DS-301f · Module 1

Establishing the Quality Baseline

3 min read

You cannot improve what you have not measured. The quality baseline is the initial measurement of all six dimensions across every critical data source. The process: identify the twenty data elements that drive the most important business decisions. For each element, measure completeness, accuracy, consistency, timeliness, validity, and uniqueness. Record the scores. These scores are the baseline against which all future improvements are measured. The baseline will be alarming. In most organizations, the first baseline reveals accuracy rates in the 70-80% range for key fields. That means 20-30% of the data driving decisions is wrong. The alarm is appropriate. It is also the catalyst for investment.

Do This

  • Measure the baseline for the twenty most decision-critical data elements first
  • Record each dimension separately — the specific scores reveal specific problems
  • Share the baseline with leadership as the justification for data quality investment

Avoid This

  • Try to measure every data element at once — focus on the twenty that drive the most decisions
  • Suppress the baseline because the numbers are embarrassing — the embarrassment is the catalyst for change
  • Measure the baseline once and never re-measure — the baseline is the starting point, not the endpoint