Government and academia: How government can engage with academia more effectively

Alongside a spurt of activity from research institutions, learned societies and funding bodies, government is creating more opportunities for researcher engagement. But existing mechanisms are not enough on their own to create effective and cohesive researcher engagement systems. Expanding researcher engagement initiatives may yield positive results, but more activity is not always better activity and an uncoordinated approach can take up resources that could be better allocated.

To improve outcomes from researcher engagement, departments need to be strategic in how they use limited resources. This means being clear on why they are engaging with researchers and what they want to get out of it. When departments can articulate what they want from academia, they can create routes for engagement that help fulfil these objectives. As part of this, departments also need to think about how engagement mechanisms can be joined up and who has responsibility for the research engagement agenda. To this end, departments need to:

• create their own researcher engagement strategies 
• provide clear, visible leadership to ensure the policy–research interchange is sustained 
• align culture and incentives to give officials the resources, permission and space needed to access the most up-to-date evidence available and feed it into decision making.

Departments need strategies to articulate what they want from academic input

Different departments will have their own reasons for seeking academic input and researcher engagement needs to cater to departmental contexts. But proactive thinking about how academic input can meet these objectives is not always evident across government.

While specific officials, teams or directorates should have the scope to create their own networks and localised initiatives, the development and deployment of researcher engagement infrastructure at a departmental level – which can include ARIs, secondment schemes, What Works Centres, advisory groups or expert networks – require more resources and should be developed with broader goals in mind. Departments need to articulate more clearly why they are engaging with researchers and identify the routes and mechanisms that can best attract evidence to meet their varied needs.

Academic literature points towards the positive impacts that engagement with researchers can have on policy making. ¹⁹¹ For example, on secondments, see Benson-Egglenton J and Flinders M, ‘Understanding the dynamics of research policy fellowships: an evaluative analysis of impacts and ecosystem effects’, Evidence & Policy, 2025, vol. 21, no. 2, pp. 257–78 (at pp. 268–71), retrieved 29 June 2026, https://doi.org/10.1332/17442648Y20 24D000000040. On ARIs, see Oliver K, Boaz A and Cuccato G, ‘Areas of research interest: joining the dots between government and research at last?’, F1000Research, 2026, 11:1509, retrieved 29 June 2026, https://doi. org/10.12688/f1000research.127542.2  There is an emphasis on building a stronger pipeline of evidence into government from external research organisations through an expansion in engagement initiatives, but the ecosystem is varied and can be confusing. If departments are not clear about what they want from researchers and the routes that officials can use to access external evidence, then already time-poor officials will be less likely to engage with external researchers. A messy ecosystem coupled with a lack of clarity from departments means that the government is not extracting as much value from the UK’s world-leading research system as it could. ¹⁹² Oliver K, Hopkins A, Boaz A, Guillot-Wright S and Cairney P, ‘What works to promote research–policy engagement?’, Evidence & Policy, 2022, vol. 18, no. 4, pp. 691–713, retrieved 26 June 2026, https://doi.org/10.1 332/174426421X16420918447616

To this end, departments should develop researcher engagement strategies. These should set a clear direction from senior leadership about what the department wants to get from researcher engagement. The strategies do not need to follow one template strictly and different departments will have different approaches. But they should inform how mechanisms for researcher engagement are developed, or reformulated, to meet the department’s goals, while clearly setting out the department’s priorities for engagement to external researchers and knowledge mobilisers.

The ARI commissioning process should already bring departmental leaders with responsibility for evidence and engagement together to identify and disseminate their department’s research requirements. This could provide a starting point for carrying out a wider needs assessment, which can underlie researcher engagement strategies. There are already examples, such as at the Ministry of Justice (MoJ), where ARIs are being used to co-ordinate a wider academic engagement strategy. Some departments, such as the Department for Energy Security and Net Zero (DESNZ), also engaged broadly with other parts of Whitehall as they developed mission-specific ARIs. Such examples of best practice in co-ordinating across government should be applied to the process of creating the kind of researcher engagement strategies we are advocating for – identifying overlapping evidence needs and synergies in the mechanisms used to service them.

There is a risk that orientating researcher engagement strategies around tightly defined outcomes does not take into account the kinds of unexpected insights and connections that academic–policy interactions can generate. ¹⁹³ Sasse T and Haddon C, How Government Can Work with Academia, Institute for Government, 2018, retrieved 26 February 2026, pp. 11–13, www.instituteforgovernment.org.uk/publication/how-government-can-work-academia  Strategies should therefore set the direction of travel, with defined outcomes sitting alongside broader goals such as exposing officials to new approaches or introducing a challenge function.

Strategies for researcher engagement need to explain how different mechanisms join up

A coherent system for academic–policy engagement needs to be more than the sum of its parts. We heard many examples where mechanisms have been developed in an ad-hoc way, where they exist in parallel with other engagement routes, rather than as part of an integrated system within a departmental evidence ecosystem. The result is unwieldy and overlapping routes for engagement. This can lead to duplication and hamper the development of longer lasting collaborations.

‘Joined up’ can mean a lot of things. In this context it means that the different mechanisms within a department are designed to take account of the existing researcher engagement offering and, where possible, directly interact with each other to promote positive outcomes. For example, engagement with a department via ARIs could lead to interactions that feed into expert networks, or to commissioning a piece of research. The aim is deeper and more sustained relationship-building between researchers and policy makers through the full suite of routes available, rather than isolated one-off interactions.

There is some evidence that newer forms of engagement have already been designed with greater join-up in mind. The college of experts, for instance, is a relatively new and flexible advisory group structure. We heard examples of secondees being invited to join a department’s college of experts, or of the structure allowing departments to commission pieces of research from its members. Taking part also includes using their profile “in order to improve two-way communications within their respective stakeholder communities”, suggesting that the colleges can also be used as a mechanism for building expert networks between a department and the research community. ¹⁹⁴ For example, Department for Culture, Media and Sport, ‘DCMS college of experts – terms of reference’, GOV.UK, (no date), retrieved 16 June 2026, https://assets.publishing.service.gov.uk/ media/61fa5c7de90e07689b389055/DCMS_College_of_Experts-_Terms_of_Reference_November_2021.pdf  There is a strong case for having a similar model in other departments.

Government has an important role in creating the conditions for academic–policy interactions and evidence flows through the development of engagement mechanisms that have a range of beneficial impacts. To maximise these benefits, thought needs to be given to how these mechanisms operate and interact with one another, as creating a joined-up system requires co-ordination and planning. Getting this right will help generate a more cohesive and coherent evidence ecosystem and create more impactful interactions. Strategies for researcher engagement provide an opportunity for departments to work out what mechanisms they need to meet their evidence needs and how these should join up.

Clear leadership is needed for government to engage more strategically with academia

We found many individual examples where strong leadership has helped develop researcher engagement efforts. But as highlighted in our previous report on how government and academia can work together more effectively, ¹⁹⁵ Sasse T and Haddon C, How Government Can Work with Academia, Institute for Government, 2018, retrieved 26 February 2026, www.instituteforgovernment.org.uk/publication/how-government-can-work-academia  departments often lack a leadership structure with clear lines of accountability for researcher engagement. Oversight of researcher engagement and evidence use is fragmented, with responsibilities falling with different senior departmental leaders, from permanent secretaries to CSAs, heads of analysis or policy profession leads. This has held back the development of more strategic researcher engagement strategies. The result is an uncoordinated system characterised by ad-hoc initiatives, where officials do not necessarily know when or how to leverage academic expertise. This means that opportunities where researcher engagement may have helped improve policy outcomes have not been seized. ¹⁹⁶ Ibid.  Responsibilities should be more clearly delineated at the departmental level so that individual leaders can be held to account for delivering researcher engagement strategies and that they have the authority to reorientate the existing engagement structures in ways that align with strategic goals.

A one-size-fits-all approach to how responsibilities are split among senior leaders will not suit different departmental contexts. Departments will lean on researcher engagement mechanisms for different reasons and some departments – such as the Department for Environment, Food and Rural Affairs (Defra) – already have mature evidence and engagement systems in place. ¹⁹⁷ Shaxson L, Hood R, Boaz A and Head B, ‘Knowledge brokering inside the policy making process: an analysis of evidence use inside a UK government department’, Evidence & Policy, 2025, vol. 21, no. 1, pp. 6–25, retrieved 29 June 2026, https://doi.org/10.1332/17442648Y2024D000000028  The ability of different actors to lead researcher engagement efforts will also depend on the structure and status of relevant roles within departments. While some chief scientific advisers have carved out a prominent role – for example, the chief scientific adviser at Defra sits on the executive committee and is director general for science, data and analysis – previous Institute for Government research found that some chief scientific advisers had less access to resources, staff and clout within departments and this appears to still be the case. ¹⁹⁸ Sasse T and Haddon C, How Government Can Work with Academia, Institute for Government, 2018, retrieved 26 February 2026, www.instituteforgovernment.org.uk/publication/how-government-can-work-academia  There is also a risk that where chief scientific advisers adopt senior civil service grade responsibilities, their challenge function is diminished.

Regardless of how responsibilities are split in particular departments, senior civil servants running academic engagement schemes should receive adequate resourcing to ensure they can effectively co-ordinate the activity happening within their department.

Evaluating researcher engagement is not easy – but departments can do better

Alongside being clearer about what they are creating engagement mechanisms to achieve, and who is responsible for driving that through, departments can be more intentional in evaluating whether they are achieving their goals. ¹⁹⁹ Oliver K, Hopkins A, Boaz A, Guillot-Wright S and Cairney P, ‘What works to promote research–policy engagement?’, Evidence & Policy, 2022, vol. 18, no. 4, pp. 691–713, retrieved 26 June 2026, https://doi.org/10.1 332/174426421X16420918447616  What evaluation is currently carried out tends to be informal and in departments with sufficient capacity. It is also worth noting that there is a distinction between evaluating how mechanisms are operating and the wider impact of researcher engagement. The former ensures that schemes are run effectively – and there is scope here to collaborate with knowledge mobilisation experts and learn from a growing body of academic literature assessing the effectiveness of different mechanisms in practice. ²⁰⁰ Chaytor S, Parker R and Stevenson O, The Art of the Possible: Catalysts, collaborations and capabilities in academic-policy engagement, CAPE, 2025, www.cape.ac.uk/wp-content/uploads/2025/01/CAPE_ TheArtofthePossible_Final_Interactive.pdf  But as we noted in 2018, to evaluate the success of engagement mechanisms, departments should collect relevant data such as the number of genuine responses to ARIs or secondments. ²⁰¹ Sasse T and Haddon C, How Government Can Work with Academia, Institute for Government, 2018, retrieved 26 February 2026, www.instituteforgovernment.org.uk/publication/how-government-can-work-academia

The types of mechanisms in place today are similar to those flagged in previous Institute for Government research, but in many cases, evidence for best practice is not being implemented across departments. ²⁰² Ibid.  This is a missed opportunity that a greater focus on the evaluation of existing mechanisms could address. As departments iterate engagement routes or set up new ones, they should draw on what has worked well across government. GO-Science has led efforts to share best practice, which has led to more consistent ARI documents across departments, rendering them more accessible to researchers and providing clearer standards about what makes ARIs effective. A dedicated ARI team at GO-Science, the establishment of a cross-government ARI officials network and the publication of guidance have facilitated this. ²⁰³ Oliver K, Boaz A and Cuccato G, ‘Areas of research interest: joining the dots between government and research at last?’, F1000Research, 2026, 11:1509, retrieved 29 June 2026, https://doi.org/10.12688/ f1000research.127542.2  GO- Science should carry on developing guidance and convening networks for researcher engagement across government.

Impact evaluation is more complex. Although departments should be much clearer on what they want from researcher engagement, it is crucial to note that positive impacts from researcher engagement on policy spill out beyond defined outcomes. It is difficult to capture and measure such benefits and these cannot be sacrificed for the sake of a restrictive approach to evaluation that may privilege linear research to policy outcomes. But the importance and value of evaluation in this space are increasingly recognised – GO-Science has been developing internal tools that departments can use to begin monitoring and evaluating the impact of their engagement activities.

There is a tension between orientating researcher engagement structures towards defined outcomes and creating space for benefits to cascade through the system in ways that can be difficult to anticipate. How departments decide to evaluate the researcher engagement system will contribute towards this tension. We argue that an expansive approach to impact evaluation should be prioritised. The Economic and Social Research Council’s (ESRC’s) approach to evaluating impact could provide a model of best practice for capturing the range of outcomes that researcher engagement can produce. The ESRC defines research impact as the contribution that research makes to society and the economy and lists three broad kinds of impact:

• “instrumental impact – influencing the development of policy, practice or services, shaping legislation and changing behaviour 
• conceptual impact – contributing to the understanding of policy issues and reframing debates 
• capacity building through technical and personal skill development”. ²⁰⁴ UK Research and Innovation, ‘Defining impact’, 24 February 2026, (bold added) retrieved 16 June 2026, www. ukri.org/councils/esrc/impact-toolkit-for-economic-and-social-sciences/defining-impact

At a minimum, departments should be collecting data consistently, even if impact metrics or processes have yet to be fully developed, so that they can better understand how researcher engagement has an impact on policy making (within and beyond defined objectives) in a departmental context. Evaluating the effectiveness of researcher engagement will require new tools and staff time. Central engagement co-ordination units in departments, along the lines of the Evidence and Partnerships Hub in the MoJ, may be best placed to lead on this.

To create effective engagement, departments need to institutionalise points of contact – this will require staff time and resources

The work of co-ordinating engagement mechanisms and evaluating them effectively will not happen unless departments dedicate time and resources to it. While external research institutions have tried to make policy engagement less ad hoc by developing roles with relevant expertise (such as knowledge mobilisers), departments do not always have corresponding knowledge mobilisation functions. This makes it more difficult to create deeper interactions between researchers and policy makers. In many cases, there is a strong desire to engage with researchers, but departments lack the capacity to absorb and use all the available evidence.

Departments can do more to actively work with external knowledge mobilisers, by inviting (or seconding) them into the department to build networks that facilitate subsequent engagement activity. But to move beyond the ad-hoc approach that has characterised researcher engagement within government, departments need to embed and institutionalise points of contact. Part of this is about being clear about whose job it is to manage, develop and evaluate engagement mechanisms. Some departments, such as the MoJ, have already made moves to create dedicated structures with clear responsibility for managing researcher engagement.

An alternative approach is DHSC’s use of the now 20-strong policy research units of the NIHR – university-based teams providing both longitudinal research and rapid evidence across specialisms. ²⁰⁸ National Institute for Health and Care Research, ‘NIHR awards over £100 million to new policy research units’, 1 August 2023, retrieved 16 June 2026, www.nihr.ac.uk/news/nihr-awards-over-ps100-million-new-policy-research-units  Beyond publishing, policy research units engage in knowledge exchange through workshops and seminars with DHSC officials, and each unit has a dedicated liaison officer to manage evidence flows between the department and the unit. ²⁰⁹ National Institute for Health and Care Research, ‘Policy research programme – invitation to tender National Institute for Health and Care Research [NIHR] policy research units’, 8 July 2022, retrieved 16 June 2026, www. nihr.ac.uk/policy-research-programme-invitation-tender-national-institute-health-and-care-research-nihr-policy-research-units  To ensure that researcher engagement is properly embedded and sustained, departments should create dedicated knowledge mobilisation functions by setting up evidence units, employing dedicated knowledge mobilisers or by including such responsibilities as part of existing roles.

Government should be more targeted in its researcher engagement efforts, but bringing many voices in remains important

Government is generally understood to be better at bringing in academic expertise, insight and evidence from science, technology, engineering and mathematics (STEM) fields than it is at drawing on knowledge from the social sciences or the arts and humanities. This is because the ‘hard’ sciences enjoy more political clout, and the evidence infrastructures built around them are more mature, more complex and used more consistently.

Hierarchies also seem to persist among social sciences, humanities and arts for people and the economy (SHAPE) fields. Economics and behavioural science have, for instance, become more firmly established as analytical professions in government, compared with other SHAPE disciplines. ²¹⁰ Academy of Social Sciences, Enhancing the Evidence Ecosystem: Improving outcomes for government and citizens, 2025, retrieved 19 June 2026, https://acss.org.uk/publications/enhancing-the-evidence-ecosystem-improving-outcomes-for-government-and-citizens  There is also a continued preference for quantitative over qualitative evidence ²¹¹ MacKillop E and Downe J, ‘What counts as evidence for policy? An analysis of policy actors’ perceptions’, Public Administration Review, 2023, vol. 83, no. 5, pp. 1037–50, retrieved 29 June 2026, https://doi.org/10.1111/ puar.13567  – even when this is not always justified.

One way to combat these disciplinary hierarchies is to carve out more space among civil service leadership for champions of SHAPE research, and ensure that they have the resources and standing to effectively advocate for the use of evidence and insight from non-STEM fields. The increased diversity in chief scientific advisers’ professional backgrounds is a welcome development in this sense.

Even within the same discipline, opportunities for engagement are not spread equitably among academics. There is a risk that government tends to engage predominantly with more senior researchers from more established institutions. There are understandable reasons for this: more established academics are easier to access and have had the time and resources to develop networks in government. But it is in officials’ own interest to cast the net as widely as possible when engaging with academics – to get a comprehensive picture of the existing evidence and ensure they are exposed to a diversity of viewpoints and perspectives. A more streamlined approach to researcher engagement would make it easier for other researchers to get their ‘foot in the door’.

There are examples of best practice in this space. For instance, the Knowledge Exchange Unit in the Parliamentary Office of Science and Technology (POST) has carried out extensive consultation activities with key stakeholders outside government to identify barriers to engagement for researchers from under-represented groups and it has set out various changes to the way it conducts academic engagement to ensure that everyone has an opportunity to take part. ²¹² UK Parliament, ‘Working to support more diverse and inclusive engagement between UK parliament and researchers’, (no date) retrieved 16 June 2026, www.parliament.uk/get-involved/research-impact-at-the-uk-parliament/diverse-and-inclusive-engagement-between-uk-parliament-and-researchers  As in other areas of academic–policy engagement, government departments could do worse than learn from parliamentary practice.

Changing civil service culture requires more opportunities for officials to spend time with researchers

Some aspects of civil service culture continue to inhibit researcher engagement. This is not a challenge unique to the UK – government and academia have long been understood as two communities with different cultures and diverging norms and expectations in and around the work they do. ²¹⁹ Caplan N, ‘The two-communities theory and knowledge utilization’, American Behavioral Scientist, 1979, vol. 22, no. 3, pp. 459–70, retrieved 29 June 2026, https://doi.org/10.1177/000276427902200308  Although there is evidence that sizeable proportions of public sector bureaucracies are more strongly aligned with the research community than might be expected, ²²⁰ Newman J, Cherney A and Head BW, ‘Do policy makers use academic research? Reexamining the “two communities” theory of research utilization’, Public Administration Review, 2016, vol. 76, no. 1, pp. 24–32, retrieved 29 June 2026, https://doi.org/10.1111/puar.12464  more can still be done to overcome remaining cultural barriers between government and academia.

Across the civil service, there is a lingering perception of academic research as being of limited use to policy makers because it is often too theoretical, not focused enough on practical recommendations or unwilling to engage with the constraints and trade-offs of government – or otherwise because there is a ‘skills gap’ among academics in communicating research findings. ²²¹ See Guerrero-Torres L, Ismail A, Savedoff W, Sheikh K and Schleiff M, ‘Enablers and barriers for policymaker engagement in health research from the perspective of policymakers: a scoping review’, BMJ Open, 2025, vol. 15, no. 8, retrieved 29 June 2026, https://bmjopen.bmj.com/content/15/8/e099720; Suazo-Galdames IC, Saracostti M and Chaple-Gil AM, ‘Scientific evidence and public policy: a systematic review of barriers and enablers for evidence-informed decision-making’, Frontiers in Communication, 2025, vol. 10, retrieved 29 June 2026, https://doi.org/10.3389/fcomm.2025.1632305

This perception often compounds other institutional or system-level barriers that inhibit interchange between academia and policy makers, such as differing timescales for research and policy work, and the lack of consistent lines of communication. ²²² Ibid.

This perception is, of course, not completely misplaced and picks up on genuine challenges in knowledge translation. But the intensified focus on knowledge mobilisation within academia over the past decade has meant that researchers have increasingly taken those challenges more seriously. Civil service leaders could, in partnership with research institutions, do more to dismantle continuing misconceptions among officials and build understanding across the research–policy divide.

There are already examples of successful initiatives in this space. Participants in the Capabilities in Academic Policy Engagement (CAPE) project reported greater connectivity across government and academia as a result of taking part, particularly through policy fellowships and knowledge exchange activities. ²²³ Mäkelä P, Boaz A and Oliver K, ‘The Capabilities in Academic Policy Engagement (CAPE) programme in England: a mixed methods evaluation’, Evidence & Policy, 2026, vol. 22, no. 1, pp. 2–22, https://doi.org/10.1332/17442648 Y2025D000000050  This included building or consolidating relationships between policy officials and researchers, and developing new ways to bridge the gaps in different ‘languages’ surrounding evidence. ²²⁴ Ibid.  Across different interventions, participants shared that they had gained a greater understanding of, and built stronger rapport with, professionals on the other side of the divide. ²²⁵ Ibid.  Participants also signalled that there were shifts in culture or attitudes at the individual, team or department levels. ²²⁶ Ibid.

CAPE’s success is one example of how offering civil servants the chance to interact with academics can improve mutual understanding and help bring about a culture that is more open to, and interested in, external expert input. Government can go further to support officials’ exposure to academics and the research community. Knowledge exchange activities such as bringing academics in for workshops, trainings and other events during ‘peace time’, without tying them to immediate policy needs, are valuable in this space, and should be part and parcel of broader departmental engagement strategies. Policy-to-research secondments are also a way for officials to develop a better understanding of the research community. Civil service leaders should provide more cover for officials to develop those kinds of relationships, and recognise that they are likely to have substantial benefits, even if they take time to materialise.

The use of evidence should matter more for professional advancement in the civil service

Criteria for professional advancement within the civil service are also critical to shaping officials’ incentives and behaviour. Engagement with researchers and evidence is not consistently valued for career progression purposes across Whitehall. This is particularly the case for policy officials, whose ability to navigate the machinery of government to deliver at speed on ministers’ priorities is in practice often more highly prized than other skills that are integral to a robust policy making process, such as having subject matter expertise or being able to use evidence effectively. ²²⁷ Sasse T and Thomas A, Better Policy Making, Institute for Government, 2022, retrieved 26 February 2026, www. instituteforgovernment.org.uk/publication/better-policy-making

While, in theory, strong analytical teams can counterbalance this – the analysis function’s career framework centres around the generation and communication of research insights ²²⁸ Government Analysis Function, ‘Career framework’, (no date) retrieved 16 June 2026, https://analysisfunction. civilservice.gov.uk/careers/career-framework  – it is policy professionals who make the ultimate decisions on whose input is sought and integrated into decision making. So the extent to which external expertise, advice and insight inform the policy making process will often depend on whether officials in charge of various policy areas are incentivised, or otherwise inclined, to prioritise it. Policy professionals who know the use of evidence is central to their career trajectories are more likely to bring analysts into decision making earlier in the process and to integrate their insights more deeply into their work.

Steps to create such incentives in the system have already been taken. ‘Data, analysis and scientific advice’ has been included as part of the ‘strategy’ pillar of the 2021 policy profession standards. ²²⁹ Policy Profession, ‘Policy profession standards’, GOV.UK, updated 12 April 2024, retrieved 16 June 2026, www. gov.uk/government/publications/policy-profession-standards  This is welcome, but questions remain about how strongly embedded these standards have become within the system. To make a difference, appraisal against the standards needs to be central to promotion and recruitment practices for policy makers. This should feed into broader thinking about performance management in the civil service, which the Institute for Government has previously argued needs to be strengthened. ²³⁰ Worlidge J, Urban J, Clyne R and Thomas A, 20 Ways to Improve the Civil Service, Institute for Government, 2024, retrieved 29 June 2026, www.instituteforgovernment.org.uk/publication/ways-improve-civil-service

Along with more robust incentive structures around researcher engagement, the civil service should also do more to support learning and development for officials. Developing a better understanding of data and research methods, the ability to critically and constructively engage with research, as well as deeper subject matter expertise in particular policy areas, should become more prominent in future iterations of the civil service curriculum. The overhaul of civil service learning and development through the creation of the National School of Government and Public Services could provide a platform for this shift. ²³¹ Cabinet Office and Jones D, ‘Move fast. Fix things’, speech at What3Words, London, 20 January 2026, retrieved 16 June 2026, www.gov.uk/government/speeches/move-fast-fix-things; see also Institute for Government, ‘Event: Move fast and fix things: Modernising Whitehall to deliver for Britain’, 20 January 2026, retrieved 16 June 2026, www.instituteforgovernment.org.uk/event/modernising-whitehall-deliver-britain

The development of the new school’s training offer should be carried out in a way that brings together work done on improving learning and development in various parts of government, including the policy profession, ²³² Government Skills, ‘Government policy profession’, (no date) retrieved 16 June 2026, https://prospectus. governmentcampus.co.uk/04-professions/government-policy-profession  the analysis function ²³³ Government Analysis Function, ‘Learning’, (no date) retrieved 16 June 2026, https://analysisfunction. civilservice.gov.uk/learning-development  and the Evaluation Task Force. ²³⁴ Evaluation Task Force, ‘ETF evaluation academy 2.0 resources’, GOV.UK, 13 February 2025, retrieved 16 June 2026, www.gov.uk/government/publications/etf-evaluation-academy-20-resources  It is important that this training is offered to – and expected of – not only analysts, but also all officials, especially those working in policy.

Easy access to scientific advice can make a big difference for time-poor civil servants

Civil servants face substantial time constraints – especially as, in recent years, rapid policy making has become the norm, rather than the exception, across Whitehall. ²³⁵ Dzimwasha V, Davison N and Hodgkin R, Rapid Policy Making: How civil servants can make effective policies under pressure, Institute for Government, 2025, retrieved 26 February 2026, www.instituteforgovernment.org.uk/ publication/rapid-policy-making  While there are examples of successful policy being made at pace, especially in emergency settings, the pursuit of speed carries risks. Institute for Government research has shed light on the experiences of policy officials delivering rapid advice to ministers who, among other things, told us they often had little choice but to sideline evidence and data. ²³⁶ Ibid.

There is a strong case against excessively compressed timelines for policy making in non-crisis contexts. But time pressures will likely continue and researcher engagement will often be deprioritised.

When pressed for time, policy makers will be far more likely to turn to evidence and research if access is as easy as possible. For one, government should ensure widespread awareness across the civil service about existing mechanisms and initiatives for sourcing researcher expertise. Although the landscape has expanded substantially in recent years, knowledge about the available routes for engagement remains uneven across the system.

However, some positive steps have already been taken in this direction. The Universities Policy Engagement Network’s (UPEN’s) interactive map of the academic–policy engagement ecosystem is among the most detailed of its kind to date. ²³⁷ Universities Policy Engagement Network, ‘The academic–policy ecosystem’, (no date) retrieved 16 June 2026, https://upen.ac.uk/the-academic-policy-ecosystem  GO-Science:

• maintains a webpage setting out the various routes for academic engagement with central government ²³⁸ Government Office for Science, ‘Routes for academic engagement with government’, GOV.UK, updated 10 March 2026, retrieved 16 June 2026, www.gov.uk/government/publications/routes-for-academic-engagement-with-government/routes-for-academic-engagement-with-government  
• has mapped out scientific advisory committees, councils and arm’s-length bodies ²³⁹ Government Office for Science, ‘Map of scientific advisory committees / councils and arm’s length bodies associated with government departments’, GOV.UK, updated 19 March 2024, retrieved 16 June 2026, www.gov. uk/government/publications/scientific-advisory-committees-code-of-practice/map-of-scientific-advisory-committees-councils-and-arms-length-bodies-associated-with-government-departments  
• has created directories of academia-to-policy ²⁴⁰ Government Office for Science, ‘Directory of exchange schemes from academia to policy’, GOV.UK, updated 19 February 2026, retrieved 16 June 2026, www.gov.uk/government/publications/directory-of-academic-and-policy-exchange-schemes/directory-of-exchange-schemes-from-academia-to-policy  and policy-to-academia ²⁴¹ Government Office for Science, ‘Directory of exchange schemes from policy to academia, GOV.UK, updated 19 February 2026, retrieved 16 June 2026, www.gov.uk/government/publications/directory-of-academic-and-policy-exchange-schemes/directory-of-exchange-schemes-from-policy-to-academia  secondment schemes.

The Scottish government has also brought together the various kinds of engagement opportunities it offers up to academics. ²⁴² Scottish government, ‘Academic engagement: factsheet’, GOV.SCOT, updated 17 June 2025, retrieved 16 June 2026, www.gov.scot/publications/academic-engagement-factsheet

However, many of these initiatives are externally facing. Officials across government also need to have a clear sense of how they can leverage these engagement mechanisms, what the internal process for doing so looks like and when it makes most sense to use them. Forthcoming guidance from GO-Science on academic–policy engagement is a welcome step.

There are also specific tools whose expansion and proliferation would radically improve the timeliness of analysis and evidence provision to policy makers. Rapid evidence synthesis infrastructures, including those powered by AI, could substantially improve evidence take-up over the medium to long term.

More transparency – in data and evidence use – would be in government’s own interest

Parts of Whitehall can be parochial and closed off from many external sources of input into decision making. ²⁴³ Sasse T and Thomas A, Better Policy Making, Institute for Government, 2022, retrieved 26 February 2026, www. instituteforgovernment.org.uk/publication/better-policy-making  Highly pressurised working conditions exacerbate this, often due to a fear of information leaks. ²⁴⁴ Dzimwasha V, Davison N and Hodgkin R, Rapid Policy Making: How civil servants can make effective policies under pressure, Institute for Government, 2025, retrieved 26 February 2026, www.instituteforgovernment.org.uk/ publication/rapid-policy-making  The upshot of this is a civil service that is less transparent than it should be, either because it is not interested in external input or because it is nervous about it, forgoing the benefits of better decision making that transparency often brings about. ²⁴⁵ Savur S, Baker F and Durrant T, The Benefits of Transparency: Why being more open is good for government, Institute for Government, 2024, retrieved 29 June 2026, www.instituteforgovernment.org.uk/publication/ benefits-transparency

For one, the civil service has not always been proactive about collating and publishing data that it holds and collects. This lack of data transparency has meant that government has missed out on opportunities to increase alignment between its priorities and research. While the engagement mechanisms outlined in this report are all key to stimulating the production of policy-relevant evidence, most evidence generation happens outside of direct policy–research collaboration and academics’ use of publicly available data drives it.

One example of best practice in this space is the Department for Education (DfE), which has a history of consistently compiling and publicising complex datasets. For instance, the Longitudinal Education Outcomes (LEO) dataset, ²⁴⁶ Department or Education, ‘Longitudinal education outcomes (LEO) data’, GOV.UK, 17 April 2024, retrieved 16 June 2026, www.gov.uk/government/publications/longitudinal-education-outcomes-leo-dataset/ longitudinal-education-outcomes-leo-data  which brings together data on educational attainment from the National Pupil Database and individual-level income data from HM Revenue and Customs (HMRC), has proven to be an especially rich and insightful data source for researchers interested in the impact of education on labour market outcomes. Research that has been produced using this data has then been fed back into government decision making. ²⁴⁷ Britton J, Dearden L, Waltmann B and van der Erve L, The Impact of Undergraduate Degrees on Lifetime Earnings, Institute for Fiscal Studies, 2020, retrieved 29 June 2026, https://ifs.org.uk/publications/impact-undergraduate-degrees-lifetime-earnings  The latest iteration of DfE ARIs steers researchers towards a variety of search engines, tools and inboxes intended to facilitate access to useful data. ²⁴⁸ Department for Education, ‘Department for Education: areas of research interest’, GOV.UK, 24 April 2025, retrieved 16 June 2026, www.gov.uk/government/publications/department-for-education-areas-of-research-interest

Compiling and publishing government and public sector datasets is not straightforward, and can be a labour-intensive process. But the recent focus on improving data sharing, including through the National Data Library, ²⁴⁹ GOV.UK, ‘Building the platforms to share and use data effectively’, GOV.UK, (no date) retrieved 16 June 2026, https://roadmap-for-modern-digital-government.campaign.gov.uk/digital-and-data-infrastructure/platforms-to-share-and-use-data  could help make this easier. Ultimately, beyond questions of technical expertise and resourcing, civil service leaders should be proactive about creating and maintaining an institutional culture that encourages and empowers officials to work across government to share data and, wherever practical, to make more government data publicly available. ²⁵⁰ Office for National Statistics and Infrastructure and Projects Authority, Data-sharing: The beating heart of a successful public sector, GOV.UK, 2025, retrieved 16 June 2026, www.gov.uk/government/publications/data-sharing-the-beating-heart-of-a-successful-public-sector

Improved evidence use in government could also be bolstered through greater transparency about evidence use. The Institute for Government, Sense about Science and the Alliance for Useful Evidence have previously developed an evidence transparency framework that can be used to assess the extent and depth of evidence use in policy making. ²⁵¹ Sense about Science, ‘Evidence transparency framework’, (no date) retrieved 16 June 2026, https:// senseaboutscience.org/transparency-of-evidence/evidence-transparency-framework; Sense about Science, ‘Transparency of evidence: spot check’, 31 January 2018, retrieved 16 June 2026, https://senseaboutscience. org/activities/transparency-evidence-spot-check  There is considerable variation in how willing different parts of government are to publish the evidence bases, and analytical work, underpinning their policy decisions.

Complete transparency about evidence bases may prove too onerous, given security and classification concerns and the time and resource requirements that it often entails. But there is also a positive case for publishing the analytical work underlying policy decisions. In particular, more transparency can help policy makers iterate policy solutions and strengthen the analytical grounding of future decision making. For instance, the Office for Budget Responsibility’s (OBR) practice of transparency around its economic modelling has meant that it has, over time, succeeded in incorporating constructive scrutiny from academic economists when tweaking its models. This contrasts with the Treasury’s reluctance to share the analytical work behind its estimates, with secrecy around budgets and fiscal decision making largely driving (and compounding) this. ²⁵² Tetlow G and Bartrum O, The Treasury during Covid: What lessons can be learned from the pandemic?, Institute for Government, 2023, pp. 78–80, www.instituteforgovernment.org.uk/publication/treasury-during-covid

Transparency about evidence also has the added benefit that it changes the incentive structures built into the policy making process. The added layer of accountability that comes with publishing the evidence behind government decisions could prompt policy makers to take a more considered, thoughtful and rigorous approach to weighing, and thinking through, the available evidence than they might have done otherwise. While making this a blanket requirement across the system would be too blunt a tool, civil service leaders should lead by example and, wherever possible, practise and advocate for greater transparency around evidence use in policy making.