The 9 quadrant model for code reviews

Abstract

Discusses a decision-making model which can be used to determine the efficiency of a code review process. This model is based on statistical techniques such as control charts. The model has nine quadrants, each of which depicts a range of values of the cost and yield of a code review. The efficiency of the code review in detecting defects is determined by taking inputs from past data, in terms of the costs and yields of those code reviews. This estimate also provides an in-process decision-making tool. Other tools can be used effectively, in conjunction with this model, to plan for code reviews and to forecast the number of defects that could be expected in the reviews. This model can be successfully used to decide what the next step of the operational process should be. The decisions taken using this model help to reduce the number of defects present in the software delivered to the customer.