Testers are fixing bugs

So a few weeks ago, I gave the green light for my team of testers to start fixing bugs.

Contraversial, no?

“Sweet mercy, what have you done?!” I hear you cry. “If testers start fixing bugs, they won’t want to find them anymore and the developers will get careless and there will be more bugs but they won’t be found and then society will collapse and the world will explode!”

Yes, I too had some misgivings about this idea. But for our situation, it seemed like a logically good use of our time. Our test team often find ourselves with a significant amount of downtime, which we usually use to improve our automation suite and to build on our technical skills. But lately we had been experiencing a longer than usual amount of downtime and we were fast running out of valuable things to do. We all have .NET coding skills, so with a large backlog of bugs to fix it seemed like a logical move.

But yes, I had some misgivings. So I laid down some ground rules.

1. This will be a “pull” system. That is, testers can decide to fix bugs if they feel it is a good use of their time. Other team members cannot tell testers to fix bugs.
2. Testers will only fix low-priority bugs. So a tester’s bug fix will never be on the critical path for a release – it will always be optional.
3. Testing tasks will always take priority over bug fixing for testers.

So far it’s been perfectly fine. I have to admit that I didn’t actually fix any bugs and that most of the bug fixing was done by my colleague Pete, who was on a bug-fixing spree. So, the results? More bugs were fixed, team morale was higher and the world didn’t end after all.

Now that more testing work has come in, the tester bug-fixing activities are on hold. But I expect this we will pick this up again the next time we get some time to kill.

8 thoughts on “Testers are fixing bugs

  1. The Director

    I could see this having real benefit, especially if it cuts down on the number of instances where arguing over fixing the bug or whether it’s worth fixing the bug expends more time and effort than, you know, fixing the bug.

  2. Duncan

    Good post – really like the fact that there were clear ground rules.

    Can’t get a much shorter feedback loop than a Tester fixing their own defect.

    Its something we’re trying at my current client, but its in its infancy – If we (as Testers) find a bug we know the fix for, we are allowed to attempt to fix it. Its really empowering.

    That said, I still generally try & pair with a Programmer to show them the defect & the proposed fix just in case, but this is still quicker than raising a defect & monitoring through its lifecycle. It also means the fix is refactored prior to check in.

  3. Darren McMillan

    Hi Trish,

    Good move, and it makes perfect sense!

    In my last work place we had a lot of easy to fix issues. If I had the time spare I would fix these issues whilst still raising it on our internal bug tracking system. I’d then get a nearby developer to verify I hadn’t missed anything whilst testing my fix. Thankfully I never had to pull the “it works on my machine” line :-)

    It’s good fun, and as long as you know your limits and don’t waste time, or introduce risk with more complex fixes, I think it’s perfectly fine.

    Thanks for sharing,

    Darren.

  4. Chris

    I don’t see a major issue. I love the ground rules…..that made all the difference IMO. It could have had potential problems, but you nipped them in the bud with the clear set of ground rules. And, by the way, the ground rules you came up with were great. Made perfect sense.

  5. William Denniss

    If the tester breaks something are they responsible for fixing it? Otherwise you’d have a bit of a moral hazard on your hands.

    Yes, if the bug-fixing-tester breaks something with their bug fix, then they are responsible for fixing it.
    – Trish

  6. Lisa Davidson

    Thank you for sharing this post. Indeed very interesting. I like your view and the way you have explained how and why Testers are fixing bugs. It will great if you can share your experience having a QA Testing Partner or an Offshore Software Testing Partner for any of your testing assignment. Is it a good idea to ask an independent software testing partner to test and fix bugs?

  7. Lucas Ontivero

    I think it is good but I also think it should be done in the same way we devs. do it, this is that before commit the fix somebody else should take a look at the fix. If this is good for devs, it should be good for testers.

  8. Trish Khoo Post author

    Absolutely. Regardless of what your job title happens to be, if you’re committing code then a code review is generally a pretty good idea.

Leave a Reply