"Without good quality data, the industry cannot seek to improve accuracy and efficiency in any area of work carried out," says Spence & Partners' Mark Johnson.

With pension liberation fraud, auto enrolment and the defined contribution (DC) code and guidance, data appears to have been put on the back burner - for the time being at least.

In a press release in December 2013, The Pensions Regulator (TPR) said it intends to review its record-keeping guidance during 2014, but it is still unclear what this will bring.

Without good quality data the industry cannot seek to improve accuracy and efficiency in any area of work carried out, be it a retirement settlement or a triennial actuarial valuation. Data is, and always has been, the foundation that the industry is built on; if this is suspect then how can we expect it to hold up to the test of time?

Looking back, I often ask myself how things got so bad. It isn't just one factor but a combination of different factors that have compounded the issues over a number of years. Some good examples are:

- Computerisation of member records. Historically data wasn't held electronically e.g. record cards and/or member files. Sadly, for some schemes this is still the case today. How was this exercise carried out, and more importantly, how do we know that it was done correctly?

- Schemes transferring between multiple Third Party Administrators (TPAs) and data being lost /misinterpreted in translation. Schemes have often transferred on more than one occasion. What happens if every time data is lost and/or misinterpreted?

- Poor governance at a processing level. Updates to the administration system can sometimes be a final thought and perhaps forgotten. The introduction of administration standards e.g. AAF 01/06 helped in some respect but there are years of neglect to account for. Can we be certain that all manual data updates were peer reviewed?

Data has never been interesting and when TPR issued guidance back in 2010 it brought an air of optimism to the pensions industry - data was finally being put on to trustee agendas.

It is now over three years since the guidance was introduced and what has this brought? Common data was an excellent starting point and with TPR issuing strict guidelines this meant that action was taken by the majority of schemes. So we now know where our members live, what next?

My hope was that TPR would expand on its guidance and turn the spotlight onto conditional data; however in this recent press release they stated that no targets are being set for conditional data. I fully understand that it is difficult to assign prescribed targets but I believe that TPR needs to set boundaries on expectations, even if it is just to ensure that data is tested. Trustees should be taking conditional data very seriously, conducting audits and closely analysing the results of these. More importantly, where problems are arising, data management plans need to be implemented to ensure the constant improvement of scheme data quality over time.

Written by Mark Johnson, business analyst, Spence & Partners

Mark_Johnson@spenceandpartners.co.uk