Software Test & Performance Collaborative

Mind the Gap

By Matthew Heusser on October 26, 2009 | 1 comments.

(was: Metrics and Process and Templates, oh my!)

I’m back from STPCon. On the last night, I had dinner with some the attendees, and, as much as I like to talk, I tried to at least listen a little. What I heard struck me.

First I heard from a tester. She said that her main goal for the conference was to get more test ideas – examples, “how I would test this”, test exercises combined with direct coaching. In the past, I’ve noticed a lack this at test conferences, but was struck to hear it from a conference that I associate with on a very personal level.

Yes, we had some exercises and examples (Testing Outside The Bach’s comes to mind), and yes, we could have more. I heard you, Bobbie, and I’m going to work on this.

The second person I met as a director of test, and he pointed out something interesting. He said that yes, deep down, we all know that metrics and templates and process are horse-pucky. Yet senior management – people who have a cursory understanding of what we do, keep asking for them. So, as a test executive, he picks the battles he can win, and tries hard to build relationships.

According to this director, who I will call Rob, Test leaders don’t need to decide what to do; they need help answering executives who want those “traditional status tools.”

But why do executives need those ‘tools’? Why do they want them?

Those tools promise to bridge the gap created by a lack of trust.

I once knew a project manager I refer to as Biff, who slipped a project by four months at a time, every four months, for about a year and a half. He had no metrics, no documentation, no evidence at all that he could bring the project in on time. In fact, the evidence was stacked against him.

Yet it did not matter. The executives trusted Biff. Biff could hand in the same natty old gannt chart with the dates pushed out by four more months – again – and would have no consequences.

Biff didn’t use evidence. He did … other things to build trust. I would argue that those things can be done better or worse, consciously or unconsciously, with integrity or without integrity.

So, in the next few months, I’d like to try to do a little more writing about what we actually do when we test, and also how we create and build trust in all directions – with our superiors, customers, peers and reports – in software development. I may also do some writing about Miagi-Do Testing, my formal but non-commerical and zero-profit software test training program.

What do you want to hear about?