Jan 24, 2012

Avoid Trivial Bugs, Report What Works

I’ve been testing this darn thing all morning and I haven't found a single bug, or even an issue.  My manager probably thinks I’m not testing well enough.  My other tester colleagues keep finding bugs in their projects.  Maybe I’m not a very good tester.  My next scrum report is going to be lame.  This sucks, man.

Wrong!  It probably doesn’t suck.  Not finding bugs may be a good thing.  Your team may be building stuff that works.  And you get to be the lucky dude who delivers the good news. 

If there is lots of stuff that works and no bugs, you have even more to report than testers who keep finding bugs.  Testers who keep finding bugs are probably executing fewer tests than you so they know less about their products than you.  Instead of figuring out what works, they are stuck investigating what doesn’t work.  They’ll still need to figure out what works eventually, it’s just going to take them a while to get there.  And that sucks.

My manager is probably looking at my low bug count metric, thinking I’m not doing anything.  Logging bugs makes me feel like a bad ass.  There must be something I can log…hmmm…I know, I’ll log a bug for this user message; it’s not really worded as well as it could be…it has been like that for the last four years.

No!  No!  No!  That’s gaming the system.  It’s not going to work.  You’re going to get a reputation as a tester who logs trivial bugs.  Your manager is only counting bugs because you’re not giving her anything else.  She just wants to know what you’re doing.  Help your manager.  Show her where to find your test reports, session sheets, or test execution results.  Invite her to your scrum meetings. Tell her how busy you’ve been knocking out tests and how bad ass your entire project team is.

Think about it. 

Reporting what works may be better than reporting trivial bugs.

4 comments:

  1. Good points, Eric. But I don't think this is an either/or situation.

    The manager may indeed be using a (foolish, in my opinion) bug count metric. If that's the reality, it would be only human nature to submit any bugs that are found - old or new - as that's how the reward system apparently works in this shop.

    At the same time, a progress report indicating what has been tested so far, and what has not, almost always adds value.

    ReplyDelete
  2. The word you are looking for is "coverage". All those bugs mean less test coverage. Coverage reporting is a central function of the Testing Dashboard, and you can't bring your coverage up if your bug counts are too high.

    ReplyDelete
  3. Finding worthy blogs is not easy now and it's a pleasure to run across your site.More informative, explained very clearly it is very help full... thanQ very much for good post.

    ReplyDelete
  4. "Instead of figuring out what works, they are stuck investigating what doesn’t work."

    Why did you use "stuck" referring to context of the other testers? Isn't "investigating what doesn’t work" more important than "figuring out what works" (other factors being equal)?

    ReplyDelete