# Thread: Defect Detection Efficiency (DDE)

1. ## Defect Detection Efficiency (DDE)

How is defect detection efficiency calculated?

2. ## Re: Defect Detection Efficiency (DDE)

It is not really an efficiency as such, for efficiencies are usually expressed as percentages. A defect detection efficiecy would infer that you knew the total number of defects possible.

What you really want are rates and ratios.

With that take several ideas, e.g. defects per test cases, or perfects per hour of testing, or defects per \$ spent on testing.

I know not those which are used the most commonly.

3. ## Re: Defect Detection Efficiency (DDE)

Originally posted by Sulla:
It is not really an efficiency as such, for efficiencies are usually expressed as percentages.
<font size="2" face="Verdana, Arial, Helvetica">In my opinion, you're taking about effectiveness , not about efficiency. Effectiveness is usually expressed in percentages.

Originally posted by Sulla:

A defect detection efficiency would infer that you knew the total number of defects possible.

<font size="2" face="Verdana, Arial, Helvetica">Not necessarily. You could measure the effectiveness or efficiency on different test stages against total number of found defects until a certain moment (which could also include defects found during operation)

Originally posted by Sulla:

What you really want are rates and ratios.

<font size="2" face="Verdana, Arial, Helvetica">I know that the question was about efficiency, but I supposed [img]images/icons/blush.gif[/img] that the intention was to ask about effectiveness. [img]images/icons/confused.gif[/img] . Why ?
From what I know (mostly theoretically) effectiveness metrics come first, and efficiency metrics only when you want to improve something that is working, and was measured before.
I'm sure he will clarify this.

Originally posted by Sulla:

With that take several ideas, e.g. defects per test cases, or perfects per hour of testing, or defects per \$ spent on testing.

I know not those which are used the most commonly.
<font size="2" face="Verdana, Arial, Helvetica">For me these seem to be good examples of efficiency metrics. It's almost identical with what I've intended to respond.

4. ## Re: Defect Detection Efficiency (DDE)

Yes I want to know about the defect detection effectiveness (DDE), and how to calculate it.

I think It was discussed once on this forum but I was not able to find.

DDE = defect detected (during testing) / defect detected (during testing) + defect detected (during production).

Can you please me know if this is correct?

5. ## Re: Defect Detection Efficiency (DDE)

In my limited experience, comparing the # of defects detected during testing vs the number reported by customers will always show pretty good test effectiveness (unless you really released a big piece of ----!).

I've never run into any of our releases where the customers reported more than a very tiny fraction of what we found during our testing.

One method that you could try is 'defect seeding'. Have the developers inject defects into specific builds that you know customers would find and you know that the test team is looking at that area during that time period.
I've never done this either...no time..and always a risk that the defect that got seeded will get shipped, or that taking it out will have a side effect in the app.

Good luck.

Martin

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•

vBulletin Optimisation provided by vB Optimise v2.6.0 Beta 4 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.