Values and Assumptions
To help the reader understand where we are coming from, we want to clarify several assumptions, or values, that form the background of everything else in the handbook:
Measurement is good
Engineering productivity can indeed be measured, as long as it is mainly focused on understanding bottlenecks and blockers, instead of measures of output like lines of code.
Engineering metrics should never be part of the performance evaluation process. In particular, trying to measure individual contributors’ performance is not only a futile exercise, but it creates a climate of surveillance that is actively hurting productivity.
While engineering productivity is more of a manager’s concern, we believe that the data should be freely available and discussed transparently, regardless of levels or titles. In our experience, real change only happens when everyone has access to the same information.
Managerial accountability and courage
Managers should be held accountable for building productive teams. In particular, managers should both provide a system to measure productivity and lead the changes that need to happen. In some cases, changes may be complex to implement and require courage to push through the status quo.