Targeting Errors of Omission

Years ago, a manager asked me to be faster with testing. “Your colleague is working through many tests a day, and you’re not”. Comparing notes, we soon learned my colleague tested for big visible error messages whereas my theory of error – and results – was more versatile.

In years since, I have been improving my theories of error and understanding of risk, and come to notice a specific class of errors many colleagues in testing struggle with: errors of omission. While noticing something is there, like a big visible error message, is something people are good at, we need support in being good at seeing things that should be there but are not. While regression testing is a consideration of something going missing that used to be there, the most devious errors of omission are the things that are missing that should be there.

In this talk, we look through examples of omission, and tactics to improve our ability to see beyond what is there. In software development, we transform ideas to code, and our code is only as good as our ideas.