Need help understanding cumulative flow diagram

[Milica Djuricic]

Hi, I'm trying to determine ways of improving our processes in the development cycles (sprints) and what are the issues we have. Reports are enabled on our next-gen project and I'm using Cumulative Flow Diagram, trying to find some peaks or something that could point me in the right direction. The issue is I am not sure how to read this diagram and it doesn't seem to show any kind of "movement". I tried filtering past 2 weeks, past month, past 3 months and it looks pretty much the same + the number of issues axis doesn't make sense cause we never have that many tickets in the sprints..
I have attached 2 screenshots of our reports and would appreciate any kind of advice. 

The things I did notice so far is that the reports contains:

- Issues from the backlog, so not just from the board or the ones we worked in specific period..
- Issues that were completed long time ago, so not just from the period I chose for the report
- all statuses except ToDo and Done seem like nothing is happening or there is no progress whatsoever towards the end date

Am I doing something wrong and is there some other way for me to see/inspect our ways of working during the sprints? 

I would like to see how often we fail on QA stage for example, so this would give me some insight..or how often a ticket is returned from code review to know if we have issues there perhaps..anything useful really..

Also, i am open to any other advise in the area of sprint-inspection and process improvements..but first I need to inspect in order to know what to improve :) 

Thank you, 
Milica

[Bill Sheboy]

Hi @Milica Djuricic 

You are using this "Cumulative Flow Diagram" (CFD) report correctly.  The design in JIRA just limits what you can see.  This is the Next-Gen version of this report.  The Classic project version of this report has the Refine Report option, but Next-Gen does not.

What a CFD can show is:

• Arrival rate of work (slope of the top of the backlog items)
• Completion rate of work (slope of the done items)
• Work in progress (WIP) counts (height of the items, between backlog and done)
• Lead time (horizontal time distance from start of in progress to done)
• And many other indications of flow issues (spikes, flat spots, and drops in overall height)

One challenge with this CFD implementation is there is no way to start the y-axis at a specific value *and* preserve the scaling.  What you can do to improve visibility of the backward flow you note for QA stage is:

• Hide the backlog items (uncheck todo)
• Limit the time-frame to 2-3 sprint lengths, or less
• Selectively hide other status values which appear stable

If that does not help enough, your other options are to buy better chart add-ons from the marketplace or to build your own reports, using logs, automation rules, or data export.

Best regards,

Bill

[Milica Djuricic]

Thank you very much Bill, I will try with this approach you suggested and hopefully I'll get some insight :)

Best, 

Milica

[Bill Sheboy]

Hi, Milica.  If this ends up helping you, please consider marking this question as answered.  That will help others in the community find solutions faster.  Thanks!