imageMost businesses are trying to ensure that they have good data, or as they might say in Yorkshire “Reet good data” (hence the flat cap image).

Before expending lots of time (and money) polishing and honing your data to perfection, have you assessed whether you are improving the right data?

Over time business requirements change and what was once critical data for certain business activities, may no longer be critical – there may be other sources for this data, or the business process using it may no longer exist.

Similarly, new requirements for data may have arisen which do not have the required quality or quantity of data available, so this data will need to be obtained.

One organisation I have been working with reviewed the attributes that they had traditionally been gathering and managing for a particular entity and found that they were requesting 160 attributes from data providers. When an objective review of the need for these attributes was undertaken, it was found that only around 60 actually had a valid business need!

Another organisation went through a similar exercise assessing all the attributes required to support their many business processes. They had expected that many hundreds of attributes would be required, however, they found that in fact only five specific groups of attributes (e.g. dimensions) were actually required.

So, are you improving the right data?

Tagged on:     

4 thoughts on “Right good data or The right good data?

  • 6th May 2011 at 06:05

    Excellent points.

    The simple rule that makes it easy to ascertain what data is required is, “The only data that an enterprise should hold is that required to support the Business Functions of the enterprise – nothing more, nothing”.

    It is critical to differentiate here between Business Functions and Processes. The former are the core activities of the enterprise, the latter the sequence in which these activities need to be performed in order to arrive at a particular business outcome.

    Functions (not Processes) create, use and transform all enterprise data. By doing a CRUD matrix (Create, Read, Update, Delete) for each Business Function, an enterprise will know exactly what data is required to be held and can archive everything else.

    When “polishing up” their data, enterprises ought also be aware of The Pitfalls of Data Re-Use


    • 6th May 2011 at 08:29


      Thanks for the excellent comments, as you say it is essential for an organisation wide (or at least directorate wide) functional review to ensure that you are only managing required data.

      Sometimes I have seen practitioners busy ‘polishing’ existing data and claiming great success/progress etc. as this is easier to achieve than taking the wider view.

      Senior managers and executives should ensure that their teams have analysed the current business data requirements before engaging in other data improvement activities.


  • Pingback: George Mallory and Data Mountaineering | The Data Roundtable

  • 2nd March 2012 at 11:29

    Great post. We need to know weather we have the right data or we can also have a data quality analysis for our business. Thought I’d share!


Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.