The Two Types of Data Quality Assurance
Here’s a thought – whether you like it or not, your organisation is implementing data quality assurance but chances are it’s the expensive type.
Nearly 3 years ago now I wrote about how a NHS data quality mistake screwed up my newborn sons unique identifier record and got him mixed up with another kid with the same surname several hundred miles away.
For several weeks after we got appointments to visit clinics and physicians who lived on the other side of the country. We were told that there had been a data quality error discovered and that a team, yes TEAM, of people would resolve the issue.
This is a classic example of how organisations are in fact doing data quality assurance every day. It just happens to be the expensive variety.
What Are Those Staff Doing On An Industrial Scale? Data Quality Assurance
We often perceive data quality assurance initiatives as expensive, lengthy, resource hungry and invasive but just consider how much money is wasted by call centre staff in a telecoms, utilities or retail business as they take call after call of billing errors, incorrect shipping address, invalid product codes and any number of further data related headaches.
It might be labelled as service management or service assurance but a huge proportion of daily activity will relate to assuring that the quality of data is sufficient enough to support the services requested by their customers.
Data Migration – The Data Quality Assurance Wasteland
At the moment I’m creating study guides for Data Quality Pro and the next guide will focus on providing a wide range of Data Quality Best-Practices for Data Migration.
When I saw this Dilbert cartoon I instantly thought of data migration projects.
How many times have we seen our version of “The Boss” above push-back when the words “data quality” are introduced. I must have heard the same words in the cartoon many, many times.
The reality is that even if you ignore data quality assurance during your data migration the company will still need to do it at some point, it will just be infinitely more expensive.
- You fail to assure quality of product naming conventions? Products get stranded in warehouses and major warehouse stock-checks required annually.
- You fail to assure quality when linking customer data during a merger? Call centre staff cope with an increase in billing complaints.
- You fail to correctly populate the engineering data in your outside plant data? Field staff perform multiple truck rolls until the service order is complete.
The reality is that every company is implementing some form of data quality assurance but they can’t visualise the costs because it’s treated as “Business as Usual”. Changing that cultural apathy towards data has to be one of your early data quality management objectives.