The Second of the Seven C’s - Coverage
Coverage describes how many records actually have the information that you desire. It can be easily measured by understanding how populated a particular element may be.
For example, suppose you are interested in understanding your 401-k coverage. You can simply count the number of employees that are putting money into the 401-k and divide it by the number of employees who are eligible to put money into the 401-k (or you can divide by the total number of employees - just stay consistent with your approaches). Coverage is typically one of the easiest things to measure, and is often one of the best places to start.
In predictive analytics, I often will use a simple 5% rule. If the field is 5% populated, then it has enough data to be of value. This rule can be especially valuable as you introduce new fields to collect, as it can often take time to gain enough coverage for the data to be valuable. For example, suppose you want to know...
