Don’t Know the DNA of Your Data Yet? Use a DQ Scorecard

Industry experts say that, “What is not measured is not managed”, and invariably what is not managed makes it difficult to maintain and improve. Data quality should be considered as a measurable metric with quantifiable scores and well defined data sets. Devoid of assessable tools, it is not possible to ensure data quality and assess the required business impact.

The Data Quality (DQ) Scorecard is a good way of identifying your data DNA. The scorecard helps in keeping check on data quality dimensions such as accuracy, completeness, consistency and compliance. In order to develop an insightful DQ Scorecard the following dimensions need to be considered:

Data Quality Dimensions 
  • Available and Complete – To ensure that available data is complete and missing data is added. 
  • Accurate and Recent – To establish whether data is updated and correct. 
  • Consistent – To ascertain if data is related to other elements in the data set and coherent. 
  • Compliance with Standards – to verify whether data complies with industry standards.

Data Quality Metrics “Should Have’s”
  • Definition –metrics should have well defined target of business users and data quality rules.
  • Relevant –metrics should define how it improves performance and have a business context.
  • Measurable - scores should be quantifiable and measurable within a particular range.
  • Controllable – metrics should define a controllable aspect of business processes. 
  • Traceable – there should be a ‘time series’ to trace and track the results in order to measure improvements and provide insights. 
  • Data currency – currency is defined as the extent to which data is updated in the real world context. Data currency is identifying the frequency at which data needs to be refreshed, or the ‘freshness quotient’.

To identify areas of improvement and maintain quality of data, well defined data quality dimensions and metrics should be developed. An important aspect to keep in mind is that quantifying and measuring data quality elements without qualified relevance (what is known as ‘so what’ metrics) is insignificant in developing an insightful DQ scorecard. 

Therefore, through an organized and well defined approach, appropriate assessments and relevant improvements can be created, to build a strong foundation for businesses, which is Quality Data!

Measure not because you can, do it to manage data relevantly
Analyze not because you should, do it to gain business value
Track not because you need to, do it to know what you can do better

No comments:

Post a Comment