The terms metric and measure are often used interchangeably. Ever wonder which term to use or what the difference is?
The proper use of terminology matters and can be impactful depending on context. One area the difference between a measure and a metric manifests itself is with scorecards. Scorecards help your group and track metric values, metric value changes over time and key dependencies between metrics.
Like anything, there are nuances but here is a brief and easy way to think about measures vs metrics.
Measures are simply numerical values. They are the quantitative values or facts in star schema parlance. Think of quantities, costs, amounts. They can be additive or non-additive depending on if they logically aggregate or not.
Metrics are similar to measures in that they are numeric values, but metrics have additional characteristics that provide greater contextual information. In addition to a quantitative value, metrics have:
1. Target Values. In order for a metric to be meaningful, it needs to have an expected or desired value that provides the context to indicate how you are performing compared to your goal. For example, if you are driving your car and the speedometer reads 60 mph (measure), you don’t know if that is too fast or too slow unless you compare your speed to the posted speed limit. Once you have that context, you can take the appropriate action, in this case, either increase or reduce your speed accordingly.
2. Thresholds. Thresholds define the range around the target value where the value is acceptable or does not require action. Following our car driving example, if you are plus/minus 5 mph within the posted speed limit, you probably are within tolerance and most likely will not receive a speeding fine.
3. Trend. When looking at metrics, it is often useful to see how the value is trending over time. For instance, is your metric value static, or does it frequently change over time? Have the actions you have taken resulted in your poor performers getting better? Are there patterns that you notice and can you figure out why they are occurring?
4. Relationships. One of the most powerful attributes of metrics is their relationships with each other. Defining and understanding how metrics relate is a powerful tool in performing impact analysis and using leading indicators to be more forward-looking. It’s one thing to know that a metric is showing poor performance according to your set goal. Still, it’s even more valuable to see the upstream and downstream relationships to understand why something is happening or give you insight into what can expect to happen if you don’t intervene.
5. Owner. A metric needs to have an owner. This does not necessarily mean the owner is responsible for the performance of the metric. Still, they are in charge of the metric definition, metadata and they serve as the organizational point of contact for questions about that metric.
Suggested content for you
In the federal government, the push to embrace data driven decision-making has gained serious momentum in recent years. From a people perspective,...
My career as a Data Architect began exactly the way you would expect. I first earned both Bachelor's and Master's degrees in History. All joking...
Transforming the way you leverage your data presents government executives with a range of tradeoffs, complications, and tough decisions. The road...