In my account I defined cycles to group statuses by responsability / involved party. I use the out-of-the-box measures EazyBI creates for me, specifically “Average workdays, but I might have misinterpreted the description given here.
What I want: I want to report the average time spent in (work-)days per cycle for all issues closed (or resolved, does not matter to me) grouped my month. It does not matter when the issue was in that cycle.
What I get: some issues reported in e. g. September are reported as having no time spent in cycle X, but they did spent time in previous months. At the same time, one of these issues reports cycle Y with ~ 73 days for September which is more than September has days.
The key is to understand by which date the cycle duration measures are grouped on the Time dimension.
eazyBI can tell the cycle time when the cycle is completed. Therefore, “ workdays” measure shows the cycle time by the date when the issue exited the cycle. This might become tricky if the issue enters the cycle for the second or third time; in that case, the total time in the cycle will be displayed according to the latest cycle exit date.
You might want to use measures “ workdays of resolved issues” and “Issues with of resolved issues”. Both measures displays data based on issue resolution date regardless when the cycle started or ended or how many times issue went though those statuses. From those two, yuo can also get the Average of workdays of resolved issues by dividing them:
[Measures].[Progress workdays of resolved issues]
/
[Measures].[Issues with Progress of resolved issues]
Using those measures ensures each issue appears only once in the report (based on resolution or closing date). Usually, resolved and closed issues are no longer worked on, ensuring that data from past periods remains unchanged.