1. Main points
- Across the public sectors, managers that described their organisation as creating a collaborative working environment felt encouraged to make improvements through innovative methods.
- Managers from across the sectors felt that while their organisations were open to change, innovation was challenging when there was pressure to be productive in day-to-day tasks.
- Managers in more public-facing sectors said they and their staff found it difficult to reserve time specifically for administrative tasks and sometimes needed to work out of hours.
- Managers in central and local government sectors were more likely to use automation and Artificial Intelligence (AI), to be open to taking risks and experimenting with new technology, and to be open to taking on costs to invest.
- Managers in more public-facing sectors tended to be more cautious and uncertain about the use of AI, particularly using virtual assistants as first point of contact with vulnerable groups, such as mental health patients or victims of crime.
The findings in this article are based on qualitative analysis. Therefore, it is not possible to quantify their importance and they cannot be applied to wider population groups. Quotes represent participants’ views only.
2. Overview
This article contains summary findings from qualitative research undertaken by the National Centre for Social Research (NatCen) on behalf of the Office for National Statistics (ONS). As part of our Public Services Productivity Review, we at the ONS conducted a pilot Public Sector Management Practices Survey (PSMPS) in Summer 2024, which collected information from organisations across the public sector about their management practices. Findings from the survey are published in our Public Sector Management Practices Survey pilot, UK: 2023 bulletin.
This qualitative research complements the PSMPS with the aim to explore public sector managers’ views on their organisations’ management practices. The research particularly explores views on the types of administrative tasks carried out and their impact on productivity. It also explores opportunities and barriers to innovation, including the use of automation and artificial intelligence (AI), to improve productivity. This research also complements findings from the further analysis of the public sector time use survey released by the Office for National Statistics on 21 October 2024.
The findings are based on 15 in-depth interviews and four focus groups with managers in the public sector, and five interviews with former civil servant managers who had left the Civil Service within the last two years. Throughout this article, the education, health, fire, and police sectors are referred to as “more public-facing sectors” or those with “frontline” duties.
Back to table of contents3. Views on productivity
Managers offered different views on the meaning of productivity. Often it was highlighted that there was not a single definition or understanding of productivity, but rather several that depended on the different roles and responsibilities of staff and teams, within a setting or organisation.
Some managers advised against framing staff performance in terms of “productivity”. They felt that staff were already working “hard” within a context of staffing shortages, and that too much emphasis on productivity in terms of numbers and targets over quality could put undue pressure on staff and could undermine staff morale, retention and achieving positive outcomes.
It's all about how many you can get in and not really about the quality of the care. It's more about that list is so high and that list needs to come down.
Some managers also felt that human resource (HR) departments and trade unions should be involved in workforce planning to consider the type of workforce and skills needed to improve focus on task prioritisation (related to core job elements) and work-life balance. For example, one manager in central government said productivity was not just about using new technology, but also having a “people” strategy in place:
So, one of the reasons I have a digital director is, it's [their] job to deliver the benefits we've decided we want to realise from technology… and, actually, [my HR manager] then has the problem of wondering how do we redeploy and reskill the colleagues that become free as a result of technology taking their jobs.
4. Management practices
This section focuses on management practices such as decision making, problem solving and approach to innovation when considering ways to improve productivity.
Decision making and problem solving
Managers who have more ability to influence decision making, particularly when supported by data, felt they were better placed to identify issues and take action to improve productivity. However, the extent to which they felt able to make and contribute to decision making varied and was linked to structural and cultural factors. This included the size of the organisation and proximity of staff to senior decision makers, levels of autonomy and whether organisations encouraged and supported staff to provide feedback and collaborate.
Managers in larger and more dispersed organisations (such as NHS trusts and fire service who were accountable to national services but organised on a more local basis) said they lacked shared spaces where they could interact regularly with senior decision makers, making it more difficult to communicate ideas and influence decisions.
Similarly, managers from the health, fire, and police service said that although they had some autonomy in decision making, particularly related to productivity, it was difficult to communicate problems and ideas for change because of the remoteness of senior managers.
Some managers felt empowered to influence decision making and take action to improve productivity where their organisation encouraged and supported staff to provide feedback and work collaboratively. This was more common in central and local government.
We've set a five-year corporate plan of what we're trying to achieve strategically, and then we set, with the support of the board, and in quite a consultative and collaborative way with senior colleagues in the organisation, we set a plan for what are we trying to deliver each year.
Approaches to innovation
Managers described both informal processes built on discussions and reflections on existing practice to make continuous improvements and more formal “transformation-type” projects led by senior leaders involving data-driven performance monitoring and consultation phases. Those adopting the more formal approaches tended to be based in the central and local government and health sectors.
Managers discussed multiple approaches to promoting innovation, including a collaborative culture where staff were encouraged to think about how they could do things more efficiently, and where there were regular opportunities to share their views and lead changes. They discussed carefully piloting (and providing clear rationales for) any proposed innovation to increase organisational commitment, and providing adequate training where any innovation or new technology was introduced.
We have local team meetings. The team, all the time, are like, can we try this, can we try that, even if it's just local stuff… So, we have a senior team meeting every week, so we'd take it to the senior team that the team wish to implement this or try this. It'll always be a discussion and unless someone's got a glaring reason why it's unsafe or illegal.
Yes, [in the Civil Service] they did encourage people to get involved in changes and improvements, if you wanted to work on something like that, then they'd probably be quite happy for you to get involved, and I did do that.
Managers in central government talked about how they encouraged recruitment of ambitious staff to bring new energy, expertise and ideas related to innovation.
We recruit bright [staff] that have used lots of clever computer kit, and we encourage them to try and apply those to the technical problems we're trying to solve.
However, managers also identified barriers that could affect an organisation’s ability to encourage innovation, described under the following subheadings.
Lack of communication
Some managers described little to no regular contact between middle managers and senior leaders.
Resistance to change
This was particularly mentioned in the health sector, where managers identified resistance from staff to do things differently often because they believed that staff did not want to take time out to learn something new.
I think with the resistance [to innovation and technology], particularly [this is at] lower levels, I think a lot of it is, a lot of, they don't want their job to be harder with learning something new or doing it in a new way. It's easier to do what you've always done.
I think in public sector, you do have a bit of a barrier when it comes to introducing new technologies and having to go through either some sort of a procurement process or there simply not being a will to change.
Slow decision making
Some managers in the education sector talked about how drawn-out decision making made it difficult to plan ahead and allocate budgets for setting salaries, and investing in innovations, technology or training.
It's like playing Russian roulette every year… The later we find out [about budget allocations], the more difficult it is for us to plan, and then we're making people redundant, or we haven't taken people on that we could have done, because we were working on a higher percentage. It's really difficult, and when 80% of your budget is on staffing.
Lack of capacity and financial resource
Some managers felt that innovation-focused interventions could take a substantial amount of time to implement and noted that new technology is expensive. Managers in the education, police and fire sectors reported lacking the necessary resource to make and implement purchases.
Well, yes, cost is a concern, but it's more to do with the fact that we've only got a finite budget. I think when you look at the way budgets are broken down, particularly I'm quoting figures that – of the budget that we get as an organisation, I think 90% of our budget goes on staff wages. We haven't got the capacity to spend money freely on developing software solutions in the way that you have bespoke written products. We have to buy products off the shelf.
While many managers felt their organisations were open to change, innovation was challenging when there was pressure to be productive in day-to-day tasks. Mid-level managers in more public-facing sectors expressed the challenges of being innovative when departments set targets that were seen as being often unachievable. For example, because of the demanding nature of policing, managers said it was more important for them to prioritise the basics of their roles and achieve productivity in that respect. They felt that they did not have time to think about how they could improve productivity.
Back to table of contents5. Administration
Managers described a wide range of administrative tasks, a lot of which involved a degree of “paperwork”, such as printing forms, manual data entry and filing. Other administrative tasks mentioned included emails, people management, planning and project management. These tasks were either carried out by dedicated roles or as part of a wider role. These findings align to those from the ONS public sector time use survey where it was found that 60% of public sector workers’ time was spent on 'non-sector specific' activities, which include specialised tasks such as data analysis, research and project management, and tasks such as meetings and events.
There were some variations on administrative tasks by sector. While all managers’ roles included email correspondence, this was more prominent in the central and local government sectors. Administrative tasks within the education sector involved correspondence with parents or students, and within the health sector, this included booking appointments and managing patients.
Impact on productivity
The number of staff dedicated to administration was often described as being limited or when available, they were too centralised (for example, in parts of the health service). Therefore, managers said they often undertook these tasks themselves, which could be time-consuming. These findings align with those published in our ONS public sector time use survey, where frontline workers reported that 47% of their time spent on 'non-sector specific' tasks was perceived as being very important.
Within the central and local government sectors, managers felt they were able to manage administrative tasks as part of their workload. However, managers in sectors that are more public facing found it difficult to allocate time specifically for administration. Some said it took up 10% to 30% of their time, while others felt it was a much higher proportion of their time.
I always try to be very positive. However, it is very difficult with the police because I think the expectation, what they expect from you is far more than you can deliver really. When I think of the admin involved in my role, it's a full-time job in itself.
Feelings of being particularly “overwhelmed” by administration were reported in schools, health, and the police service. Managers in these sectors felt that administrative tasks took them away from other parts of their roles and as a result they sometimes needed to work out of hours, including evenings and weekends.
If I'm perfectly honest, I would say most of my day is administration, and when it comes to strategic thinking, I think in our school, if I have something important to do, I'll take a day or a half-a-day out of school… Or else I find in evenings or at weekends, or during the holidays… All that kind of top-level thinking, you just haven't got the time in school.
The use of delegation to reduce the impact of administration on productivity was not discussed in detail. Some managers did talk about how they would like to delegate tasks, such as placing orders for basic equipment. However, junior managers did not always have the access permissions to undertake these tasks, or in other circumstances, delegation was not possible because staff were simply not available.
As an officer, we're given certain permissions where we're able to make certain changes and do certain things on our intranet system. So often, I will want something done and I'll need some assistance, and I'll say, 'Right, look, can you just go and fill these in for me?' and they don't have the permissions to do it. Then you're left banging your head against a brick wall and doing it yourself in the end.
Opportunities to automate administrative tasks
Some managers said they made regular use of automation to streamline processes to save time and improve communication (for example, sending letters to parents or generating and sending invoices automatically). Other managers discussed the possibilities of automation to reduce the impact of administration on productivity. Although managers thought there was an appetite for more automation, its adoption within organisations and across sectors varied.
Within the central and local government sectors, managers described an approach that was focused on reducing or eliminating processes involving paper forms and manual data entry. Other examples of automation included:
- using electronic forms that could be checked by a machine
- doing pupil registers using iPads
- using speech-to-text applications, instead of typing up information
However, some managers felt that parts of the public sector are “behind the curve” in the adoption of automation.
Managers discussed some of the barriers surrounding automation of administrative tasks. Some managers pointed out that services were still reliant on old, centralised technology that did not integrate well with other systems, which led to duplication of data entry and to services being unfit for purpose at a local level.
I would probably describe us as being 20 to 30 years behind the curve. The favourite term that we seem to use in our in our IT world in services is ‘workaround’. Everything has a 'workaround' of some description because one thing doesn't talk to another. There's no two systems that can talk to each other, it doesn't appear.
Additionally, where there was investment in technology including automation, managers felt that it could be difficult to find the time for staff training. This was especially the case where staff would need to be taken away from frontline duties. Training for anything that's new also can be quite difficult, because if you're a busy team, getting people trained in anything - getting them to do the training they have to do regularly anyway is hard enough.
6. Artificial intelligence (AI) to improve productivity
Existing use and knowledge of AI
Current use of new technologies, automation and AI varied across and within sectors. Some organisations were in the early stages of introducing AI. Informally, managers were often “looking into” using AI. They described small-scale implementation with AI use restricted to particular operations, for example, testing AI for recruitment purposes. More formally, managers described AI pilots, where specific programmes, such as Microsoft CoPilot, were being rolled out with the aim of assisting with administrative tasks, such as minute taking and summarising documents. These formal pilots were common in central and local government, and within academy trusts in the education sector.
The use of generalised AI tools, such as ChatGPT, was often done at an individual level rather than an organisation-wide level. For example, using AI to synthesise meeting notes to assist with report writing. However, this would often be done “offline” and without departmental instruction.
Mid-level managers in frontline positions within health and policing were particularly unsure about the definition and use of AI. Some managers were unable to see how AI would help them and felt that there was no software available that would change their existing databases and platforms.
Our [School Information Management System] (SIMS) is quite clunky, but it's what we have, and you get used to it and make the most of it. School data have got to be sent in a specific form to the local authority. In the past, some schools have moved away from [SIMS] and then moved back again, so there is nothing really to replace [this platform] that's satisfactory.
Additionally, mid-level managers in frontline sectors such as education, health, policing, and fire service, talked about the difficulty of being innovative when departments and individuals have targets to meet that are often perceived as unrealistic. They felt that they did not have time to think about how they could improve productivity through AI.
Managers acknowledged the benefits of using AI in helping to free up staff time, reducing the administrative “burden” and improving the accuracy and speed of reporting. Within the education and fire sectors, the introduction of handheld devices, such as iPads, enabled the live inputting of data, improving reporting accuracy and removing the task of extracting data manually. For managers who had been testing out AI tools, the trialling period was felt to have increased confidence for future implementation.
One challenge of using AI included a lack of time to receive adequate training when new technology was introduced. Additionally, lack of communication and consultation with frontline workers was seen as a challenge. When a new technology was introduced, there was often no consideration about whether the decision would benefit those who use it, and managers felt that this made their working lives harder rather than leading to increased productivity.
On the other end of the scale, some managers had little knowledge of AI and therefore had no current use of AI or automation in their roles.
Managers who were aware of AI but not currently using it understood its potential benefits for future use. There was a sense of inevitability from these participants about the eventual rollout of AI. Some managers in the health sector expressed views that, while they would like to use AI, the technology was not yet ready to be applied within their workplace setting, mostly because of a lack of technical expertise. Often participants had learnt about AI and automation from external sources. For example, some managers reported receiving emails from outside their organisations that had been written by AI.
Differences between sectors were apparent with some managers in more public-facing sectors being more cautious and uncertain about the use of AI and were therefore less likely to be using it already. These sectors had concerns over the safety of their pupils, patients, or members of the public, which led to many managers thinking about regulations and the barriers to AI, instead of how it could be implemented. By contrast, managers in central and local government reported having greater capacity to experiment with new technology.
Potential and future uses of AI
Where managers considered how AI could be used to increase productivity, reasons given mostly centred on a desire to reduce the administrative “burden” of their roles. Managers reflected on how AI could remove “laborious” tasks, freeing up their time. Even those who were more wary of AI were open to technology that could help to automatically produce minutes of meetings.
I know that they [the associate directors within the organisation] have been exploring AI. Particularly, obviously, in relation to automation, there is a lot of – and a lot of it is quick wins in terms of process and systems that are focused on operations. Yes, I think there is a strong focus within the business of, particularly around the processing bit of it, in terms of quality and timeliness about the opportunities that AI and automation can bring.
Managers suggested that the implementation of new technology and AI is dependent on the management structure of their organisation, particularly in the more dispersed organisations, such as the police force or NHS trusts, where decision making takes place at a local level. Some managers from the police and health sectors suggested that the use of AI existed but it was not rolled out at the national level.
Risks and barriers to using new technology and AI
Participants from across all sectors expressed reservations over the implementation of AI and new technology.
One of the main barriers to introducing new technology and AI was cost. Those in central and local government sectors were generally more open to these costs, managers in organisations, such as schools and hospitals, were more apprehensive about the financial commitment. Some managers, particularly those in sectors such as health and policing, questioned why the money needed for AI could not be invested into people instead.
The amount of money that it would take to invest in AI and implement it in hospitals and wards, and the maintenance of it because you'd need to employ somebody to maintain that massive system, why can't it be put into people? Why can't they employ more nurses? Why is it even up for discussion to replace a person with personal care over a computer that could possibly go wrong?
There were also suggestions that fewer senior staff were apprehensive towards change and therefore were not ready to adopt AI and new technologies. While managers in health suggested that they will always need people, they described the process as being a change management issue, where senior staff would need to provide reassurance over job losses. However, other managers did not see staff readiness as a barrier, not because they felt ready for AI, but because management decisions were often implemented regardless of wider staff opinion. Some middle managers, particularly in the police service, felt that senior leaders’ main priority was cost saving and not whether people were positively engaged with an idea.
Other barriers to using technology and AI were around trust in accuracy of outputs and data security. Some managers were against the idea of using AI because they believed that they would still have to check for mistakes and therefore would not be saving any time. For managers in more “high-risk” sectors, such as health, this lack of trust was amplified where they were worried about potential errors that could put patient lives at risk. Managers were concerned over adherence to General Data Protection Regulation and the increased risk of cyber-attacks or data breaches when using an unfamiliar and untested software. This concern was held most strongly within the health, education, police and fire service sectors, where the risk to public data was perceived to be stronger.
People are using AI but then they're checking it and editing it. Why are we using AI? You might as well just write it yourself in the first place. There is no productivity [gain] if you're using an AI package which you're then having to check and edit… I don't get the logic of that.
Some managers felt that automated technology, such as virtual assistants, were not appropriate in some circumstances, for example, when dealing with victims of crime. There were concerns that vulnerable groups might not be served effectively by an automated service. Humans leading first contact were able to pick up on non-verbal cues or “read between the lines” in ways that machines could not, which was considered important by managers in terms of delivering good outcomes.
I definitely think there's a place for technology in health, but sometimes I find going towards this very digitalised world is, I think there's a lot of negatives that we're not necessarily thinking about in terms of communication, human connection. In mental health, loneliness is huge, social isolation is huge. I just sometimes wonder what that world might look like if we become very digital.
These barriers were suggested across all sectors, however, those in education, health, fire and policing sectors described these more seriously than managers from central and local government. They viewed their sectors as high risk and thought that consequences of bad technology would be much higher. Former civil servants also said that the stakes of making poor investment decisions were higher for them in the private sector compared with the public sector, because of the increased accountability over costs.
I'm at a new start-up so obviously, one of the things, it is impeding in terms of full automation or the whole system, how you run the business, is capital. We don't have the huge capital yet.
7. Glossary
Artificial Intelligence (AI)
Computer programs or machines that can learn from data and perform tasks usually completed by humans. Artificial Intelligence (AI) is currently used in a variety of ways, including:
- online product recommendations
- facial recognition
- self-driving vehicles
- medical diagnostic tools
- chatbots that interact in a conversational way and can answer complex questions
Automation
A set of technologies that can substitute routine, non-cognitive tasks or jobs (for example, the introduction of the telephone switchboard replacing switchboard operators, or accounting software).
Back to table of contents8. Data sources and quality
Methods
The National Centre for Social Research (NatCen) carried out this research on behalf of the Office for National Statistics (ONS). NatCen conducted in-depth individual and focus group interviews with senior managers in the public sector. Interviews took place between August and September 2024. They were carried out online, using semi-structured topic guides agreed with the ONS. Interviews lasted around 60 minutes and focus groups lasted around 90 minutes.
Vignettes about possible uses of automation and technology were used in the focus groups to stimulate discussion. All interviews and focus groups were audio recorded with consent and transcribed verbatim. They were then analysed thematically according to the aims of the study (using a top-down approach).
Sampling and recruitment
Public sector managers in executive level management positions (responsible for an organisation or service), or service delivery managers (responsible for a team or department) were recruited for this research.
Participants were recruited in the following ways.
Those who had taken part in the public sector management practices survey (PSMPS) and agreed to take part in further research carried out by third-party organisations. This involved 15 in-depth interviews with:
- six participants from central government (including a paired interview, involving two participants)
- one participant from local government
- five participants from the health sector
- four participants from the education sector
Participants recruited from a recruitment agency. This included four focus groups with managers in the following sectors (number of participants shown in each sector group):
- education – 7 participants
- health – 8 participants
- police – 7 participants
- fire – 6 participants
There were also five in-depth interviews with managers who had left the civil service in the last two years to work in the private sector. This group was included to enable additional insight and to understand whether experiences in management practices differed between the public and private sector.
Back to table of contents10. Cite this article
Office for National Statistics (ONS), released 21 October 2024, ONS website, article, Public sector managers' views on management practices, Great Britain: August to September 2024