QA Challenge #6: Missing Measurements
Posted on Tue, Sep 13, 2011 @ 01:56 AM
Well, you might have heard this one before: If you can measure it, you can improve it. Yet, when it comes to implementing this rule to improve business results many people and organizations struggle.
My colleague, NK (http://www.linkedin.com/in/nkshrivastava), gives a great example: Typically, we measure our weight, BMI, blood sugar, etc before we actually start working on a “health improvement” program. But applying this rule to our business is often challenging. Why? Because we fail to understand what are the key measurements, where to begin, whether we have time, and much more.
In my experience, Quality Assurance metrics are the best way to improve not only the testing process or the QA team’s performance but also the organizational performance. If quality and improvement is part of your organizational culture, here are some of the standard quality metrics that I have found to be relatively easy to measure and implement:
1. Total Defects By Stage Found (during requirements, design, development, testing, after deployment) – the earlier the better so you know what area needs improvement
2. Defect Detection Effectiveness (DDE) – This is the one that can be measured at any phase. So let’s say in QA, DDE-QA = Defects found by QA/ Defects found post QA. It’s similar to the one above but it’s relative and could be one of the productivity measures
3. Defect Backlog by Month – measuring how many defects you verify and receive in a month can help you know your capacity thus helping to make decisions on your estimates or priority
4. Repair to Build Ratio – Time spent to fix and re-test defects as compared to time spent to develop the solution or product. This can help you know your unit’s effectiveness as well as provide accurate estimates
Though metrics are important and, in fact, a key ingredient for success, many process improvement initiatives still fail (just like many weight management programs fail). Here’s a good article which explains it well. http://online.wsj.com/article/SB10001424052748703298004574457471313938130.html