Attaching the test data on which I am trying to create some Pivot tables.
1.Parent - Child Count(1) --- In this PIVOT table the counts are mostly ok except for when the TYPE is a bug eg - Child task 18109 as one entry is for a developer and the other entry is for a QA. So in this pivot table the Bug count is duplicated. I looked for Count removing Duplicate in internet but not getting any such option. Is there a way to achieve this
2. Count By Child Task by Sprint(2) - Same is the scenario out here. The counts by SPRINT are ok except for when there are bugs.
3. Count By Child Task(3) - This is where the count gets exorbitantly high. I can have a DEV task which spans across SPRINT and is OPEN and the that task eventually closes. I would like the task which is closed to be counted in CLOSED but not in OPEN. Similarly if a DEV task in open across SPRINTS and has not closed it should be counted only once which is not happening currently.
4. Count By Parent Task(4) - This is the pivot by PARENT task. I would not like the PARENT task to be counted as CLOSED unless all the CHILD task in it are closed. Which is not happening currently.
Would also like to add that I cannot take the last Sprint to calculate the open task count as a task which was open in the previous SPRINT might not necessarily come in the next Sprint because it lost it's priority. But still I need to count it as an OPEN task in the consolidation.
Is the above directly achievable via Pivot or I need to create some intermediate table to get what I am looking for.
Any help or direction would be appreciated.
Sunil
Hi Sunil,
Thanks for sharing your file, however I'm not following what your desired results are. Your first example references Child Task 18109 and you say "one entry is for a developer and the other entry is for a QA", but actually 1 entry is for Dev and 4 are Bug.
Please review your file and provide clear examples of where the problems are AND what you expect to see, so we can follow the logic back to the source data.
Mynda
Hi Mynda,
Thanks for the quick response. What I intend to say is that when I count the number of Bugs it should be counted 18109 as 1 and not 4 (Sorry for the typo. I ended up calculating by Sprint). 18109 is a Bug but it has 4 entries across the Sprint but I would like it to be counted as 1. So I would like to count the Bug 18109 as 1 and not 4 because it is a single task.
However when I look into Count By Child Task by Sprint(2) pivot table, for Sprint 22 in the pivot table I would like the count of Bugs to be 2 and not 3 as 18019 is a single task and not 2 individual task. Also the Grand total of Bugs should be 5 and not 10. As in the excel the distinct count of Bugs is 5 and not 10. Similarly the count of overall Dev task is 23 and not 29 as being calculated.
Sorry for creating the confusion.
Thanks
Sunil
Hi Sunil,
Why would count of child task 18109 bugs for sprint 22 be 2 when there is only 1 in the dataset? Sprint 21 has 2 bugs, but 18109's sprint 22 only has 1.
Mynda
Sorry I mean 18220. Just too many numbers around. getting confused. So total bug count should be 2 and not 3 what is currently being displayed for Count By Child Task by Sprint(2) pivot.
Similarly for Sprint 21 the total count of bugs should be 2 and not 4.
Hi Sunil,
For 1 and 2 you can load to the data model/Power Pivot and use distinct count. For 3 and 4 you need to rearrange the data to remove the items you don't want counted and or classify the tasks into open/closed. In the attached file I've done this with Power Query.
Note: with 3, it's not clear what 'Type' you could retain after removing duplicates that were open. Anyhow, I hope that points you in the right direction.
Mynda
Thanks Mynda for the soluction. So basically I need to create an intermediate table to have the calculations I want. Cannot use the raw data and in one shot create the Pivot's that I am looking for.
Thanks
Sunil
Correct.
