the JRC, enabling researchers to experiment in their approach to EIPM. Another example is Greece’s Science4Policy Lab, discussed in Box 3.1. (European Commission, n.d.[12]).
The Policy Lab at the National Documentation facilitates communication and collaboration by bringing together policymakers, scientists and stakeholders, and therefore engages in co-creation of policies The lab aims to:
Understand the nature of problems more deeply and contribute to their resolution.
Analyse challenging issues, approaching them from a variety of perspectives.
Create bridges with actors in civil society, the economy, politics and evidence.
Increasing capacity to work with evidence (policymakers)
Skills to supply reliable, policy-relevant evidence are essential if policymakers are to use such evidence effectively. However, these skills are not always present in government, leaving them deprived of high-quality expertise. Even when the will and capacity to supply evidence is present, skills, frameworks and mechanisms are needed to request, access and understand this evidence, as well as to translate it into policy decisions. It is vital for policymakers to understand how evidence is formed, and to recognise that supply of evidence does not necessarily mean that this evidence is used effectively.
To provide high-quality evidence, governments need to ensure that civil servants and policy analysts in the public sector are equipped with EIPM-relevant skills. The specific skills required depend on the nature and accessibility of available evidence, but can include the ability to identify evidence needs, to connect to experts, and to gather and assess evidence. The JRC Competence Framework found in Figure 3.2 provides a comprehensive overview of these skills.
There are cases where some EIPM-related training is conducted, but it is often ad hoc and scattered. This is the case in the Czech Republic, where support for any development is ministry-specific, with limited cross-institutional training, and in the Netherlands, where some researchers receive training on policy impact, but the kind of training received varies widely even between individuals. In Belgium, the Federal Public Service for Policy and Support (BOSA) conducts a number of trainings, although these do not examine EIPM issues. In Greece, the National School of Public Administration’s existing training courses on policy analysis have been complemented with a series of trainings on EIPM.
In cases where supply of evidence is considered in policymaking organisations, it does not necessarily follow that organisations will consider how to best utilise it. This is the case in Belgium, where while several government organisations and units within organisations produce high-quality evidence, policy analysts rarely consider how to best make use of it. Similarly, in the Czech Republic, discussions around evidence that do occur regularly focus on generating it, rather than how to make effective use of it. Furthermore, policymakers’ lack of knowledge on research means they often don’t understand the processes or timeframes needed for good research and can thus have unrealistic expectations on output speed.
Certain training programmes would benefit from being organised at a more centralised level, to encourage systems thinking to a greater extent. A strong example here is the Czech Republic’s Strateduka, a hybrid training course aimed at developing competencies in strategic planning and management (Box 3.2). While not all the skills it teaches are directly relevant to EIPM, its holistic perspective is valuable in ensuring trainees learn how to work with processes and people even outside of their professional scopes. The ability for civil servants to be able to ask researchers the right questions is also critical. This involves developing an understanding on how policy questions are formed, an awareness of knowledge gaps and a capacity to frame questions in a way that does not bias answers.
Strateduka is a course in the Czech government focused on the development of competencies in the field of strategic planning and management. The course is provided in a hybrid format and lasts three days. The programme has three courses per year, all of which were fully filled in 2023.
The programme enables participants to deepen their understanding of strategic work from its conception, i.e. identifying the need to prepare a strategic document, to its evaluation. The course provides a set of tools useful for strategic work across the public sector and highlights the value of looking at strategic issues from a holistic perspective, including by identifying the impact of global trends on such issues, as well as how different actors are involved in the preparation and implementation of strategic documents.
As part of this project, a dedicated training session on ‘working with evidence’ was organised as part of the capacity building workshop for policymakers. Most of the participating policymakers underlined the need to more closely engage with the tools provided at an EU level, deepen their knowledge and skills regarding the use of evidence in policymaking. They further reported that theoretical knowledge should be clearly connected with practice and that using specific policy examples enhances learning. Finally, they highlighted the value of the interactive sessions within the training, and in particular the peer learning opportunities among countries.
Increasing data literacy
Due to the wealth of knowledge produced and collected in policymaking processes, data literacy is of critical importance. This is evident in Belgium, where the findings of the project show that while it is in theory possible for academics or civil servants to process at least some data sources themselves, this requires a level of expertise regarding data production that is low across most organisations. This often leads to researchers requesting data that isn’t needed and thus slowing down data approval processes. Such inefficiencies also hinder the timely production of evidence, which in turn impairs the ability of decision makers to inform their decisions on such evidence. As a result, planning becomes reactive rather than proactive, and policy measures may be based on incomplete or misinterpreted data. This lack of data literacy can also impact the supply side, as many civil servants are not able to update data platforms within their own organisations, and points of contact with the required technical knowledge about these databases are sparse.
Such a lack of knowledge can also mean that policymakers do not always understand the importance of data quality. In theory a quick fix for this issue could be to hire more statisticians and data experts who already have these skills. However, while such roles could certainly help, particularly in reducing data access bottlenecks, it is also important for those making policy decisions themselves to have some of these skills, to allow for a more holistic consideration of the data. One such approach can be seen in Lithuania (Box 3.3), where the government created a set of trainings around data literacy deliberately designed to reflect changes to its data collection policies.
In January 2023, an amendment to legislation was implemented that requires Lithuanian ministries to justify with data why proposed measures should be applied when submitting draft legislation to government. Furthermore, it turned Statistics Lithuania into the State Data Agency, and made it responsible for the collection and use of high-quality data for policy decisions (it was already responsible for their production). This means that all state institutions are now able to carry out analytical tasks in a secure environment without having to invest in installing additional analytical tools in their own systems.
In order to ensure ministries could make the most of this new data source, the State Data Agency incentivises competency trainings in ministries. These training programmes focus on giving participants analytical skills, including using statistical packages such as SPSS and programming languages such as SQL.
Increase capacity of individuals working as knowledge brokers
Researchers and policymakers can be assisted in their interaction and co-operation by knowledge brokers, individuals and organisations able to bridge the gap between the scientific and policymaking worlds. Other terms that are frequently used to describe such a role are intermediaries, boundary organisations, and knowledge translators.
Knowledge broker positions should be filled by people well-versed in both academic and policy-related ways of working. It can be helpful if they are relatively senior, so that they can act as clear points of reference for other employees. Furthermore, it is important to ensure that they have sufficient time to dedicate to this task of translating evidence into policy – for example, while Estonia has several science advisers who facilitate exchanges between academia and policymakers, their heavy administrative workload means they are often not able to fulfil their role as effectively as they could. In Greece, the work of the Unit of Experts Employment, Social Insurance, Welfare and Social Affairs MEKY and the Council of Economic Experts has been well acknowledged in policymaking and the need to deploy similar units in other ministries consists one of the major recommendations in the Greek report. In the Netherlands and Belgium, where the concept of knowledge brokering is more developed, the main challenge is co-ordination, as fragmentation appears to hinder effective utilisation of evidence.
Some issues of fragmentation can be attributed to the lack of widespread consensus on the definition of a knowledge broker and where they should be positioned within a knowledge ecosystem. Definitions in the literature on the topic highlight a variety of possible roles, including putting complex scientific data into accessible language, creating networks between researchers and policymakers, helping involve policymakers in the research process, maintaining the independence of scientific advice, and adapting research to political contexts. The style and location of work can also vary – some tasks can be carried out on an individual basis, while others can occur within organisations. Knowledge brokers can also work directly for intermediary organisations, or with more traditional knowledge supply and demand organisations such as universities, ministries or independent bodies. (Van Eerd and Saunders, 2017[15]) (European Commission, 2019[16]) (Torres and Steponavičius, 2022[17]) (Stier and Smit, 2021[18]) (Tellmann and Gulbrandsen, 2022[19]) (Pedersen, 2023[20]) (Gluckman, Bardsley and Kaiser, 2021[21]) (Davies et al., 2017[22]). There would thus likely be value in clarifying of the concept of knowledge brokerage (Box 2.2), and in some countries, ensuring clearer recognition of the role. This project has contributed to identifying this issue, which will be further explored in the future (see Chapter 4).
Several countries have created knowledge broker positions within universities and ministries. In Estonia, over ten science advisers are currently tasked with advising ministers on matters related to R&D, including fostering co-operation on research, developing research plans and implementing them in co-operation with different stakeholders, as well as representing Estonia in international initiatives for R&D co-operation. Lithuania’s Research and Innovation Adviser (R&IA) network, although relatively new, envisions a similar approach, with plans for 15 advisers strategically placed in Lithuanian ministries, guided by criteria emphasising expertise in governance and networks between science and policy. Both countries expressed their ambition to strengthen the position of these networks and work on stronger positioning of these knowledge brokers within their ministries.
The project identified a need to ensure individuals that perform these roles are not isolated, in particular in view of the often-scattered knowledge on the profile and positioning of knowledge brokers. Training programmes for knowledge brokers can once again be of value here, in order to foster understanding of the particular characteristics of such intermediary roles. For example, knowledge brokers should be able to understand the perspectives of both researchers and policymakers, negotiate and resolve potential conflicts. In order to encourage this, training programmes could help ensure that the following skills are present:
a broad understanding of policymaking and scientific procedures.
capacity to negotiate and conflict resolution skills.
scientific experience to assess the validity, appropriateness of evidence.
time-efficiency to be able to respond effectively to deadlines and urgent needs in policymaking.
collaboration skills and capacity to manage teams.
effective communication skills.
There may also be value in fostering the development of managerial and leadership skills, in particular for senior scientific advisors and chief scientists whose role is to steer processes. Further professional development could also be supported through the creation of knowledge broker networks. Such networks could give individuals the possibility to exchange best practices and reflect on their respective tasks, and thus strengthen their collective capacity. A network of science advisors could also help to bridge ministerial siloes. Additionally, the use of secondment programmes discussed in the previous sections could also provide value here.
Career incentives to engage in EIPM work
A working environment conducive to EIPM within universities and research institutions
When EIPM activities are considered irrelevant to professional development in research careers, there is naturally a lack of interest and motivation to engage in such activities. In many of the countries that formed a part of this project, such a disengaging environment exists, in which EIPM-relevant skills acquired by researchers are rarely taken into consideration in their career development. This can mean that, in cases where EIPM-relevant skills are already present to some extent, they may not be recognised properly and thus do not offer any advantages to the researchers that possess them.
More significantly, incentives for researchers to partake in policy work are often lacking. In the Netherlands, an overall focus on international academic excellence appears to predominate, to the extent that stakeholders sometimes feel that the academic system prioritises international recognition over domestic issues, with many academic courses conducted in English. In the Baltic states, heavy reliance on European funding means that relevance to policy can also be built through the requirements that exist in the Horizon programmes, as mentioned above. Thus, researchers’ focus on such policy work can have minimal impact on, and even be detrimental to their career. This means that many researchers who might otherwise be motivated to work on policy-relevant research and dedicate time to refining their policy-relevant skills are likely to avoid it or consider it a low priority.
In order to reduce such barriers between researchers and policymakers, it is important for universities to recognise the benefits of policy-related work and provide development paths for academics that are interested in pursuing such a career path. This appears to be the case for some universities in Flanders in Belgium, with a particularly strong example at Ghent University, where the criteria used to plan and evaluate research consider impact on policy (Box 3.4).
Ghent University’s Research Dimension Portfolio was created to aid in the description, planning, and evaluation of research. The portfolio includes a variety of dimensions, including leadership, ability to work in an interdisciplinary way, scientific impact and socioeconomic impact. Professors are able to select competencies within this portfolio that best match their profiles. These dimensions can be used for context in deciding on the appointment and promotion of university members of staff. The portfolio splits into two main sections: 1) design and development of research; and 2) impact of research. This highlighting of the importance of impact, including impact in policy, is a valuable incentive for researchers to dedicate greater amounts of time to policy-related work.
Source: (Ghent University, n.d.[23])
At a personal/professional level a rewarding evaluation system that identifies value in EIPM skills can trigger the attention of researchers that are looking for alternative career paths. The Coalition for Advancing Research Assessment Framework is a good example of an assessment framework that considers the impact of research (see Box 3.5).
The Coalition for Advancing Research Assessment (CoARA) has developed an Agreement on Reforming Research Assessment that sets a shared direction for changes in assessment practices for research, researchers and research performing organisations, with the overarching goal of maximising the quality and impact of research. The vision is that the assessment of research, researchers and research organisations recognises the diverse outputs, practices and activities that maximise the quality and impact of research. This offers an opportunity to promote changes in research assessment for both researchers and research performing organisations, where engagement in EIPM activities and production of policy outputs (policy briefs, policy reports, etc.) can be used as a quality criterion.
Source: (CoARA, 2023[24])
However, policy impact cannot be a compulsory and solitary assessment criterion for researchers or research organisations. Not all research has to be policy relevant and creating pressure for this to be the case may disincentivise researchers from outside-the-box thinking. Furthermore, it is not always feasible to assess the impact of research at a particular time. Sometimes evidence can provide solutions before problems appear and the potential policy impact can only be traced in the long term (Zika, 2023[25]). It is thus important to strike a balance, ensuring that researchers are supported in efforts to contribute to policy without forcing them to do so.
In order for those with EIPM skillsets to identify jobs which take advantage of these skillsets, job profiles and openings should explicitly recognise these skills. When the acquisition of these skills is not related to professional development or career development, any EIPM activity is likely to remain a sporadic and voluntary process at the margins of academic activity.
Recognition of EIPM-related skillsets within government
As within universities and research institutions, skills to demand and effectively utilise evidence are often not adequately valued within government. This leads to situations where even if EIPM-relevant skills already exist, they are frequently unidentified or underexploited. This is evident in Greece, where despite an increase in formal qualifications such as Masters’ and PhDs, such qualifications are not well integrated into work appraisals in public administration.
All seven countries forming part of this project reported either that ability to use evidence was not considered in performance evaluations, or that analytical positions were considered inferior to managerial positions from a promotional perspective, thus disincentivising ambitious employees from pursuing them. It is thus important to put such incentive structures in place. Another strong initiative that works towards mitigating these issues can be seen in the recent creation of the ‘executive officer’ position in Greece, designed to fulfil a variety of staff functions in central government. The policy analysis branch of this position has been trained in a variety of EIPM-relevant skills, including using statistical data and reviewing secondary literature in view of making evidence-informed policy proposals. Furthermore, they are mandated to operate in this manner in their day-to-day work. In Latvia, some training programmes do incorporate EIPM-related skills, but do not explicitly recognise their value to EIPM and thus do not encourage employment of these skills to this end.
Inability to identify existing skills also means that training programmes to improve these skills become harder to create and implement. This is the case, for example, in the Czech Republic, which recently created a three-day training course looking to develop a range of EIPM-relevant skills. While the course was a welcome initiative, the unclear understanding of exactly what skills civil servants need for EIPM meant that it struggled to respond to the broad range of participants involved. Other countries are likely to face similar issues – without an understanding of which EIPM-relevant skills are important within a particular ministry or organisation, as well as which skills are already present and to what extent, any training programme is unlikely to be tailored enough to offer more than generic advice.
To remedy these issues, EIPM-relevant skills need to be formally recognised and implemented into competency frameworks such as those developed by the JRC (see Figure 3.1 and Figure 3.2). Good examples of special technical tracks, such as policy analysis tracks or digital professional tracks can be found in some countries beyond the seven countries involved in the EIPM project (see Box A.B.2 in Annex A), including the Irish Government Economic Evaluation Service or France’s “Statistician economists”. Another example is the United Kingdom’s efforts to map digital skills across government.
Once these skills are identified, rewarding civil servants could incentivise the use of such skills in their careers. As it currently stands, understanding how to make information relevant for policymaking purposes rarely forms part of performance evaluation criteria. For some countries the recently developed skills framework for public administration can be enriched with more explicit EIPM skills. For example, in Greece the recent law establishing a Common Framework of Competences is a great step towards the identification and uptake of EIPM-related competences. Similarly, in Lithuania the existing civil service competence model since 2014 encompasses competencies related to information management but fails to directly connect with EIPM.
As mentioned above, another issue is that analytical employment paths, if they exist at all, tend to offer less scope for career progression than managerial paths, acting as a further disincentive to focus on analytical skills. Ensuring that such skills do make up a portion of evaluation criteria, and that they can be rewarded to the same extent as managerial prowess, is likely to serve as a strong incentive for civil servants to focus more on them. Ideally, ensuring that such skills are adequately rewarded could also take the form of different available career paths – i.e. some managerial and others analytical – so that those who are more analytically oriented are incentivised to take this path. Identification of skills in this manner will also serve to inform the development of training courses, allowing them to be tailored to the knowledge needs and gaps that exist.
In order for such skills to become more commonplace, hiring practices need to successfully identify and attract those who already possess such skills. Across the seven countries examined, challenges in hiring for EIPM-relevant analytical roles are evident. In Belgium, for example, the capacity to translate research into policy is not considered in hiring processes. In Latvia, Greece and Lithuania, a similar situation exists, which, combined with relatively low salaries as compared to the private sector, can make it difficult to attract workers with the relevant skills. Latvia recently aimed to improve competitiveness by aligning public salaries to 80% of private sector levels, although budget constraints hinder full implementation. Among other countries, in Greece, a lack of a clear job descriptions for specialised personnel also means that it is difficult for them to become well embedded in government.
In some cases, these hiring issues have been recognised, and steps have been taken to remedy them. For example, several ministries in the Czech Republic have recently made efforts to establish analytical units. However, limited pay scales and a tight labour market mean positions within these units can be difficult to fill. In Belgium, while there are adequate salaries for analytical positions within certain federal organisations, ministries still face difficulties recruiting specialised profiles such as data scientists due to strong private sector competition.
As mentioned previously, the establishment of relevant job profiles that speak to the use of evidence in policymaking can also be an incentive in attracting talented researchers to public administration, or allowing well-educated civil servants to receive credit for their skills. Such hiring procedures and job profiles go hand in hand with the relevant analytical units that need to be part of all ministries and have a mandate to work with evidence.
Strengthening institutional structures and processes that feed evidence into policymaking process
Developing supporting systems for knowledge brokerage
While building individual capacity ensures that actors on both the supply and demand sides can effectively take on an EIPM approach, there is also the need to tackle EIPM at the organisational level, as discussed in the overall analytical network. Effective EIPM cannot be carried out in an isolated manner and requires a matrix of relevant structures to produce results.
On the supply side, i.e. in academic and research institutes, there are rarely systematic structures for knowledge brokerage able to connect research results to policymaking. However, there are some institutional and planning structures in place. For example, the strategic plans and work programmes of universities, which usually cover a 5-year time horizon, are structured using an approach that reflects both the growth and development of the university and social needs. In addition, research funding programmes, such as Horizon 2020 and Horizon Europe, tend to include impact as one of their key criteria when evaluating proposals.
Usually, universities’ international liaison and communication units undertake the role of communicating external research outputs stemming from various university departments, not necessarily with a focus on policy impact. The same patterns often exist in research organisations, with the exception of those that are involved in a knowledge brokerage role. For an EIPM perspective to flourish, dedicated structures or teams within research organisations must co-ordinate and disseminate research for informing policies.
Setting up knowledge brokerage capacity can also be done within the public sector, either within some ministries or in structures operating at arms’ length from the ministries. Several governments have effective examples of knowledge brokerage organisations, although they are often limited to specific sectors. Countries such as the Netherlands make use of specific organisations dedicated to providing evidence for policy, such as advisory councils. These organisations function as ‘translators’ of evidence-to-policymakers and can help ministries in articulating knowledge questions and translating them into research questions. In Belgium, the knowledge brokers KCE and Sciensano function highly effectively as intermediary organisations between scientists and policymakers in the public health sector (see Box 3.6). However, outside of this sector such organisations are rare. A similar issue can be seen in Estonia, where the varying requirements for the recruitment of science advisers across different ministries means that their capacities for knowledge brokerage are likely to vary. It therefore may benefit these ministries for some hiring requirements to be consistent across government.
The lack of demand for the evidence produced by knowledge brokers can also be an issue. As previously mentioned, the Belgian knowledge brokers within the public health sector are very effective. While KCE recommendations are usually followed with clear effect by decisions from the minister of health, the Superior Health Council, another organisation within Belgium’s public health sector, estimates that as few as 25% of its scientific recommendations are used, perhaps due to their cross-cutting nature that goes beyond the remit of the Ministry of Health. A similar issue can be seen in Greece, where, with the exception of a few fields such as the economy, labour, and social affairs, ministries do not generally work with officially established knowledge brokers in their fields.
Assistance on use of evidence can also be provided through dedicated analytical units, as has started to be formed in the Czech Republic. This unit, known as the Government Analytical Unit (VAÚ), partners with line ministries throughout the regulatory impact assessment cycle. In this, it provides guidance, helping ministries design RIA plans and select appropriate analytical methods, as well as providing analytical services, such as preparing evidence maps and literature reviews.
However, in many of the seven countries, including Latvia and Belgium, units dedicated to collecting and encouraging use of evidence are rare. However, several steps are being made to increase the role of knowledge brokers. For example, in Lithuania, the launch of the network of Research and Innovation Advisers by the Research Council of Lithuania is an important initiative, and represents an opportunity to strengthen the council’s advisory and knowledge-brokerage role. Lithuania’s STRATA also provides a strong example of the knowledge brokerage role within government. Its mandate to foster a culture of EIPM across governmental institutions, as well as its focus on aligning evidence with government priorities, makes it an effective way to leverage external expertise during the preparation of studies and evaluations. Greece has also made several notable steps forward – knowledge brokers within government, including the Council of Economic Experts (SOE) and the Unit of Experts in Employment, Social Insurance, Welfare and Social Affairs (MEKY) have obtained a central role in their respective policy fields – although they are only present in specific sectors. This is an important area where exchange of experience at the international level, such as through the EIPM project, particularly the needs and gaps assessment which provided many international insights, as well as the capacity building activities can help countries identify and implement innovative approaches.
KCE
The Belgian Healthcare Knowledge Centre (KCE) is an independent, but federally funded research institute that provides scientific advice to authorities on topics related to health care. While the KCE's recommendations are not binding, their advice to medical practitioners should be followed, and failure to do so could be used in the court of law to attempt to appeal malpractice. By refraining from involvement in subsequent policy choices, it ensures its impartiality as a knowledge broker. However, a report by an independent consultancy found that over 50% of KCE recommendations had political impact. A large part of the research work conducted by KCE consists of knowledge synthesis – i.e. synthesizing existing evidence about a specific topic from different sources to gain new insights and draw comprehensive conclusions. Besides this, KCE also frequently performs extensive data analyses and primary qualitative research to ensure relevance of the work for Belgium. All research conducted by the KCE is made accessible to the public.
Sciensano
Sciensano contributes significantly to the scientific community, publishing around 280 peer-reviewed articles annually, and addressed over 100 parliamentary questions in 2021, playing a noteworthy role in supporting evidence-based policies and initiatives in Belgium. Furthermore, it offers training to scientists to increase the impact of their research, by addressing four dimensions: quality, targeted communication, stakeholder engagement and cross integration. Part of this effort has resulted in the production of several health data visualization tools, following the Findability, Accessibility Interoperability and Reusability (FAIR) health data principles, to facilitate public health information dissemination to the public.
In 2020, Sciensano launched a 'Research Translation' project with the aim of improving the utilisation of their research results by decision makers, an objective written into the institute's 2023-2029 strategic management plan. Using a list of 30 criteria, they investigated the quality, dissemination, and utilisation of service outputs by stakeholders, including civil society and actors outside the health field.
Source: (OECD, 2024[26])
Removing bureaucratic barriers to getting external evidence-informed advice
Next to the presence and effectiveness of knowledge brokering structures, policymakers’ ability to address their evidence needs is influenced by the setup of procurement and budgeting processes to rely upon evidence produced externally either through academia or consulting firms. In both Estonia and Lithuania, for example, policies on public procurement were mentioned as key barriers to increased use of external evidence-informed advice by policymakers since they involved lengthy processes and were not designed for procuring research. Moreover, while tying ministerial research funds to the annual government budget cycle can be a useful way to ensure that there is dedicated budget for such projects, if there is not enough flexibility as to when such funding must be used, it can discourage ministries from engaging in multi-annual research projects. This is partially the case in Estonia and The Netherlands, which reduces policymakers’ flexibility in procuring policy-relevant evidence and discourages ministries from engaging in strategic, multi-annual research projects – thereby, substantially limiting the kinds of evidence they can procure. In Estonia, the short timeframes available for completing research projects procured through ministerial funds – a consequence of the timing of the annual budgetary process – was even mentioned as a reason for the often-low quality of publicly procured research.
Different solutions to the challenge of flexibility exist. In Sweden and Estonia, for instance, the public procurement law foresees exemptions for public R&D projects – although the actual use of such exemptions requires specific legal support, as the case of Estonia shows (Swedish Competition Authority, 2016[27]). Other solutions include mixed funding models, as in the Netherlands, where certain ministries use both tenders and dedicated arrangements with public sector research organisations, or the United Kingdom’s dynamic purchasing system for research and insights (TNO, 2025[28]) (Crown Commercial Service, 2021[29]). The latter is a dedicated pool of knowledge providers run by the crown commercial service that allows policymakers to run streamlined, R&D-specific tender processes. To encourage multi-annual, policy-relevant research, some countries, including Belgium and Spain, are setting up dedicated science for policy research funding programmes (OECD, 2025[7]).
Creating visibility and voice for the EIPM function – the role of leadership
Even when analytical skills are well-developed, leadership remains indispensable in connecting the various elements of the EIPM process. Without strong leaders to act as co-ordinators and points of contact, responsible for promoting and ensuring the use of evidence throughout the policy cycle, interest in evidence at the policy level often remains fragmented and disorganised. Therefore, this needs to be addressed at the organisational level.
Leadership at the administrative-political interface is particularly crucial. Here, individuals need to be able to navigate the distinct priorities of administrative staff and political decision makers, ensuring that evidence aligns with policy objectives while preserving its integrity and relevance. The need for leaders with these capabilities is particularly significant in contexts where political actors view evidence as an afterthought or as a tool for justifying pre-existing decisions. In many countries, the absence of clearly defined leaders or ‘champions’ at this interface undermines the integration of evidence into decision making. In Estonia, while there is willingness to listen to scientific advice before making decisions, differences between ministries’ cultures on how to engage with this advice means that knowledge on such engagement is unevenly spread, making it difficult to engage with long-term problems. In Lithuania, those demanding evidence tend to prioritise advice from experienced professionals rather than relying on scientific evidence and are more likely to consider scientific advice when it comes from scientists with practical experience. This suggests that Lithuania would likely benefit from evidence ‘champions’ able to effectively highlight the value of research findings, even when they don’t come from those with practical experience.
In some countries, staff in leadership positions who are able to effectively work at the administrative-political interface are also likely to be employed as ministerial advisers on the political side. In Belgium, for example, while technically skilled senior staff do exist at the administrative level, there are few individuals who can effectively translate evidence in politically meaningful ways, with most staff able to operate effectively at this interface are working within the cabinets. In Greece, despite increasing recognition of the value of expert input in the policymaking process, the General Secretariat faces a shortage of qualified analytical personnel able to conduct evidence-based analysis, meaning a large part of this analysis is conducted at the higher political level with the help of non-permanent advisors. While these roles can still have significant benefits, their presence within the political level means that they are more likely to prioritise evidence in line with political goals and are less likely to focus on building knowledge bases unless there is clear immediate political value for it.
Where such leadership roles do exist at the administrative level, their potential is often constrained by limited authority or resources. Such an issue is evident in the Netherlands, where while most ministries have a dedicated chief science officer or equivalent function, responsible for improving uptake of knowledge in ministries and increasing the quality of knowledge produced on behalf of ministries, these individuals or units often lack mandate and budget. This lack of institutional backing limits their capacity to drive meaningful change or influence decision-making processes. As discussed previously, development of such institutional backing requires so called ‘tone at the top’ – in other words, clear senior political interest, helping legitimise and amplify the role of evidence champions.
When countries do successfully promote the value of EIPM via the administrative-political interface, the positive benefits are clear. Countries like Lithuania provide a model of this through initiatives such as the “Public Policy Decisions and Data Analysis Group,” which formalises the integration of evidence into decision making and ensures a direct link between administrative evidence providers and political leaders. In Belgium, in the rare cases where there is a senior figure with the explicit role of operating at the political to administrative interface, the ability to absorb evidence seems greater. This can be seen, for example, with the Competition Authority, where the setting up of a formal position of Chief Economist has allowed the development of a workforce able to understand both academic research and policy, and thus able to promote research at the political level.
Another success story can be seen in Estonia, where almost all ministries have science advisors. Development of such a feature in other countries could be a feasible way to ensure Ministers are more likely to be exposed to the value of using evidence. Such advisers could also help reduce the distance between policymakers and evidence and ensure that there are opportunities to consider evidence. However, the Estonian situation could be further improved: science advisors generally lack a specific mandate to have control over research funds and most science advisors are not able to funnel advice directly to political leadership, which may contribute to difficulties in evidence-related practices gaining influence over political will. Finally, in Greece, many policy advisers exist to provide advice to Ministers, and while their role combines policy management and provision of expertise, they also perform important tasks including steering policy and supporting key co-ordination functions. Outside of the seven countries, a strong example of a similar position can be seen in the United Kingdom Chief Science Advisers, which have been established as early as 1960 and have been in operations over 60 years (Box A.B.3 in Annex A).
Improving access to data for policy relevant analytical purposes
Access to high-quality data in a timely manner, as well as the appropriate tools to use the data, is key to EIPM. However, in many of the seven countries, access to data, and in particular to linked microdata can be one of the biggest obstacles to effective utilisation of evidence, even in cases where data sources themselves are rich and varied. The following section thus identifies three key functions needed to ensure that data can be fully employed in service of EIPM – coherent and connected data ecosystems, proportionality of legal and privacy constraints, and data literacy.
Developing coherent and connected data ecosystems
Generally speaking, existence of data themselves does not pose an issue – all seven countries reported, to various degrees, the existence of a rich and diverse data landscape, with a variety of data registers. However, all seven countries also reported that the fragmented nature of these data registers created significant difficulties in accessing this data, both for those within government and researchers outside of government. In many cases, large portions of public-sector data are managed in a decentralised manner – in the Netherlands, for example, there are more than ten organisations responsible for data collection, while in Latvia there are 38. Ministries also tend to own their own data, such as in Estonia, where Statistics Estonia is the collector and keeper of national statistics, but ministries own sectoral data. In some cases, such as in the Czech Republic, some private organisations have even become default providers for certain data, with unclear requirements around if this data can be provided to the public administration. In Belgium, organisations at arm’s length from government, such as the data banks play a major role.
This complex tapestry means that data products are ultimately provided through a fragmented set of interfaces and databases, with little to no signposting. This can make it difficult for researchers requiring specific data to know where to look. Poor governance of these sources also means that metadata information systems are rarely kept up to date or compatible for linkage. This, combined with the fact that lack of collaboration between different data providers leads to data obtainment processes varying across providers, means that successful data requests often require simply knowing the right person. There are some efforts to introduce digital tools able to overcome such issues – for example, in Belgium, there are some experiments on having administrative data on a Remote Access Server, which would remove the need for organisations to have to transfer data to each other for any exchange. However, in many cases such experiments are in their early stages.
A researcher’s job is likely to be even more difficult if they need to combine different data sources. Decentralised systems and lack of collaboration across data holders mean that data quality, data definitions and metadata (even when they are kept up to date) can all vary significantly from source to source. As such, comparability, let alone integration, can be a significant challenge. In many cases, data sources were developed via external procurement for administrative service delivery purposes, meaning the knowledge on how to combine them with other datasets for research purposes is often lacking altogether. This is exacerbated by the fact, mentioned above, that in some cases data providers have strong senses of ownership over their data, and are thus unwilling to work collaboratively. Furthermore, the mismatch in timeframes often found between access to data for research purposes and policy needs can lead to issues where data production takes too long for it to be used for policy emergencies or emerging priorities.
In some cases, this difficulty in accessing registers can have significant unintended consequences. In Estonia, low ease of access has led to people trying to circumvent the official system, going to agencies with one or several registers, which has led to the creation of warehouses in these agencies which do not always fully follow GDPR. This caused a scandal in 2023 where a think tank obtained the information of thousands of childless Estonian women, sent them a questionnaire with personal and sensitive questions, and asked to link survey data with national registry data of the respondents.
These issues and their ramifications are fully recognised in many of the seven countries. Several have started taking action to remedy them. Open data initiatives in the Czech Republic have obliged certain datasets to be published and have provided the impetus for bodies to start identifying, managing and publishing their datasets. Furthermore, the newly established Digital and Information Agency is implementing data management principles across the state administration. In Greece, the National Documentation Centre, with EU support, develops and organises repositories and databases on published scientific research also megatrends indicators. Latvia’s central Open Government Data Portal has also significantly increased ease of access to many data sources and represents a strong step in the right direction. Lithuania’s recent data governance agency represents another innovative good practice.
While these initiatives to create more open, accessible data sources are important, it is also vital that solutions are put in place for access to data sources that can’t be made fully open due to privacy regulations, but that can nevertheless play an important role in EIPM. Among the seven countries, a strong example of such an approach can be seen in the Netherlands, where, while different government entities do own their own data, the central statistics agency has access to it all, making the access process for researcher significantly easier. It may also be useful to provide some comparative examples beyond the seven countries in this area, as the United Kingdom provides another strong example of how to ensure coherence and integrability of different data sources, through its statistical authority, mandating that data are compliant with an extensive code of practice if they are to be approved as official statistics (see Box A.B.4 in Annex A).
As part of the 2018 Dutch Digitalisation Strategy, the government launched the Data Agenda Government, (NL Digital) outlining plans to improve the management of personal data, open data, and big data, leveraging analysis and integration for informed policymaking, and addressing societal challenges. Implementation responsibilities were shared by central and local governments, with co-ordination by the Ministry of the Interior and Kingdom Relations (Kingdom of the Netherlands, 2022[12]).
The Dutch Statistics Agency (CBS) also has access to data from across government, and thus can ensure standardization. It provides a portal for scientific researchers at Dutch universities to access microdata for research purposes. To access the data, which is stored on remote servers after the CBS removes all individual identity description elements, the researchers must follow a multi-step process submitting a proposal for research, which is reviewed by the CBS, before they receive a physical personal token on loan to access the data remotely.
The success of the Data Agenda government saw a follow-up strategy, the Intergovernmental Data Strategy (IBDS). The programme delivering its implementation – Realisatie IBDS - includes projects such as the Advice Function Responsible Data Use, Data Dialogues, Federated Data System, Data Catalogue, and an intergovernmental knowledge centre. These initiatives contribute to the practical implementation of the IBDS and foster a dialogue on responsible data use in the public sector.
Balancing legal and privacy constraints with the value of evidence
All EU Member States are covered by the General Data Protection Regulation (GDPR), and in general ensure the security of the data they have responsibility over. However, it is important that they strike a balance between such privacy concerns and the need for researchers, both inside and outside government, to have access to data to provide the necessary evidence to design economic and social interventions and provide citizens with quality public services.
In the seven countries, while universities and research institutions have the theoretical right to request more detailed statistics, legal and administrative hurdles can mean that obtaining data can be very difficult. To name just a few examples, the process can involve requiring express permission from a data protection authority and even an ethics committee, as is the case in Estonia, or permission from Parliament with a precise explanation of what the data is authorised for, as is the case in Latvia. In some cases, as reported in Greece, legal risks can mean that public servants are often reluctant to provide data without explicit permission from senior staff, even when the data request in question is legitimate. Such strict interpretations of privacy issues can also exacerbate data integration issues. In many cases, technical capacity and even willingness to link different data sources is present, but the stringency of the data access process can make it difficult to do so.
Interpretation of privacy laws can also be inconsistent. In many countries, including Belgium, the Czech Republic and Greece, guidance for data holders on use of data and data sharing is limited. This means that decisions to grant data access are treated on an ad hoc basis and can be somewhat dependent on personal contacts – indeed, knowledge of individual datasets is often concentrated in single individuals, meaning without knowing this person access can be close to impossible. Furthermore, data approval processes are not always transparent, increasing the risk of subjectivity within the process. There is also generally a lack of resources and guidance – in some cases, a lack of resources means that there are staffing gaps to help researchers through the data access process, while in others, fear of making a mistake and facing legal repercussions means employees of data institutions can be hesitant to help.
This combination of strict and inconsistent interpretations of data privacy regulations along with lack of guidance can mean that government analysts and academic researchers alike often find gaining access to data an arduous and time-consuming process – well over a year in some cases. This means that research questions often need to be answered with old data, as researchers are unable to wait to gain access to the data they need due to output and funding deadlines. This in turn, means that sufficiently granular data can be very difficult to access, leading to situations where policy analysis cannot be fully developed in a way that could help to target specific populations.
There is therefore a need to clarify any legal and operational ambiguities regarding data management and data sharing. The example of the Netherlands in Box 3.7 is also relevant here – by combining centralised co-ordination with local implementation, ensuring consistent data governance while allowing tailored approaches, and by a thorough yet clear-cut vetting process for any data access requests, the Dutch government strikes an effective balance between accessibility and privacy. Beyond the seven countries of the project, another strong example is France Secure Data Access System (Box A.B.4 in Annex A) which takes great lengths to ensure that data remains anonymised and well-secured, while ensuring internal coherence and thus ease of access to approved researchers. It offers a concrete good practice example of striking an effective balance and securing access to linked data for policy analysis.
Improving intra-organisational communication and transmission of evidence
Effective intra-organisational communication and evidence transmission are essential for EIPM. However, while analytical capacities exist to varying degrees within the ministries of several of the seven countries, these capacities tend to be fragmented across several teams and departments, resulting in scattered expertise that is challenging to co-ordinate. This is evident, for example, in Lithuania, where while a few ministries, including the Ministry of Agriculture and Ministry of Economy, have dedicated analytical units, most ministries lack such analytical units. A similar situation exists in Latvia, where only the Ministry of Economics has a dedicated analytical team. While other ministries do handle evidence, they do so sporadically and as a secondary responsibility, limiting their ability to systematically produce or use evidence. The Czech Republic faces a comparable issue, where analytical work is often confined to domain-specific departments, with some exceptions where one analytical team serves other departments, as is the case with the Ministry of Industry and Trade, which provides statistical services. The Czech Republic’s VAU has also helped expertise and evidence transmit across government, although as discussed previously, the fact that ministries are not required to collaborate with it means that its scope has not reached its full potential. However, overall analytical capacity remains imbalanced across ministries, meaning external expertise is often required.
In Greece, the limited availability of centralised knowledge-sharing systems is compounded by the fact that scientific personnel, though present in many ministries, often lacks clear roles in central policymaking. While the Ministry of Foreign Affairs provides an exception, with an institutionalised role for scientific personnel, most ministries fail to integrate their expertise into broader organisational knowledge-sharing processes.
Even where such dedicated analytical units do exist, siloed organisational structures can mean that evidence remains confined to specific departments. Belgium’s experience reflects this challenge – as above, ministries often rely on external aid, not because they do not possess the skills to undertake the analysis needed, but because there is no capacity for such information to be distributed, as analytical functions within ministries often operate in isolation. While there are some strong examples, such as the Ministry of Economy, which has developed internal observatories to share evidence across directorates, in most cases ministries lack such structures. In the Czech Republic, these issues are compounded by functional specialisation along professional lines, resulting in limited cross-functional teamwork and sometimes siloed evidence practices.
Inadequate resources and influence can also inhibit the ability of such units to effectively contribute to EIPM. In Lithuania, as mentioned above, some planning units have dedicated analytical teams. However, their small size means they often do not have the authority and resources necessary to effectively support decision making. In the Czech Republic, analytical teams are sometimes integrated within strategic planning departments. However, as many such teams are often at early stages of their development, there can be a lack of clarity regarding their influence in the policymaking process, further compounded by limited resources that restrict their functionality.
Action has been taken to address this challenge. For example, the Czech Republic and Latvia have recently set up units close to the centre of the Government to provide a core “anchor” capacity for such analytical work and support the line ministries with significant cross sectoral expertise. In the Czech Republic, the government created the Government Analytical Unit, a cross-sectoral team of analysts providing analytical support for units within ministries. While it currently predominantly focuses on Regulatory Impact Assessment, the ambition is to create a high standard of analytical work more generally within the public sector. In the Netherlands, the directors of major advisory bodies, such as WRR, and the Planning agencies participate in senior meetings with Ministry executives. In Ireland, the Irish Government Economic Evaluation Service (IGEES) provides a cross-cutting function, organising seminars such as the Dublin Economic Evaluation meetings, and a broader support function to IGEES analysts working in line ministries. Similarly cross-cutting support and analytical functions exist through the French Statistical Institute, INSEE, and its network of economists and statisticians across ministries.
These examples highlight the value of a “Hub and Spoke” approach, where units across governments support intraorganisational communication and transmission of evidence. Such an approach can be organised by central government or within ministries, to ensure that smaller teams are connected to the main hub of expertise and analysis in the ministry.
Understanding and developing the role of tools and frameworks in enabling implementation EIPM practices
Effective EIPM relies on the integration and application of robust tools and frameworks. In particular, ex ante impact assessments are crucial to ensuring that unintended consequences of novel regulations are kept in check, thus minimising the negative effects of proposed regulations and ensuring that new proposals are able to be as effective as possible. Strong evaluation practices can help ensure policies and regulations remain effective and reduce unintended consequences, through ex post evaluation. Strategic foresight helps policymakers make decisions based on a broader understanding of future possibilities. This section therefore discusses the use of these three tools in the seven countries forming part of this project. It finds that while guidance is frequently in place, there are often inconsistencies between intended models and the reality of implementation.
Impact Assessments
All countries in the EIPM project reported, to various extents, that they have guidelines for regulatory impact assessments (RIA) in place. However, they generally report a discrepancy between what their guidelines aim to achieve and the capacity of these guidelines to ensure that regulatory impact assessments are effectively informed by relevant evidence. In Greece, for example, while guidelines on impact assessments for primary laws exist, their successful implementation has proven to be challenging, with guidance not well-integrated into the policymaking process. Similarly, while the Netherlands’ impact assessment framework is meant to be used for policy processes, in reality policymakers often did not feel a strong urge to use it. However, new guidance was developed in the Netherlands during the course of the project related to the “Kompass”, with the intent to strengthen the implementation of the requirements. Guidance also exists in Latvia and in Belgium, in both cases requiring impact assessments to capture a wide range of economic, social, and environmental impacts. In Belgium, the RIAs produced are generally conducted after all decisions are made and can be lacking in quality and rigour. In Lithuania, impact assessments must be conducted for new legislation and substantive changes to existing legislation, while guidelines in the Czech Republic stipulate the need for comprehensive collection and analysis of data when assessing potential impacts of proposed regulations. However, Lithuanian ministries sometimes carry out such impact assessments during or even after the drafting of the relevant legislation, while in the Czech Republic, incorporation of evidence and scientific expertise has been limited to date.
While such guidelines are also, in many cases, intended to ensure uniformity in approaches to RIA, several countries see variation in such approaches across different government bodies. As examples, Estonia and the Czech Republic both report that the quality of impact assessment reports can vary significantly, and that there are inconsistencies in how guidelines are applied. The findings of the Estonian report identified that, despite the guidelines, many impact assessments were perceived to fall into a grey area, where it was not clear how much scientific data would be needed to carry them out, whether they would be eligible for funding, and whether they should be conducted internally or externally. Similarly, in Greece, impact assessments are not conducted within a systematic framework, and there is need for further guidance material and familiarisation with relevant processes.
Such a discrepancy between guidelines and implementation is also evident regarding proportionality principles. These principles offer guidance on how thorough impact assessments should be for different types of regulation, with more attention needed for regulations with major economic, social or environmental impacts. Therefore, proportionality is crucial for ensuring that resources are allocated in an efficient manner. Both Latvia and Lithuania make use of guidance that says resources must be proportional to the impact of the legislation, with Lithuania compiling a list of high-impact draft legal acts. However, in both cases this guidance has not been fully absorbed by the ministries – in Latvia, impact assessment format and analysis tend to be broadly the same for all draft laws, while in Lithuania impact assessments appeared unlikely to be conducted for most of the acts in the high-impact list. In Belgium, a proportionality principle does not exist at all, meaning analytical resources do not necessarily correlate to the issue at hand, making it harder for federal ministries to effectively target their resources.
These discrepancies between guidance and reality can create broader issues. While in many countries there is a formal requirement to conduct Regulatory Impact Assessment, often, implementation of impact assessment guidelines tends to be seen by policymakers as an add-on, rather than an essential part of policymaking. Indeed, ministries in several countries view impact assessments as a formal exercise. This means that even where RIAs are conducted, they are done in a formalistic manner, with little effort to integrate their findings into the policymaking process. Reasons for this vary – in some cases the pressure for policymaking to happen at a fast pace means that attention isn’t paid to perceived ‘extra steps’ such as RIA, while in others the perception is that the results of such assessments are unlikely to be considered in policymaking decisions, meaning civil servants are not motivated to pay attention to them. In others still, such as Belgium, the process is viewed as a way to verify political decisions, as opposed to scrutinise them and present alternatives.
Another key issue is that of stakeholder engagement. Several countries have, to varying extents, good practices here. Stakeholder engagement is incorporated into the RIA process in Estonia, with consultations with practitioners, researchers, and other stakeholders, as well as the requesting of additional research. In Greece, public consultation is a requirement for primary laws, with draft laws frequently posted on their consultation portal and guidelines on how to conduct stakeholder engagement provided. In the Czech Republic, the impact assessment system is transparent, with all legislative drafts submitted to the government published on a portal accessible to the public. However, all countries would benefit from a greater, more diverse level of stakeholder consultation, including from traditionally underrepresented groups. In Latvia, the development of the Single Portal for Development and Harmonisation of Draft Legal Acts has allowed the country to greatly improve its public consultation practices (see Box 3.8), although evidence on such consultation leading to substantive alterations of impact assessment documentation is limited. In Belgium, there is a system of strongly organised consultation, with defined and well-structured stakeholders. However, outside of these stakeholders, the system is not necessarily inclusive. Furthermore, there is no portal to allow people to follow and comment on the preparation of regulations.
The Single Portal for Development and Harmonisation of Draft Legal Acts, known as the TAP portal, provides the public with opportunities to follow all stages of legislative drafting. In 2021, the Cabinet of Ministers approved a regulation mandating any ministries drafting a regulation to publish a notification on the possibility to participate in the consultation process on this portal. This notification has to be published no less than 14 days before official submission of the document. Any interested party may take part in the consultations, including individuals, companies, and Non-Governmental Organisations. This has increased the transparency of the law-making process and allows various stakeholders to contribute to the drafting process in a clear-cut manner. Furthermore, it allows different organisations with different specialisms to contribute to legislation, allowing different types of evidence to be used.
Source: (State Chancellery, 2021[31])
In some of the countries, impact assessments conducted are overly qualitative in nature, with limited quantitative analysis. This is the case in Belgium and Latvia, where they only see quantitative requirements for analysis of administrative burdens. This lack of quantification also means that assessments are rarely based on hard data – although this is in part due to difficulties in accessing the necessary data (see relevant section). Furthermore, there is little utilisation of cost-benefit analysis and other methodologies. While qualitative assessments are important and necessary, this lack of quantification can reduce the usefulness of an impact assessment, in particular when it comes to comparing the regulation in question with possible alternatives.
The issues that many of these countries face are well recognised. Several countries are taking steps to remedy them. For example, in several ministries in the Netherlands, CSOs have organised activities to stimulate the uptake of impact assessments, while the 2023 update of the Policy Kompass ensured it was supervised by an interdisciplinary team from different ministries, ensuring it is applicable to a variety of contexts (see Box 3.9). In 2021, Estonia’s impact assessment methodology was updated to clarify the different stages of carrying out an impact assessment. Recognising the issue of consistency and uniformity. In Lithuania the government aims to clarify the legislative process for decision making, and establish a unified legislative monitoring system, to ensure that legislative processes are organised in a coherent manner.2
In the Dutch Government, to facilitate information exchange across ministries for the RIA process (entitled the “Policy Compass”), there is a cross-departmental Policy Compass working group that meets monthly to exchange best practices. Representatives of the expertise teams from all departments are members of the working group. The composition can change depending on the agenda. The chairmanship and secretariat of the working group lies with the directorate of the Ministry of Justice and Security.
In addition, the working group prepares the agenda of the more senior Policy Compass Steering Group, which meets several times a year and consists of directors of all departments, from policy directorates as well as implementing organizations and supervisors. Furthermore, each department has a team that ensures that there is sufficient knowledge and expertise in the organization to be able to use the Policy Compass effectively. This team, also known as the expertise team, can support users with advice and assistance in applying the Policy Compass.
Source: (OECD, 2020[32])
Ex post evaluation
Evaluation practices vary significantly across the seven countries involved in the EIPM project, often reflecting differing levels of prioritisation, resources, and guidelines. In Estonia, such evaluations are rare, with no mandatory framework or standardised methodology, although there is an increasing recognition of its importance, as evident through its identification as an important goal in the National Development Strategy 2035. Latvia similarly lacks a systemic programme for ex post evaluations – while guidance on conducting such evaluations is in place, it is not enforced. This, along with limited analytical capacity, means such evaluations are rarely conducted effectively. The Czech Republic also provides voluntary guidelines, although the Czech government has committed to applying them more widely. Belgium shows some isolated instances of effective evaluation—such as through the Court of Audit—but lacks cohesive co-ordination and a strong demand for evaluations across government. Greece remains unsystematic in its evaluation approach, underscoring the need for integration into policymaking. These findings point to persistent gaps in institutionalisation of evaluation, an issue recognised in the 2022 OECD Recommendation on Public Policy Evaluation, which calls for the carrying out of evaluations in a systematic manner, and for planning it early on in the policy process.
Some countries have made significant strides in the evaluation field. In 2018, the “Operation Insight into Quality” initiative in the Netherlands led to the Strategic Evaluation Agenda (SEA), requiring ministries to outline four-year evaluation plans. However, inadequate resources often limit the agenda’s efficacy. In 2023, Lithuania’s Ministry of Finance approved Methodological Guidelines on Evaluation, defining evaluation processes and highlighting evidence gap identification. Latvia and the Czech Republic have set up analytical units close to the centre of government with the aim to support the implementation of policy evaluation, or even contribute to high profile and high impact evaluations.3
Mobilising strategic foresight to inform policy development
The use of foresight practices for EIPM purposes also varies significantly, with some countries developing sophisticated approaches while others are in early stages of implementation. The Netherlands has a longstanding tradition in forward-looking approaches, with the establishment of the WRR in 1972 and other, more recent enhancements, such as the Rathenau Institute’s recent publications examining policymakers’ anticipatory practices. However, there is no central in-house foresight unit or dedicated team at the level of the national government. Estonia’s Foresight Centre, founded in 2017, operates independently of government, enhancing its ability to provide unbiased insights. While it influences the political discourse, as evidenced by its inclusion in party programmes, its collaboration with ministries remains limited in practice. There is also a lack of foresight training among the teams in the ministries. Greece’s new Special Secretariat of Foresight provides advice on emerging challenges and opportunities at the centre of government, though its novelty and need for skilled staff mean it will require time to gain influence.
In Latvia, while policy planning is well-developed, it does not necessarily imply substantive foresight efforts in many cases, beyond mechanical planning approaches as required to meet European Commission funding requirements. Only some ministries incorporate analytical forward-looking approaches into their planning programmes via use of modelling. The Czech Republic also has decentralised foresight practices, most notably related to R&D, regional development, and environmental policy, with methodological guidelines acknowledging foresight’s value. While a cross-ministerial unit exists, it does not co-ordinate effectively. In other country cases, strategic foresight exists, but tends to be isolated to very few areas of government. For example, Lithuania institutionalised the application of strategic foresight through developing the State Progress Strategy “Lithuania 2050” (Nakrošis et al., 2025[4]). The institutional structure includes a range of institutions across the executive, the centre of government and Parliament. STRATA serves as a partner for other institutions in integrating future-oriented approaches (OECD, 2025[33]). Similarly, while Belgium’s Federal Planning Bureau has strong foresight practices, use of strategic foresight outside the Bureau are limited. A particularly strong example of a well-organised foresight system can be found in a European country beyond the seven involved in the project, namely in Finland, where many foresight capacities across different ministries and policy areas of government combine the Finnish National Foresight Network and do inform future coalition agreements and coalition programmes (Box A.B.6 in Annex A).
Strengthening co-ordination of EIPM practices across government
Many countries face significant challenges in co-ordinating EIPM practices, resulting in inefficiencies, duplication of efforts, and a lack of alignment across ministries. This is often due to strong functional specialisation along professional lines, and little culture of cross-silo project working. The need for cohesive, whole-of-government strategies that promote collaboration and ensure the effective use of evidence in policymaking is thus increasingly urgent.
One significant challenge regarding co-ordination of EIPM practices is the fragmentation of advisory networks and policymaking bodies. Many such bodies exist, and while they generally perform effective roles within their area of competence and specialisation, lack of co-ordination means their roles often overlap. In Estonia, the Science Adviser Network aims to improve the capacity of research and development institutions to conduct socially relevant research, while another network, the Coordination Network for Research and Development, is intended to discuss all relevant R&D issues, using a whole-of-government approach. However, many members of the Estonian government find the distinction between these two networks to be blurry, contributing to uncertainty and inefficiency in how evidence is mobilised across the two.
This fragmentation is not unique to Estonia. In Latvia, advisory bodies often operate within narrowly defined policy domains, limiting their ability to address complex, multidisciplinary issues that require comprehensive evidence. While numerous advisory bodies exist, they typically focus on specific areas, such as economic or social policy. This results in a siloed approach that hinders collaboration between different sectors and agencies. This highlights the need for more spaces for multidisciplinary discussions, emphasising the importance of integrating diverse expertise in policymaking processes. Similarly, there is a perception in Greece that the horizontal operation of the administration needs to be strengthened, with few interorganisational networks dedicated to mobilising Evidence and EIPM practices. Although each ministry has specialised scientific personnel embedded within their structures, there is no cohesive network across ministries. In Lithuania, the absence of a central co-ordinating body results in ministries navigating EIPM practices independently. This lack of overarching guidance increases the risk of duplicative research efforts and inconsistent evidence use.
In some cases, cross-domain advisory bodies do exist, and they are starting to integrate and develop EIPM approaches. In Belgium, the Economic Council has a broader mandate and greater capacity for cross policy engagement across the economic and social issues and is currently expanding its capacity to co-ordinate on matters related to addressing the need for the green transitions. On the positive side, Belgium’s Science Policy organisation funds a range cross-institution initiatives and has made a priority in its strategic workplan, which remains subject to funding constraints. In the Netherlands, several networks exist that try to foster co-operation between ministries, including an informal network of CSOs. However, knowledge providers feel that it would be beneficial if networks around EIPM were stronger, particularly when it comes to multifaceted issues such as green transitions and migration. Lithuania also has several networks, including an evaluation co-ordination group, a network of strategic management co-ordinators, and a network of citizen services.
While overarching EIPM co-ordination organisations and mechanisms are rare, several sector-specific organisations and mechanisms exist. In the Czech Republic, for example, the Research Development and Innovation Council plays an important role in identifying priorities for research and co-operating with different institutions working on Research and Development. It thus plays a valuable role in the evidence-for-policy ecosystem in the country. In Estonia, the establishment of the Research and Development Council (TAN) plays a crucial role in guiding national development policy and advising the government on matters related to research, development, and innovation. While this Council focuses on shaping policy for science rather than synthesising scientific policy decisions, its existence highlights the importance of high-level structures in facilitating evidence access.
The Netherlands has a rich and well-developed landscape, although it is also fragmented and characterised by silo-thinking. It is thus difficult for the demand side to have an overview of which knowledge resides where, and for many suppliers to have an overview of how their work could complement that of other suppliers. While some actors effectively work across siloes, they can have difficulty in finding the right recipient for their output due to silo-thinking and low absorption capacity on the demand side. In Lithuania, the Expert Committee of Science and Higher Education Policy in the Research Council of Lithuania effectively provides recommendations on improvements for the science and higher education system. However, its role could be expanded to a general evidence-for-policy platform providing policy relevant advice and analysis in different policy fields. Finally, in Greece, a number of structures for the mobilisation and analysis of data are established in the field of economic and social policies, with the Council of Economic Experts and the Unit of Experts in Employment, Social Insurance, Welfare and Social Affairs (MEKY). While the existence of all these organisations is valuable, their sector-specific scopes mean they do not always aid in overcoming the siloed way of thinking that often acts as an obstacle to effective co-ordination of EIPM practices, and they cannot be fully counted as interorganisational structures.
There is generally a recognition of the issues this siloed approach to EIPM causes, and efforts to mitigate it. In Latvia, for example, the Cross-Sectoral Coordination Department's aims to improve co-ordination by creating a unified space for evidence-based discussions across ministries. In the Czech Republic, the Government Analytical Unit (VAU), discussed previously, represents a strong step in the right direction. The Unit has several functions, including helping policymakers and analysts in ministries with various problems related to analysis for policymaking. The existence of a central organisation with such a role has great potential to harmonise practices, processes, and approaches to EIPM across government, facilitating greater co-ordination. However, role, functions and mandates of the VAÚ are not formalised, and collaboration with ministries is voluntary and thus can vary from ministry to ministry. In Lithuania, the Office of Government has developed an informal network of civil servants who fulfil analytical functions spanning all line/sectoral ministries, encouraging the exchange of knowledge. Furthermore, the Office of Government has established a working group spanning several government organisations to address the economic consequences of Russia’s war in Ukraine, which facilitates the utilisation of analytical capabilities to deliberate an evidence-informed policy response.
The COVID-19 pandemic provided strong impetus for cross-coordination approaches of EIPM. In Greece, for example, the National Public Health Organisation (EODY) operated as a central hub for EIPM. In Belgium, some sectors with more developed use of scientific evidence, such as Public Health, did see increasing collaboration. As a result, several key knowledge institutes in the health sector have built on the links made during COVID-19 and intend to continue strengthening these collaborations and identifying topics that they can all work on.
Co-ordination heavily relies on the proper mapping of existing organisations and, if necessary, of organisations that need to be established when they are missing. Some organisations operate within government as ‘in house’ advice, enjoying privileged access to information before it reaches the public sphere. These organisations have a policy relevant agenda, and their mandate is clearly one of an advisory organisation with evidence as its basis. The other locus can be found on the other end of the spectrum, in research organisations external to government, or in political institutions that in addition to their research agenda produce policy-oriented research and respond to critical policy questions. Their profile tends to be that of a research organisation with relevance to policy. Identifying the scope, capacity, and focus points of these organisations, is an important starting point for then highlighting where research gaps exist.
However, there is not necessarily alignment between the timelines of research and policymaking, which can hinder their policy impact. The more distant the knowledge brokers are from the political level, the more independence they enjoy. They also tend to enjoy higher resilience, given their distance from the political level allows them to remain unaffected by political turbulence and changes of government. On the other hand, these organisations, situated outside political structures, need to work harder to formulate a policy relevant agenda and to ensure an ‘effective hearing’ process from the policymakers that will increase the potential of their research outputs to achieve an impact. In addition, they need to invest extra efforts and resources to effectively sense and have access to the evidence needs of policymakers. This extra effort usually involves interaction with reliable policy networks and constant evaluation of policy debates in the public sphere. Interactions with the international level in terms of data and expertise can also provide some form of a unifying framework, as scientific results and evidence can also be valid and relevant across borders, though local circumstances must of course be taken into account.
Nurturing an EIPM-conducive culture in public administrations
Promoting good knowledge management and governance of evidence
Good knowledge management and governance of evidence reduces the barriers to entry for its uptake, incentivising such uptake and thus helping to foster a culture of evidence use. One key aspect of good governance is adequate capacity to effectively track, store, and share evidence. Any research conducted for policy purposes should be noted and logged in such a way that different teams can use it, and understand what it was previously used for, with ease. It should also be stored so that it can be referred to at a later date.
Such systems are relatively underdeveloped in seven countries of the EIPM project. Almost all countries reported either not having research tracking and storage systems in place or experiencing deficiencies with their systems. Where such systems do exist, they tend to be isolated examples – for example, a unit within the Department of Labour in the Czech Republic keeps track of all research performed within the line ministry, while the Ministry of Social Security in Belgium maintains a knowledge cadastre to keep track of ongoing and recently completed research. In both cases, the research that is kept track of is specific to the organisation itself. The Netherlands also reported that it lacks formalised structures for archiving evidence systematically. In other cases, such as in Latvia, centralised databases do exist, but awareness and use of them is very low and they remain limited to publicly funded research. This leads to situations where knowledge tracking is inconsistent and recall of previous research is dependent on individual memory rather than institutional memory.
Another important component of good co-ordination and governance of evidence is ensuring quality standards across government, both in terms of the quality of evidence and evaluations and in how evidence is to be used. To do this, it is important to have clear, well-developed, and widely followed guidance documents and frameworks in place so that policy analysts have strong points of reference. Generally speaking, guidance documents for policy processes are relatively common in the EIPM countries – the Czech Republic, for example, has developed various methodological guidance documents, including for Regulatory Impact Assessments (see dedicated section) and EU-funded projects, while Greece’s Integrated Plan of Government Policy includes annual action plans for each ministry. However, while such guidance certainly has the potential to have a positive impact on EIPM, it does not necessarily include explicit quality standards for evidence, or for its use. Such standards can also have value in guiding policymakers on how best to commission evidence from research organisations.
Where evidence-specific documents do exist, they are often not compulsory. For example, in the Czech Republic, one methodology provides guidance on the evidence required to define a problem when preparing strategic documents. However, it is not binding, and as such its deployment varies. Similarly, the Netherlands’ Toolbox for Policy Evaluations, designed to support evidence-informed evaluations, has seen its application vary across departments, leading to inconsistencies regarding if and how evidence is integrated into the policymaking process. Other cases lack enough concreteness, such as Estonia’s National Development Strategy, which recognises the need for increased use of research in decision making but does not provide an action plan on how to reach this goal.
On the other hand, Lithuania’s Analytical Information Library provides a good example of effective evidence storage (Box 3.10). Similarly, Latvia also runs a government wide library of evidence.
The Analytical Information Library (AIL) is a collaborative effort between the Government Chancellery, the Government Strategic Analysis Centre, and the Martynas Mažvydas National Library of Lithuania. The AIL aims to facilitate EIPM and promote the use of research in the field of public administration. AIL provides digital access to approximately 1,700 publicly funded research reports, policy assessments, and analytical studies produced in Lithuania between 2000 and 2023. By hosting a comprehensive collection of research in one accessible platform, AIL helps avoid duplication of efforts and encourages decision makers in the public sector, including politicians and employees of ministries and municipalities, to utilise evidence in analysing societal trends and formulating policy decisions. The platform fosters a culture of evidence-informed governance, aligning academic research with policymaking efforts, and serves as an essential resource for effective policy development, implementation, and evaluation. With continuous updates and future expansion, AIL is poised to become an integral part of the country's public information resources and data system, fostering effective communication between policymakers, researchers, and intermediary organisations.
Source: (Nakrošis et al., 2025[4])
Identification of knowledge needs at a cross-ministerial level is also key for the effective governance of evidence. However, existence of such cross-ministerial knowledge planning is rare, although several organisations within the seven countries have their own knowledge agendas. In Belgium, the Ministry for Social Security has an annual research agenda which aims to finance policy-oriented academic research in relevant domains, based on evidence needs. However, there is no whole-of-government annual working plan to identify evidence needs. Similarly, in the Czech Republic, the Ministry of Labour and Social Affairs supervises two government-funded public research organisations which perform research and co-operate with the line ministry, whose individual departments agree on a research agenda with the institutes for a two-year period. However, nothing like this exists at the whole-of-government level.
Perhaps the best example of cross-ministerial knowledge needs being addressed can be found in the Netherlands. Here, the Dutch Research Agenda, established in 2018, aims to jointly ensure scientific and societal impact, and is carried out across the entire knowledge chain in an interdisciplinary manner. On top of this, certain individual ministries do their own knowledge planning – for example, the Ministry of Social Affairs has a knowledge agenda to set plans for the next year, and is discussed every six months, bringing together policymakers from across the whole ministry.
Conflict of interest mechanisms are also a key facet of good evidence governance – without them, the risk of political or other types of capture can significantly undermine the legitimacy and uptake of evidence in policymaking. Existence of such mechanisms varies significantly across the seven countries, In Belgium, a few advanced knowledge broker organisations, namely KCE and Sciensano, require conflict of interest declarations. In Estonia, conflict of interest of scientists was identified as a major issue – in particular, the fact that many scientists have difficulty understanding their role in the policymaking process means that they end up acting as advocates of their own scientific field or institution, rather than acting as an independent expert. In the Czech Republic, the regulation of conflicts of interest for advisory bodies remains unclear.
Ensuring recognition of the value of EIPM through leadership
For EIPM to be effective, political factors and executive leadership must align with and actively support evidence-informed approaches. Political support and leadership of EIPM not only sets a precedent for evidence-driven decision making with a long-term view, but it also promotes a culture of learning and adaptation within government, enabling policies to evolve in response to new information and shifting societal needs.
However, in many cases, a political focus on short-term achievements over sustained, evidence-backed decisions is a significant barrier to EIPM. In several countries, political timelines and the pressure to deliver quick legislative results encourage rapid policy output that often does not provide sufficient opportunity for integrating evidence into decisions.
In some of the seven countries, for example in the Netherlands, Belgium, and Latvia, coalition governments may create a specific context for the uptake of evidence in policymaking. In particular, while there might be significant demand for evidence when such coalition agreements are crafted, the window for supplying evidence is often very short. In the Netherlands, perception of knowledge as a commodity able to be supplied by consultancies has led to a loss of substantive knowledge within ministries, an issue exacerbated by an emphasis on generic skills over knowledge in specific subject areas. This issue is compounded by excessively high staff rotation amongst civil servants. Staff rotation in and of itself can have positive impacts such as sharing expertise across government, and preventing the forming of siloes. However, when rotation periods are too short, civil servants end up not having enough time in a team to fully gain such expertise, and this can also impact organisational memory. For example, Greece suffers from the fact that policy formulation and design is usually entrusted to Ministers’ offices, and therefore is generally carried out by political advisers. As previously discussed, this is exacerbated by the lack of career incentives among civil servants to invest in skill development, as generally possession of such skills is not rewarded as compared to managerial paths.
However, the project also shows that foundations needed for a pro-EIPM culture already exist. In Estonia, findings from the project show that public administration staff are often highly educated, interested in learning, and in favour of using evidence in their decision making. The Netherlands is working on strengthening the use of evidence with the introduction of the Policy Kompass. In Greece, while experts have historically had only a marginal role in the decision-making process, there is rising consensus over the importance of evidence in policy, and indications that politicians increasingly seek to involve experts in policymaking. In the Czech Republic, despite frequent preference for personal experience over research findings by public officials, there is a long-established tradition of using, to some extent, research in policymaking. There is also growing awareness of EIPM in Belgium, Latvia and Lithuania, including as a result of this project.
References
[24] CoARA (2023), “The Commitments”, https://coara.eu/agreement/the-commitments/ (accessed on 30 January 2024).
[29] Crown Commercial Service (2021), “Research & Insights”, https://www.crowncommercial.gov.uk/agreements/RM6126 (accessed on 16 April 2025).
[22] Davies, S. et al. (2017), “Knowledge into action – supporting the implementation of evidence into practice in Scotland”, Health Information and Libraries Journal, Vol. 34/1, pp. 74-85, https://doi.org/10.1111/hir.12159.
[2] Dorren, L. et al. (2024), Building capacity for evidence-informed policymaking in governance and public administration in a post-pandemic Europe: Country analysis, policy recommendations & implementation roadmap - The Netherlands, Scharfbillig, Mario; Smits, Paul, editor(s) Publications Office of the European Union, Luxembourg, https://doi.org/10.2760/5012478 (accessed on 16 April 2025).
[16] European Commission (2019), “Scientific Advice to European Policy in a Complex World”, https://www.kowi.de/Portaldata/2/Resources/fp/sam-scientific-advice-european-policy-in-a-complex-world.pdf (accessed on 31 March 2025).
[12] European Commission (n.d.), “EU Policy Lab”, https://policy-lab.ec.europa.eu/index_en (accessed on 31 March 2025).
[23] Ghent University (n.d.), Portfolio of research dimensions, https://www.ugent.be/en/research/research-strategy/evaluation/portfolio-research.htm (accessed on 27 September 2023).
[21] Gluckman, P., A. Bardsley and M. Kaiser (2021), “Brokerage at the science–policy interface: from conceptual framework to practical guidance”, Humanities and Social Sciences Communications, Vol. 8/1, https://doi.org/10.1057/s41599-021-00756-3.
[30] Government of the Kingdom of the Netherlands (2022), Data Agenda Government - Digital Government, https://www.nldigitalgovernment.nl/overview/new-technologies-data-and-ethics/data-agenda-government/ (accessed on 20 September 2023).
[8] JRC (2023), “Competence frameworks for policymakers and researchers”, https://knowledge4policy.ec.europa.eu/projects-activities/competence-frameworks-policymakers-researchers_en (accessed on 28 March 2025).
[1] Koppel, A. et al. (2025), Building capacity for evidence-informed policymaking in governance and public administration in a post-pandemic Europe: Final integrated country report - Estonia, Kock, Elias; Olajos-Szabo, Alexandra editor(s), Publications Office of the European Union, Luxembourg, https://doi.org/10.2760/4815761 (accessed on 16 April 2025).
[20] Krieger, K. and L. Melchor Fernandez (eds.) (2023), An evaluation framework for institutional capacity of science-for-policy ecosystems in EU Member States, Publication Office of the European Union, Luxembourg, https://publications.jrc.ec.europa.eu/repository/handle/JRC136095.
[3] Ladi, S. et al. (2025), Building capacity for evidence-informed policymaking in governance and public administration in a post-pandemic Europe - Country analysis, policy recommendations and implementation roadmap - Greece, https://doi.org/10.2760/8311826.
[14] Ministry of Regional Development (2024), “Strateduka”, https://mmr.gov.cz/cs/microsites/portal-strategicke-prace-v-ceske-republice/strategicke-projekty/strateduka (accessed on 30 October 2024).
[4] Nakrošis, V. et al. (2025), Building capacity for evidence-informed policymaking in governance and public administration in a post-pandemic Europe - Country analysis, policy recommendations & implementation roadmap Lithuania, Almeida, Mara; Olajos-Szabo, Alexandra editor(s) 2 Publications Office of the European Union, Luxembourg, https://doi.org/10.2760/0030376 (accessed on 16 April 2025).
[13] National Documentation Centre (2023), “Knowledge4Policy@EKT”, https://knowledge4policy.ekt.gr/knowledge4policy (accessed on 31 March 2025).
[33] OECD (2025), Building Anticipatory Capacity with Strategic Foresight in Government: Lessons from Lithuania, Italy, and Malta, OECD Public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/d7eb0bb6-en.
[7] OECD (2025), “Building capacity for evidence-informed policymaking in Belgium: Assessment and recommendations roadmap”, OECD Public Governance Policy Papers, No. 71, OECD, Paris Cedex 16, https://doi.org/10.1787/223b01a8-en.
[34] OECD (2025), OECD Regulatory Policy Outlook 2025, OECD Publishing, Paris, https://doi.org/10.1787/56b60e39-en.
[26] OECD (2024), “Building Capacity for Evidence Informed Policymaking in Governance and Public Administration in Belgium: Preliminary Findings from the Diagnostic Report & Needs and Gaps Assessment”, https://bosa.belgium.be/sites/default/files/content/documents/Formatted-Diagnostic%20%2B%20Needs%20and%20Gaps%20For%2030th%20with%20cover%20(002).pdf (accessed on 15 April 2025).
[6] OECD (2024), “Building capacity for evidence-informed policymaking in Latvia: Assessment and recommendations roadmap”, OECD Public Governance Policy Papers, No. 65, OECD Publishing, Paris, https://doi.org/10.1787/2b43bc19-en.
[32] OECD (2020), Review of International Regulatory Co-operation of the United Kingdom, OECD Publishing, Paris, https://doi.org/10.1787/09be52f0-en.
[10] Pearson, H. (2024), “Science could solve some of the world’s biggest problems. Why aren’t governments using it?”, Nature, Vol. 636/8041, pp. 26-30, https://doi.org/10.1038/d41586-024-03906-0.
[5] Petr, B. et al. (2025), Building capacity for evidence-informed policymaking in governance and public administration in a post-pandemic Europe - Country analysis, policy recommendations and implementation roadmap Czech Republic, Olajos-Szabo, Alexandra; Cannata, Giuseppe editor(s) Publications Office of the European Union, Luxembourg, https://doi.org/10.2760/9257621 (accessed on 16 April 2025).
[11] SAPEA (2024), “Report on challenges for early and mid-career researchers in the provision of science advice”, https://scientificadvice.eu/reports/report-on-challenges-for-early-and-mid-career-researchers-in-the-provision-of-science-advice/#elementor-toc__heading-anchor-14 (accessed on 16 April 2025).
[9] Scharfbillig, M. et al. (2024), Science-for-policy ecosystems through the eyes of professionals, https://doi.org/10.2760/7490500.
[31] State Chancellery (2021), “Public Portal on Draft Legislation [Tiesību aktu projektu publiskais portāls]”, https://tapportals.mk.gov.lv/# (accessed on 17 January 2024).
[18] Stier, J. and S. Smit (2021), “Co-creation as an innovative setting to improve the uptake of scientific knowledge: overcoming obstacles, understanding considerations and applying enablers to improve scientific impact in society”, Journal of Innovation and Entrepreneurship, Vol. 10/1, https://doi.org/10.1186/s13731-021-00176-2.
[27] Swedish Competition Authority (2016), “Swedish Public Procurement Act”, https://www.konkurrensverket.se/globalassets/dokument/informationsmaterial/rapporter-och-broschyrer/informationsmaterial/swedish-public-procurement-act.pdf (accessed on 16 April 2025).
[19] Tellmann, S. and M. Gulbrandsen (2022), “The other side of the boundary: Productive interactions seen from the policy side”, Science and Public Policy, Vol. 49/4, pp. 621-631, https://doi.org/10.1093/scipol/scac013.
[28] TNO (2025), “Collaboration with TNO”, https://www.tno.nl/en/collaboration/ (accessed on 16 April 2025).
[17] Torres, J. and M. Steponavičius (2022), “More than just a go-between: The role of intermediaries in knowledge mobilisation”, OECD Education Working Papers, No. 285, OECD Publishing, Paris, https://doi.org/10.1787/aa29cfd3-en.
[15] Van Eerd, D. and R. Saunders (2017), “Integrated Knowledge Transfer and Exchange: An Organizational Approach for Stakeholder Engagement and Communications”, Scholarly and Research Communication, Vol. 8/1, https://doi.org/10.22230/src.2017v8n1a274.
[25] Zika, E. (2023), “Eleni Zika on curiosity-driven research and its contribution to policy”, https://scientificadvice.eu/podcast/eleni-zika-on-curiosity-driven-research-and-its-contribution-to-policy/ (accessed on 31 March 2025).
← 1. As clarified earlier the engagement with citizens and stakeholders was not a particular angle of analysis in this project. However, the trainings have made strong reference in the importance of engaging with citizens early in the process to grasp public attitudes towards policies and inform policies with citizens perspectives.
← 2. These steps are consistent with the broader picture offered by the latest 2025 OECD regulatory outlook, which allow to put these efforts to strengthen RIA in a wider perspective, with Lithuania, Estonia and the Netherlands among the countries to have pursued recent reforms to improve their RIA frameworks (OECD, 2025[34]).
← 3. The 2025 OECD Regulatory Policy Outlook also offers some cross-cutting wider information in terms of the ex post evaluation of primary laws and subordinate regulations (OECD, 2025[34]). Often, several issues remain. Ex post evaluation tends to follow long timeframes and be-resource intensive, rather than the shorter, more frequent, and lighter-touch rapid evaluations which would be more appropriate to fast-moving policy processes. Furthermore, in most countries evaluation considerations are not integrated at the design phase of policies, which limits their effectiveness.