We have optimised the actual test process. It’s now responsible for catching 70% of defects. But despite this 10% of defects still get into live. We need to do better. We need to look at the subtle patterns, to find areas that can be optimised to reduce the number of defects reaching test phases altogether. AI or Cognitive Testing is one way that this can be done.
Optimisation is not just about the defects or the code, but the way people are perceiving the system. Defect reports and the choice of language reflect what users and testers feel. The sentiment shifts over time, by measuring that we move towards making the SDLC a three-dimensional space, rather than two dimensions, as we take code and defects and then add in the emotional reactions buried in the interactions between the developers, the testers and the users.
This talk will touch on this aspect and some of the wider learnings we have learned so far in building out systems capable of analysing many different and disparate data sources, to find patterns and identify actions that can really help you to optimise the SDLC earlier in the cycle. It’s a talk from someone who has already done the walk.