Pay per bug

 Subscribe in a reader

Tuesday, June 12, 2007

One thing we like to do at DrivenQA is to find new ways of working. Anything that encourages more testing can only be a good thing.

Project patterns

We've had a look at selective projects we have been working on and have discovered a pattern: the cost per bug is remarkably similar.

What does it mean?

As a software testing consultancy, the potential is to work in a different method. Instead of, for example, working on a fixed price or time and materials basis, it is possible to charge per bug found. For the client it guarantees that bugs will be found and that the money is well spent.

It also means that the focus is on finding bugs. This is an important point to remember. Software testing is mostly about finding bugs. However, in reality many software testing aspects of projects don't involve enough hands on testing to find bugs, and not just any bugs - the bugs that matter.

What it does not mean

It doesn’t mean that there is a universal cost per bug. The cost is directly related to the type of project and mission requested by the client. For example, content bugs are much easier to find and reproduce (and would cost less) than functionality or security bugs. This also means that clients could essentially pick and choose the areas testing should focus on – or of course, we could recommend what and how to test.

Yeah, but could it work in practice?

It can and DrivenQA has done it, this is how:

• Tell us about your project – and we’ll tell you if it is suitable for the ‘Pay per bug’ approach
• You tell us our mission - e.g. what are we to test? what is the timeframe/budget?
• We quote you on how much each bug will cost. We can even quote for different levels of bugs. For example, a content bug might cost you half the price of a functional bug.
• We make a plan and go bug hunting (setting ourselves a limit and keeping you informed at all stages).
• After our round of testing we will review everything to make sure everyone is happy.

Our Approach

The problem with software testing is that it is often perceived that it takes too long and documentation is excessive. It’s no wonder people don’t want to pay for testing!

Our core approach is to use exploratory testing techniques. Experience shows this is a very effective approach to software testing that places emphasis on skilled testers who learn and create tests on the go. As testers our main output of our work is bugs. It is what we want to find and is probably the most enjoyable part of testing. Exploratory testing is ideal for this scenario.

The best bit is...

The best part of it is that you are guaranteed to get bugs for your money!

Get in touch to see if this approach could work for you!