Sunday, 28 December 2008

DDP is my favourite metric. It is very easy to measure, easy to understand, and gives you an idea of the quality of the testing, in terms of defects found. Many people have found it very useful. If you are one of those people, please respond to this blog and tell me about your experiences!

What is DDP?
DDP is Defect Detection Percentage. It is a measure of the effectiveness of testing (or reviewing) at finding defects. The calculation is very simple: The number of defects (bugs) found in testing* divided by the number of defects found in total so far**.

*DDP can be calculated for different iterations, different testing stages (e.g. integration and system test) or testing by a defined set of testers.

**You can choose any point in time to calculate DDP - for example, one month after an iteration is released (or when the next iteration is released).

There are lots of variations on DDP - if you have any questions, please ask!

Dot

4 comments:

Randy Rice said...

Hi Dot,

I'm also a fan of DDP because you don't have to get into the sizing measures such as lines of code or function points. I have one client that got a rather nice bonus based on his own self-initiated effort to measure DDP on a number of projects. He was able to show an increasing trend of test effectiveness for his department. It was impressed his senior management!

Dot Graham said...

Thanks Randy. Nice to hear when people impress their managers!

I had a client who used DDP at Board Level as one of their Balanced Scorecard measures. I think he got promoted!

Pedro Montenegro CorrĂȘa said...

Hi Dorothy,
Fyi, I'm currently implementing DDP in a Testing and Auditing Center in a government body in Brazil. I hope it's as successful as Randy's experience.

Thanks for an excellent speach about cognitive biases in Sao Paulo/Brazil.
I was responsible for the only question you've got that day, regarding applying behavior economics to test/quality context.
On that regard, I want to submit the following Forbes article on the subject as food for thought.
http://www.forbes.com/2010/03/22/behavioral-strategy-decisions-leadership-managing-mckinsey.html?boxes=leadershipchannellatest
I'm looking for co-authors on the subject... in Brazil is considered too "out there" for IT profs.
I'm about to put out a paper and would love to have your/your group's input on the subject.

Pedro Montenegro CorrĂȘa said...

Hi Dorothy,
Fyi, I'm currently implementing DDP in a Testing and Auditing Center in a government body in Brazil. I hope it's as successful as Randy's experience.

Thanks for an excellent speach about cognitive biases in Sao Paulo/Brazil.
I was responsible for the only question you've got that day, regarding applying behavior economics to test/quality context.
On that regard, I want to submit the following Forbes article on the subject as food for thought.

http://www.forbes.com/2010/03/22/behavioral-strategy-decisions-leadership-managing-mckinsey.html?boxes=leadershipchannellatest

I'm looking for co-authors on the subject... in Brazil it's still considered too "out there" for IT profs.
I'm about to put out a paper and would love to have your/your group's input on the subject.

Best regards,
Pedro