What if you could find hidden patterns in the data your development teams create? What if you could discover ways to improve your team’s performance?  To answer questions like these, we are introducing technical preview using Watson Analytics with Collaborative Lifecycle Management (CLM) 6.0.2.

To see this new integration in action, go to https://jazz.net/rs and run a report.  Once the preview is displayed, use the Export menu to “Export to Watson Analytics”.  Before using this feature, you will first need to sign up for Watson Analytics. It’s free to try it out with small amounts of data.

So how do you use Watson Analytics with CLM? First, let me say that the insights you can gain from analytics improves with the more relationships you have between the artifacts you want to analyze. CLM provides a very strong solution for facilitating analytics on your lifecycle data because of its strong support for traceability and reporting on cross domain artifacts.  For example, when our team runs a test tracked in Rational Quality Manager, and it fails, we create a defect in Rational Team Concert and link the two artifacts.  By using the Jazz Reporting Service (which comes with CLM), we are able to quickly generate a report identifying all test cases and any associated work items.  This sounds simple enough, but when we bring this report into Watson Analytics we are able to analyze this data to explore questions like:

• Which tests are finding defects?
• Which tests have never found defects?
• How frequently are these tests being run?
• What is the breakdown of severity of defects produced by a test or a test plan?
• Which tests are primarily finding defects which provide no value (e.g. only resolved as invalid, works as designed, etc.)?

We are asking these questions to try and optimize our own testing strategy.  If you would like to know more about this kind of analysis and the outcomes, please leave me a comment on this blog entry.

There are so many possible ways we can utilize analytics on our lifecycle data, and we are looking to showcase more.  I encourage you to try it yourself with data from jazz.net or your own CLM data. Please read my article here for details on how to get started: Gaining insights into CLM Data with Watson Analytics and the Jazz Reporting Service.  This article explains how to use a spreadsheet to get data into Watson Analytics using any version of CLM.

Adam R Neal
CLM Analytics Architect
nealadam@ca.ibm.com