The Pensions Regulator’s latest review of scheme data quality, published at the end of last year, sets out a challenge - one that many schemes are still underestimating. While dashboards may be the spark for current data activity, the fire needs to burn brighter and longer. The regulator is no longer politely suggesting improvement - it’s expecting demonstrable action.

Trustees have been making efforts to clean up the 'match' data needed to identify savers on dashboards - names, dates of birth, National Insurance numbers and the like. That’s a solid start. But few have seriously tackled the ‘value’ data - the actual pension figures members will see.

Without good value data, dashboards risk disappointing users and undermining trust. TPR is clear - getting members onto dashboards is one thing, but if schemes can’t show them anything meaningful, what's the point?

Data quality is only as good as the method used to test it - and this is where many schemes continue to fall down.

Some still calculate their data scores based on the number of fields failed, rather than the number of members impacted. Others submit only partial data or combine common and scheme-specific items into a single score. This creates a misleading picture and undermines the value of the exercise.

PASA’s Data Scoring Guidance, published in November 2024, reinforces that scoring must reflect data quality from the member’s point of view - if a single field is wrong or missing for a member, that’s a failed record.

The guidance goes further: trustees should test all relevant data at least annually, and schemes should resist the temptation to exclude difficult or known-problematic items. Hiding the truth helps no one.

While common data is broadly well understood, scheme-specific data remains a grey area. PASA highlights the need to move beyond boilerplate checks and assess data that reflects the scheme’s unique structure - including salary history, tranching, special terms, GMPs, AVCs, and historic transfer-ins.

What’s more, trustees shouldn’t wait until a project like a buy-in or GMP rectification kicks off. They should anticipate future needs - whether that’s dashboards, strategic transactions or compliance demands - and factor those into data testing and cleanse activity now.

One of the most consistent weaknesses identified in both TPR’s findings and PASA’s guidance is trustee detachment. Many are still overly reliant on administrators to define, test and report on data - often without proper scrutiny or challenge.

This has to change. Trustees don’t need to become data scientists, but they do need to understand the scoring methodology, question the rationale behind exclusions, and actively monitor progress.

PASA advises that trustees and administrators should work together to ensure data scoring is meaningful, consistent and repeatable. Where scoring is high, trustees should interrogate whether that reflects genuinely strong data - or just weak testing.

Clean data isn’t just about dashboards - it’s the foundation for every scheme decision and member experience.

It helps avoid costly mistakes, supports accurate benefits, enables technology adoption, and underpins risk-reduction exercises like buy-ins and consolidations. It’s not a one-off project - it’s a strategic necessity.

The direction of travel is clear. TPR and PASA both want to see data treated as a long-term governance priority - not an annual compliance chore.

If you can’t confidently say what your data score means, how it was calculated, and what’s being done to improve it, you're not ready for dashboards. More importantly, you’re not ready for the future.

Data is no longer an admin issue - it’s a trustee responsibility.

Claire Montgomery, Client Projects & Data Manager – Trafalgar House