Ranking bugs

[Brian Burki]

Hello!

This seems like it's a stretch but here goes...

I'm working on ranking bugs in a backlog.

In a perfect world this is what I'd like to happen:

When a user submits a bug they're required to make a choice from three menus:

type
5: accessibility/pricing problems
4: major usability
3: minor usability
2: visual/aesthetic issues
1: documentation: a documentation issue

impact
5: major harm to the bottom line
4: missed opportunity
3: important but has a workaround
2: a nuisance
1: just weird

ubiquity
5: will affect all users
4: will affect most users.
3: will affect average number of users.
2: will only affect a few users.
1: will affect almost no one.

Note that each option has a numeric value.

The user chooses the appropriate value from each menu and bugs are then ranked in the backlog based on the product of the numeric value of the choices.

For example...

type:5 x impact:5 x ubiquity:5 = 125

vs

type:1 x impact:1 x ubiquity:1 = 1

Thus a bug with a value of 125 would be ranked higher in the backlog vs one with a value of 1.

Thanks for any input!

[Deleted user]

Brian,

We did the same thing for awhile...

Being on the Cloud we couldn't add code so we added a "Score" field and calculated it manually (sigh).

We then setup an issue report for Bugs and sorted based on the Score, showing the relevant information to make decisions as to which Bugs went into the sprint during Sprint planning.

Once the sprint contents were defined, we went back to the "top down" order for all the issues.

Just one approach.

[Brian Burki]

Thanks, Tom. Good feedback.

[Nic Brough -Adaptavist-]

If you were not using Cloud, you'd be able to add code (scripted field or a listener to populate a field) that could do this calculation and tag it into the issues.

However, Cloud doesn't support this yet.

As a design point, it's not really going to be a rank either.  JIRA Software implements ranking, with all issues having a unique position in the list.  With this calculation, you're getting more of a priority, as an issue with 5x5x1 will be in the same position as a 5x1x5 or 1x5x5 and so-on.

[Brian Burki]

Thanks Nic!

I kinda thought the answer would be something like that.

And thanks for the rank vs priority input. 

I knew that some combinations of values would result in the same priority. But I'm just looking for a way to prioritize a list of bugs based on some (imperfectly) objective criteria. 

I see that issues are ranked by the "Priority" field. But I'm looking for a more nuanced and granular way to prioritize bugs. If there were only 5 or 6 bugs at any given time the standard "Priority" field would probably do the trick. But (hopefully) like other software platforms, we've got a few more bugs than that 

If you know of any resources that you think would be valuable, like articles or videos on bug management, feel free to recommend.

Thanks!

[Nic Brough -Adaptavist-]

Well, JIRA Software has a ranking system in it too, and plain JIRA has a priority.  I've tended to use JIRA's priority to help me rank issues, in that high issues generally rank higher than medium, medium higher than low and so-on.  Your calculation is similar but a lot more granular as you've got a lot more than the usual 3-7 values most use for priority.  And, if you could do it, it would be sortable of course.  It's a good guide.