As we move through the connection phase of pensions dashboards, and with regulatory pressure mounting across the industry, pension schemes face growing scrutiny over their data. This article, as originally featured in Heywood’s Pension Pulse report, explores why data accuracy is critical and what’s at stake if it’s overlooked.
Accurate pension scheme data is the foundation of effective scheme administration and operation. It ensures the correct benefits are paid to members – and that the scheme does not continue to pay members who have died. It allows DB schemes to calculate pension transfer values more quickly. It also ensures that appropriate member communications materials are sent to the correct postal or email addresses.
Inaccurate, poor-quality data can:
Inaccurate data also compromises decisions made by a scheme’s trustees, managers or sponsoring employers, because those decisions will be based on incorrect information.
But in 2025 the consequences of inaccurate data could be even more damaging. Inaccurate data has always created regulatory risks, but the risk of breaches and punishment has increased because poor data could undermine efficacy of the Pensions Dashboards, with connection of the first schemes due to start this spring. In addition, inaccurate data could delay, complicate or even prevent a scheme completing de-risking activities including bulk purchase annuity (BPA) or consolidation transactions.
Pensions dashboards have the potential to drive a huge, positive change in the way people save and plan for retirement by providing members with a secure view of pensions information, all in one place.
Dashboards’ success hinges on the accuracy of schemes and providers member records. In October 2024 a survey conducted by the Pensions and Lifetime Savings Association (PLSA) found that 90% of schemes surveyed were confident they could integrate securely with the pensions dashboards – but 49% cited data quality as a concern.
Dashboards, when live, will likely expose some unpleasant secrets, says Heywood Chief Operating Officer Louise Donohue:
“A lot of the consequences of having poor data are hidden at the moment, but when dashboards come along that data’s going to be exposed in real time to members, all the time.”
That could create big problems for many schemes, because consumers expect online systems to work. If poor data stops the dashboards working properly, consumers will be disappointed and irritated. They may also be alarmed if the system can’t match them to the pensions they think they have, or worse, if it produces possible matches with the wrong pensions. If these scenarios were to play out, schemes and providers could end up spending a lot of time and money dealing with unnecessary enquiries.
Ensuring data is accurate is not a one-off exercise. Member data needs to be right all the time, because it will never be static: we know that when some members move home, or get married or divorced, they often don’t inform pension schemes. This means data cleansing strategies must become part of business-as-usual activities, to ensure that data provided to the dashboards is fit for purpose.
For DB schemes approaching the endgame, accurate data is essential in enabling de-risking exercises, including scheme consolidation, buy-ins and buy-outs.
“Accurate data is critical, because the insurer wants to understand the risk and the liabilities of the scheme,” explains David Rich. “That analysis is complex, but in the end, it rests on knowing who’s dead and who’s alive; and where they live, which is really important for assessing forward-looking liabilities.” It is also essential to ensure the correct benefits are paid. If the scheme provides benefits to a member’s spouse after the member’s death, accurate information about those individuals, particularly their ages, is crucial for assessment of the scheme’s total liabilities, and for paying correct benefits.
Heywood’s Chief Strategy Officer, Chris Connelly, says that gaining an understanding of the true potential costs of inaccurate member data is driving many trustees and managers to ask for help in devising data cleansing strategies.
“Schemes have often struggled to write the business case to fix data,” he says. “But when a derisking transaction may be around the corner they can see a price tag, but they can also immediately see the upside, because the transaction price is often then cheaper.”
In addition, schemes may soon be under more pressure from The Pensions Regulator to accelerate work on improving data accuracy. The Regulator’s General Code and guidance outline existing requirements for reviewing data regularly and including common quality data scores provided by scheme administrators in the scheme returns submitted to the Regulator. These scores are useful in a limited way: they show if there is some data present in data fields and check that it meets some basic requirements, such as National Insurance numbers being in the correct format.
As Connelly warns: “Don’t rely on your Pension Regulator data quality score – it measures the presence of data, not the accuracy of data.”
The Regulator is also now likely to take a tougher approach to schemes that do not appear to be making adequate progress in dashboards preparation. In October 2024 it announced it had contacted some schemes to remind them of expectations set out in the General Code, and to warn that if they cannot demonstrate those expectations are being met, “regulatory action may be taken”.