True issue close date from the list of statuses

[macin]

Hi,

I hit the wall while trying to calculate true cycle time per JIRA issues.

What I managed to do

1. I started with the list like this

I managed to get to the point where next to the start date, I have an end date using the index shift of records sorted by date (based on https://youtu.be/map7WjGPXoI)

3. I mapped granular statuses to status category

4. Now what I am struggling with is the next step. Observations

* Earliest start date - to know when something was created

* Earliest "In progress" occurence to understand when someone started working on the issue 

* Earliest "Done" occurence, which is not followed by any other statuses different than "Done"

I have a special case which I have no idea how to report on corectly:

1. There is an early status change to "Done" on 18/05/22. But seeing the rest of the entries, I cannot take this initial Done as completion date, because the true completion date is 15/07/22. 

2. Additional challenge I have is that there may be many status changes within a day. I should report the latest one on the day only. Not sure how to approach this problem either.

I am not looking for step by step instructions (although anything like that would be nice). Could you direct me please to some blogs or videos, where similar challenge is well described?

I'd love to learn to solve above.

My final report should consist of number of started issues and number of closed issues per day and/or month.

Thanks!

[ronrsnfld]

I don't think you need your intermediate steps, unless for some other reporting requirement.  Just GroupBy the Key and then apply your algorithm to each subtable, using a custom aggregation.

After you've done that, you can check on the started/finished numbers per month.