Hi all,

regarding the pdca(plan-do-act-check)-cycle there's written a lot about plan, act and check but you don't find a lot about do.

What I am searching for is how to get from the bunch of single/raw data to reasonable metrics. There are different ways for calculating for example customer satisfaction indices with weights and dividing them afterwards to get a weighted average. But what I have never seen is a methodology that helps to decide how to calculate a ratio based on an existing framework.

How do you calculate your metrics?

Kindly,
Dirk

Welcome to The Cove! :bigwave:

I will give you an answer that may seem maddeningly vague.

The metrics are calculated against however it is you would know the process went well. A target time frame, no errors made, customer satisfaction after a service call is at the "I would definitely buy again" level, and so on.

The difficulty lies in calculating qualitative results.

For this I assign a value for each level of success, like a rubric does. I have given a sample of what that would look like in this little paper, Some Options for Risk Based Auditing (http://elsmar.com/Forums/showthread.php?t=20536).

Please let me know if this helps or not.

Thank you Jennifer.

It's an interesting and helpfull example for the special situation and gives an insight of on special way to calculate metrics. So it was helpful to me as it has shown one example of a topic with a large scope.

Actually, I am searching for an overview about the possibilities of getting from raw data to metrics. Different calculation methods maybe with a decission tree, helping to decide which method has to be used in which case.

I help a lot of folks develop metrics, and do have the following presentation available on our Department of Energy website:

http://www.hanford.gov/rl/uploadfiles/VPP_3_Choosing_PIs.ppt