Data Quality Metrics – don’t blow up the whale!


Faced with a dead 14 ton whale rotting on their beach a town council came up with what they thought was a brilliant plan. They loaded the carcass with dynamite and blew it up – thinking this would solve the problem. Unfortunately the end result was to spread rotting whale meat over a vast area – creating an overwhelming clean up job that is still carrying on today.

For many data governance organisations faced with setting up metrics it may seem like a good idea to apply every conceivable metric, and apply these metrics to all data. After all, if you can’t measure it you can’t manage it.

However, if metrics aren’t clearly thought out and applied in to meaningful data sets you may unwittingly “blow up the whale” for your data governance initiative by reporting an overwhelming and ultimately irrelevant set of exceptions.

For example, assume that you business requires date of birth to be populated to comply with local legislation. You test for empty “Date of Birth” against the 10 million records in your Client database and return 3 million exceptions – each of which must be addressed by the Client Relations team. However, when you hand them the report you discover that the business has only got 1 million active clients – so at least 2 million of your exceptions will never be acted upon. More importantly, the million records that require action are now hidden in the mess and will probably also not be addressed. This can be extremely demoralising for business – and in many cases will suggest that the problem is either to big or too poorly defined to be addressed.

It is important to understand that  a data quality metric should  be used to manage performance and ultimately to measure improvement of data quality. Make sure you get answers to key questions such as “Who will own this metric?”, “How will we measure success?”, “How will we track improvement?”, “Is this relevant?” and even – “Can this be addressed?”

Setting appropriate metrics is a critical success factor for almost any data governance or data quality initiative and you should consider getting support from appropriately experienced consultants.

Advertisements

One thought on “Data Quality Metrics – don’t blow up the whale!

  1. Pingback: How many data stewards does it take to screw in a light bulb? | Data Quality Matters

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s