Blog 
Spotlight Pension administration teams work to provide members with a better service while keeping up with changing regulations and day-to-day demands. Investment in pension technology is growing, but ageing systems, staffing challenges, and inconsistent data quality can make that job harder.
So what do the figures tell us about the scale and cost of pension administration, and where automation is making a difference? We’ve brought together UK pension administration statistics from regulators, government, and industry bodies to take a closer look. We’ve included the reporting periods and made clear which schemes and parts of the UK each figure covers.
Pension administration statistics at a glance
- 32.8 million memberships were held in occupational defined contribution (DC) and hybrid schemes in 2025, excluding micro schemes with fewer than 12 memberships.
- The Pensions Regulator (TPR) recorded approximately 5,060 private-sector schemes in its 2025 defined benefit (DB) and hybrid pension scheme landscape report.
- The Local Government Pension Scheme (LGPS) spent £226 million on administration in England and Wales for 2024/25.
- 65% of surveyed administrators had increased technology or automation investment over the previous two years, according to TPR’s 2023/24 administrator survey.
- 87% of surveyed master trusts had increased administration technology or automation investment over the previous two years, according to TPR’s 2025 defined contribution research.
- More than 70 million workplace and private pension records were connected to the Pensions Dashboards infrastructure, according to the July 2026 progress report.
How many pension schemes and memberships need administering?
Good pension administration helps members understand what they have and access the support they need. Behind that service are teams managing the records, calculations, and day-to-day details that keep schemes running smoothly.
TPR’s 2025 landscape publications show just how many schemes and memberships depend on that work:
| Scheme category | Number of schemes | Number of memberships |
| Private-sector defined benefit (DB) and hybrid schemes | Around 5,060 | 9.174 million |
| Public service defined benefit schemes and locally administered sections | Around 200 | 19.784 million |
| Occupational defined contribution (DC) and hybrid schemes (excluding micro schemes) | 790 | 32.8 million |
Sources: TPR’s 2025 DB landscape and DC landscape. The DC figures exclude schemes with fewer than 12 memberships. TPR treats locally administered public service sections as separate schemes for reporting purposes.
These figures count memberships, not individual people, as one person can belong to several pension schemes. Both publications also include hybrid schemes, so the rows should not be added together to avoid double counting.
Why are deferred memberships important?
Administration isn’t finished when someone leaves an employer or stops contributing. In the non-micro occupational DC and hybrid schemes covered in TPR’s report, 21.3 million memberships were deferred in 2025, or 65% of the total. Active memberships accounted for around 11.4 million.
Private-sector DB and hybrid schemes also had significant ongoing obligations. 47% of memberships were pensioner memberships, while 46% were deferred.
For administrators, this shows why active contributors aren’t the full story; records and benefits must remain manageable long after a member’s employment ends.
How much does pension administration cost?
One useful benchmark for understanding the cost of pension administration comes from the Local Government Pension Scheme in England and Wales. Its 2025 annual report recorded approximately £226 million in administration expenditure for 2024/25.
| Financial year | Total administration expenditure | Administration cost per member |
| 2022/23 | £178 million | £27.41 |
| 2023/24 | £195.7 million | £29.28 |
| 2024/25 | £226.03 million | £33.18 |
Source: LGPS Scheme Annual Report 2025. Figures cover England and Wales and include all membership categories in the per-member calculation.
From the published figures, administration costs per member rose from £29.28 in 2023/24 to £33.18 in 2024/25, while total administration expenditure increased from £195.7 million to approximately £226 million.
The Scheme Advisory Board links rising costs to pressures including McCloud remedy implementation, Pensions Dashboards Programme preparation, digital service expectations, cyber security, and recruitment and retention.
These are LGPS benchmarks, rather than average costs across UK pension schemes, so they should not be treated as an industry-wide administration rate.
Do larger pension schemes have lower costs?
Larger pension schemes can benefit from lower running costs per membership. TPR’s May 2026 analysis of 29 active master trusts found that total costs per membership generally fell as scheme size increased. Smaller schemes also showed greater variation in costs, with the smallest recording higher median costs.
These findings point to a practical benefit of scale: supporting more memberships does not necessarily mean costs have to rise at the same rate. Larger schemes may be able to negotiate lower service provider fees and operate more efficiently, creating opportunities to deliver better value for members.
The figures cover administration, investment, and operational costs rather than administration alone, so they are best viewed as an indication of the wider economies of scale available to larger schemes.
Are administration costs the same as pension scheme charges?
What members pay and what schemes spend on administration are different things. An annual management charge can cover both investment management and administration, so it does not tell us the cost of administration alone.
The Department for Work & Pensions’ Pension Provider Survey 2024/25 puts some figures behind those charges. For providers’ largest default funds, the median annual management charge was 0.24% for single-employer trusts and 0.32% for multi-employer providers, including master trusts and group personal pension providers.
The figures from this annual survey came from 30 single-employer trusts and 10 multi-employer providers. They offer a useful snapshot of reported charges, rather than a standard industry rate.
For trustees comparing costs, a useful starting point is to check what each figure includes. Keeping member charges separate from administration spending makes it easier to compare like for like and understand where the money goes.
How widely are pension schemes investing in automation?
Pension automation can take some of the manual work out of running a scheme. Here, we mean using pension software to handle administration tasks, rather than automatic enrolment into a workplace pension.
TPR’s 2025 research into DB and DC schemes shows that larger schemes were more likely to have increased their investment in administration technology or automation over the previous two years:
| Scheme group | Schemes that increased investment in technology or automation (%) |
| DB schemes (overall) | 36 |
| Large DB schemes | 64 |
| Large DC schemes (excluding separate master trusts) | 57 |
| Master trusts | 87 |
Sources: TPR’s 2025 DB and DC scheme surveys. “Large” means at least 1,000 members. The master trust figure is based on a subgroup of 15 schemes.
These figures measure increases in investment, rather than the proportion of administration tasks being automated. A scheme that maintained its budget may already have established systems, while increased spending may involve upgrading existing technology rather than introducing automation.
What benefits have administrators reported?
In TPR’s 2023/24 survey of 169 administrators, 65% had increased technology or automation investment over the previous two years. The 110 administrators who increased investment reported the following outcomes:
- 85% reported better service for members
- 67% experienced efficiency gains and cost savings
- 55% benefited from higher member engagement
- 42% had fewer errors or complaints
These were self-reported outcomes, not audited savings. In particular, 67% reporting efficiency gains and cost savings does not mean that costs fell by 67%.
For schemes assessing their own investment, useful measures could include calculation turnaround times, repeat enquiries, and cases requiring manual correction. Establishing a baseline makes it easier to judge whether a new system improves the work that matters.
What can make pension technology harder to implement?
Only 37% of administrators in the 2023/24 survey had a documented IT or technology strategy.
TPR’s September 2025 market oversight report provides more recent context. Following engagement with 15 administrators, it identified legacy systems as a continuing barrier to automation and scalability. Updating those systems could also require substantial time, money, and staff capacity.
The report described emerging uses of AI, including chatbots and predictive tools, but found that governance arrangements varied between administrators. Some had established structured oversight, while others lacked clear governance.
For schemes, the starting point should be the administration problem they want to solve, rather than the technology itself. For example, a scheme seeking faster retirement quotations should establish where delays occur, which data the calculation requires, and how exceptions will be handled. That creates a clearer basis for assessing a system rather than a general ambition to use the latest automation technology.
How much attention are schemes giving to data quality?
Technology is only as useful as the information provided. That means investment in software needs to go hand in hand with investment in data quality. TPR’s 2025 DB research found that 60% of schemes had increased spending on managing or improving data over the previous two years, rising to 83% of large schemes. In its separate DC research, 71% of large schemes had increased their data management or improvement budgets over the same period.
However, increased spending alone does not guarantee better data quality. TPR’s data quality initiative, running from October 2024 to September 2025, included 847 targeted letters asking schemes to take action and 211 requests for further information. These were targeted interventions, not a representative estimate of how many schemes across the industry had poor data.
The regulator found inconsistencies in testing and scoring. It emphasised that checking whether a field contains information is not enough: schemes also need to assess whether that information is accurate.
For example, a date of birth can be present but incorrect, and a record can look complete while still missing information needed for a particular benefit calculation. The more useful question is not whether data exists, but whether administrators can rely on it for the task at hand.
How far has the Pensions Dashboards Programme progressed?
Pensions Dashboards are intended to let people see pension information securely in one place. The July 2026 progress report recorded more than 70 million connected workplace and private pension records, representing approximately 85% of records in scope. This was alongside tens of millions of State Pension records. These are pension records, rather than unique individuals.
Is the connection deadline the same as the public launch?
The connection deadline and public launch are different. The connection deadline for schemes and providers in scope is 31 October 2026. The July 2026 report said that, on current plans, the MoneyHelper Pensions Dashboards was expected to become publicly available during the 2027/28 financial year.
Connecting pension records and launching a service to the public are different milestones. Schemes should not treat the connection deadline as confirmation of a public launch date.
What does this mean for administration teams?
TPR’s April 2026 Pensions Dashboards oversight report highlighted how concentrated the occupational pensions market is: 51 schemes held around 80% of occupational pension records in scope for Pensions Dashboards. Its engagement found that preparations for matching users to records were generally more mature than preparations for providing pension values.
This makes operational readiness just as important as technical connection. Pension administrators need to maintain reliable matching data and pension values after connecting. They should also prepare for increased member enquiries as dashboards encourage people to investigate pensions they may not have reviewed for years.
TPR’s July 2026 guidance specifically encouraged schemes to consider workforce planning, training, and the support members may need.
How prepared are schemes for cyber incidents?
TPR’s 2025 DB survey found that 95% of schemes were covered by a cyber security incident response plan. However, only 31% said that plan had been tested within the previous 12 months.
This is important because a response plan must work in practice, not just exist on paper. For trustees and administrators, regular testing can help uncover weaknesses before an incident happens, from unclear responsibilities to gaps in service continuity and member communications.
It is also worth noting that the remaining schemes were not necessarily untested, as some respondents did not know when testing had last taken place.
Make pension administration work better for everyone
Good pension administration should work for the people running the scheme as well as the members relying on it. That means reducing unnecessary manual work, keeping information accurate and connected, and making it simpler for members to access the information they need.
Mantle brings all aspects of pension administration into one connected platform. Schemes can spend less time managing disconnected systems and more time delivering an effective service to members.
Want to see how Mantle could support your scheme? Start a conversation with our team.
