I have been thinking for some time about an easy-to-generate metric that seems to be rarely-considered in real-world practice.

UN-PLANNED vs PLANNED WORK

The idea is simply to track the percentage of time that a dev or a team is compelled to work on things that were NOT planned for the current sprint.

This allows you to easily calculate how much of the total time was actually available for planned work.

More often than not I wind up working on teams and projects that pull DEVs into a lot of Production Support work, typically UNPLANNED.

Then folks will go off and try to calculate VELOCITY or something similar, and completely ignore the UNPLANNED work factor -- and even try to increase 'VELOCITY' without any consideration of UNPLANNED.