How to display number of issues in historical statuses over time?

Note that “Status” dimension returns the current status of an issue even if “Time” dimension would be used in the report. Therefore measure “Open issues” may not give you expected results when “Status” dimension is filtered.

To display a number of issues in historical statuses over time, we recommend importing issue change history (via eazyBI import settings).
That will create some new dimensions and measures in your eazyBI JIRA issues cube.

Next, you could use measure “Issues history” together with “Time” dimension and filter your report by “Transition Status” values.

3 Likes

@martins.vanags

Hello :smiley:,

how can I display the transition status of linked members depending on the actual issue? The actual issue have issue type “A” and the linked member issue type “B” because of different Projects.

Hi Milad,

Please find my answer to your question here:

Martins / eazyBI support

Hi @martins.vanags,

I did it exactly your way, and I got a question about the time dimension:

When time level is “Day”, then eveything is accurate, for example for May 2018 it looks like this:

But I try to get the count per month. When I just select “Month” as time hierarchy level, it looks pretty wrong, it shows me just 2 WIP items for May :frowning: Seems like he chooses the count of the last day of the month?:

Best would be to show an average per month. Do you maybe know how to do that?

Thank you in advance.

BR,
Markus

Hi,

You are right with your asumption. Measures with sufix history would display the count of the last day of the month.

I believe you should create a new calculated measure that would return the average number of issues history for displayed month and calculate it from all days from the month.
Try this code:

NonZero(
Avg(
Filter(
Descendants([Time].CurrentHierarchyMember,[Time].[Day]),
[Measures].[Issues history]>0
),
[Measures].[Issues history]
)
)

However, I find this calculation misleading when I need to see the development of issues in a status for a month.

Martins / eazyBI support

Hello Martin,

oh waow thank you very much for the fast and great response.
It works :+1:t4:

Yes you have to take care how you use that averages, I agree (background is the little’s law topic, where you always use averages).

For an accurate view, daily is definitely the best.

Weekly also has the drawback of the last day of the week, so good to know how eazy BI works here.

But now I can use the suitable metrics depending on the topic.

Thank you again and best regards,
Markus

Hi Martins,

Just need to check the benefit or use case of this report “[How to display number of issues in historical statuses over time?]”.What exactly User can abstract from this report.

Thanks & Regards,
Gaurav

Hi,

You could follow the development of issues in a particular status over a time.
To understand what is the average number of issues in “In progress” status month by month (week by week).
How else would you know how many issues there was in “In Progress” status two months ago?

Martins / eazyBI

1 Like

Thanks so much Martins i got it.

@martins.vanags How your issue history works for counting the #of issue at particular date over time.

Example: I want to ko the #of issues at Assigned on 10th of January

How does your issue history work?
Is it calculate based on EOD of 10th like till 11:59 PM of 10th of January?

Hi @asudhakar493

You can still use the measure “Issues history” with “Assignee” dimension members and filter the report by the “Time” dimension.
If you select Jan 10, this would calculate the number of issues for the assignee at the end of Jan 10 (11:59pm)

Martins / eazyBI

Thanks, martin for the reply.

I see some cunt issue with history.
Ex: #of the assigned status issue on 10th Jan 2021 is 30
When I click drill through issues I can see 33 records received and lost column having -1 and 1 this is causing the count issue i guss.

Can we eliminate -1 from the count so that count will come correct i guss.

Hi @asudhakar493

please reach out to support and provide more details including exported report definition
Issues history should not return negative values unless the context for the report is built wrongly or there is something wrong with the changes coming from Jira changelog.

Martins / eazyBI support

NonZero(Sum(
Filter(
Descendants([Time].CurrentHierarchyMember,[Time].[Month]),
[Measures].[Issues history]>0
),
[Measures].[Transitions to status]
)
)

How can I specify this to the transition status of In Progress? It won’t accept it if I change [measures].[transitions to status] to [measures].[transitions to status],[transition status].[in progress]
I would like it to count/identify the issues that were moved to the In Progress status each month. Then how could I also include those that remained in the In Progress status each month?

Hi @Alyssa_A

Try this formula to count issues that were in the “In Progress” status at the end of the time period (month/year/day).

(
[Measures].[Issues history],
[Transition Status].[In Progress]
)

And this would count issues that moved to “In Progres” in the time period:

(
[Measures].[Transitions to status issues count],
[Transition Status].[In Progress]
)

Martins / eazyBI support

@martins.vanags I see issue with the history and I have rectified the history and now count is showing fine.

Next, you could use measure “Issues history” together with “Time” dimension and filter your report by “Transition Status” values.

Hi @martins.vanags,
I don’t understand what is the difference between combining “Issues History”, “Time”, “transition Status” and combining “Issues History”, “Time”, “Status”.
I can see that the results are different, but I don’t know why.

Thanks a lot for your help,
Marilou

@Marilou

The “Status” dimension would always group issues by their current status today, however “Transition status” would look at historical status in the report.
You notice the difference only when start using “Time” dimension, because one dimension would group issues based on their current status, which can change for the issue over time.

Martins / eazyBI team

1 Like

@martins.vanags

Thank you so much for the explanation.
Does that mean that combining ‘Time’, ‘Issue History’ and ‘Resolution’, the issues would also be grouped by the current Resolution today?
If it’s the case, is there a equivalent of ‘Transition Status’ for resolution?

Thanks again :slight_smile:

Marilou

@Marilou
Resolution is one of the few dimensions eazyBI automatically ipmorts change history about.
The Resolution would play along with “Time” dimension and group issues by historical resolution.

Martins

1 Like