Software testing metrics can be very difficult: there are thousands of them out there, it is hard to know which ones are important, many are hard to collect, its often difficult to extract meaning from the metrics, and everyone has an opinion on the usefulness of any metric that you might choose. So, what is one person to do?
This talk presents a methodology for thinking about which metrics are the most important, principles to use for collection and dissemination, and practices to insure that quality gains from your metrics are sustainable. The methodology is illustrated through a case study for building a Customer Quality dashboard – measuring quality on several dimensions. The core methodology for identifying the right metrics is the use the Goal, Question, Metric methodology, where you start off with the quality goals of your organization then design the metrics to support those goals. This method has proven successful with the team, and the wider organization, gaining acceptance of these metrics – and directly tying the metrics to your business goals. The talk concludes with several danger signs for metrics misuse, and show how gains made in quality, by using metrics, is sustained over time.
Video producer: http://www.pnsqc.org/