I have occsssionally espoused "metric-driven-development" where a metric is defined and measured and then a release is tagged with that metric - and over a peri

I have occsssionally espoused "metric-driven-development" where a metric is defined and measured and then a release is tagged with that metric - and over a period the release is seen as a success or fail if it moves the metric. This keeps everything focused on the important usually externally facing issues.