Code churn
The rate at which code is added, modified, or deleted over a period, particularly code rewritten shortly after it was written, which signals unstable requirements or design thrash. High churn in a file often correlates with defects; tracking it highlights where a team is reworking the same area repeatedly instead of converging on a stable design.
Code churn, how often the same lines are rewritten shortly after being written, is a quiet signal of unclear requirements, exploratory design, or a struggling area of the codebase. High churn around a release is a risk indicator; sustained high churn in one module is a "this needs attention" flag. Like most process metrics it's a question, not a verdict: the right response is to look at why the code keeps moving, not to punish the churn.