It Ain’t Necessarily So by Nicola Askham

We often read insightful articles written by journalists who have based their article on data provided by a particular respected industry association. A lot of people make decisions based on such articles, but have you ever thought about how reliable that data might be? Just because the data comes from say the British Bankers Association, the British Retail Consortium etc. should we necessarily believe it?

By definition any report issued by a trade association must be based on the combined data of their members, but do you know how good all or some of that data is? Does it contain data from all their members or just a sample?

Data Management as a discipline is still relatively immature so how do you know that the data the report is based on is any good, or that it has combined the same things or compared like with like, to come up with an aggregated view?

So, what’s the solution to this predicament?

In Insurance Regulation the Solvency II Directive recognises this issue and makes the requirement that external data must be treated as though you had created that data yourself. This requires knowing how good that data is and how it has been controlled before you received it. On a company basis you can look at including contract terms regarding the quality of data which you purchase and maybe even include a right to audit that data periodically. But that doesn’t help us when we are reading our daily newspapers.

I think that we are a long way from such reports coming with some kind of assurance of quality of the data both contained within the report and the sources of that data. I’m not sure that it will ever happen, so can you do anything to increase your confidence in the data? Well to be honest, I think you are limited in your options, but if you want to make an important decision based on such reports my advice would be as follows:

  • Check for other sources of the same or similar data to compare.
  • Look at previous years reports and see if they have changed as you would have expected.
  • Look for changes in trends and look for independent explanations of those changes. For example if say growth in a certain sector has been 5% year on year and then suddenly is 12% this year, are there reported reasons for such an increase?

So, while I’m not saying that you should distrust every report, a healthy dose of cynicism would not go amiss. Remember what Harold McMillan is reported as saying: “ I read a great number of press reports and find comfort in the fact that they are nearly always conflicting”.


Nicola is an independent data management consultant, whose primary area of expertise is data governance. Nicola has experience in designing and implementing full data governance organisations in both regulatory and non-regulatory environments. She has worked in Data Management for ten years, initially for a leading UK Bank, before becoming a consultant at the beginning of 2009. Most recently Nicola has spent most of her time delivering data governance for Solvency II and also offers data governance coaching.

Nicola presents at Data Management Conferences, is a Director and Committee Member of DAMA UK and is on the Expert Panel of Dataqualitypro.com.

KDR Recruitment is the home of the best data, technology and analytics jobs. For more data news and views follow KDR on LinkedIn.

You might also like: