Dear community,
Dear AppFire,
This is a complex / longer one. I hope you bare with me.
BACKGROUND:
We are using BigPicture for 2 years now. We spent roughly half a million Euro in consultants ... and yet, we cannot seem to get happy with BigPicture and try to code our way around it.
I believe we are using the app wrong. However, we cannot seem to find a (automated) way to get it right.
We have development projects which run for 1-3 years, depending on complexity.
= exhibit A: time horizon is longer than anyone can plan or estimate beforehand with precision = non-agile / non-iterative approach = classic project management.
OUR PROBLEM:
Before starting a project, we try to roughly estimate things to do. For example, create a feature list (we use Epics for this, but it could be any issue type).
Let's say we estimate feature A with 100 h.
= exhibit B: rough estimation is done prior to project start.
After a project gets approved (e.g. budget is approved), detailed planning and estimation commences and is extended over time. This is done by adding Stories to the Epics.
Here is where it gets wild-ish: some teams estimate the stories, some don't.
Lets take route 1 first: it gets estimated, e.g. 10 h. Now we have 100 h in Epic A and 10 h in Story B. However, the 10 h - in our way of thinking - are actually part of the 100 h! Now, if we look at resource and team workload, we see 110 h, not 100 h.
= problem 1) estimations are entered multiple times, at multiple stages with different granularity.
Lets take route 2: it doesn't get estimated. Cool. Problem 1) does not manifest. However, people start logging time, e.g. 5 h. Since the estimation of the story does not exist (OE = 0), any time spent reduces the remaining estimate but never below 0. Since we didn't estimate the story, the RE stays 0.
= problem 2) the remaining estimate of the Epic is never reduced even though time was spent in the hierarchy "below".
From 20+ years with Jira I am thinking: this is wrong. Estimations on one issue only belong to that issue. Period. There is no "hierarchy" of time estimations between issues, neither in Jira nor in BigPicture, even though BigPicture suggest a tree-like hierarchy visually.
Having said that, thinking in (time) budget, the estimates of one issue in the work breakdown structure should deduct or be part of the estimation of a "higher up" issue in the hierarchy. Otherwise, summing up efforts adds "the same hour" multiple time.
The default for Jira, BigPicture, SumUp, RichFilter, and other apps is: Sum.
This only works, if estimations aren't entered more than once at various levels.
= otherwise it is an over-reporting of workload
BigPicture and RichFilter have the "sum without parent" feature. This doesn't work (well) for 2 reasons: 1) it isn't the default and people have to manually switch to it every single time increasing human error by magnitudes; 2) it assumes that every of the 100 h of Epic A are re-estimated in a story. In the example above, only 10 h got estimated at story level. The other 90 h would get "lost"
= under-reporting of workload.
Any ideas on how to solve this?
a) what could be a change in behaviour on our end to solve it?
b) what could be a technical change (e.g. automation rules) in the system to solve it?
c) can this even be solved or is it picking plaque or cholera?
Doing it manually could look like this: if an estimation is done, then deduct it from a "higher-up" issue, if possible.
Doing this automatically with Automation rules, Java Code, events, ... didn't really work for month now. It always breaks other features or UI. I believe, because it goes against Jira's design to begin with.
Thanks,
Janek
Online forums and learning are now in one easy-to-use experience.
By continuing, you accept the updated Community Terms of Use and acknowledge the Privacy Policy. Your public name, photo, and achievements may be publicly visible and available in search engines.