How’s it going with testing on the project? To understand this, you need to analyze its effectiveness in terms of product quality and processes.
You can calculate the density of defects, breaks, leaks, the effectiveness of test cases, RC, FDP, DDP, PTC, MTTD, TDE, and dozens of other test metrics. But to determine the profitability of such testing, you need to count the money.
Money and its increasing flow are the main goals of the client in most cases of software development. To make management decisions correctly, a test manager needs to fully understand the cost of testing activities, see development zones, and ways to optimize processes. It is also important for the client to understand what they pay for and why, where they lose, and where they earn. No man is an island, and the task of the so-called architect is to try to make the bees really realize how much money they bring to the client and how much they helped to save. Saved costs are not necessary but they can form a fund for a potential increase in the salaries of the same bees.
Any quality has a price: Cost of Quality = Cost of Poor Quality + Cost of Good Quality. Costs of Poor Quality include internal and external failure costs; Costs of Good Quality – appraisal and prevention costs.
The total cost of testing is quite high but once you estimate the cost of poor testing, it already seems quite affordable.
Warren Buffett once said that price is what you pay and value is what you get. And they do not always coincide. Quality does not mean value. Try to sell a very high-quality typewriter today or convince the client that it will be more valuable for them to release the feature not the day after tomorrow, but in a year, because during this time you will test everything even better and the quality will be higher. This is not going to work. A stitch in time saves nine, and do not forget about time to market.
The challenge is to achieve the optimal price/quality ratio for the client. Why optimal? Because as the cost of finding defects and fixing them increases, the cost of failure will decrease until the optimal point is reached. After that, it becomes economically impractical to further increase testing activities.