Sizing Up the Enterprise


As teams begin to estimate User Stories, they may explore different approaches such as T-Shirt sizes and Fibonacci sequences that stop at 8 or go much higher.

This freedom to choose a relative sizing style allows a team to adopt what fits well within their work environment.

While this flexible approach is quite useful at the grass roots team level, it does pose an interesting challenge in the Enterprise setting where Agile teams roll up into an Agile PMO.

How do those in charge of the overall strategic planning, make informed cross team decisions with Product Backlogs of such varying size criteria? How can they identify the features that involve the least complexity, effort and doubt coupled with a high return on investment?

scrumology.net agile pmo sizing

Before we go into my suggestions on how to address this, let’s explore a common fallacy that is being evangelized in the Enterprise today.

Mandating Story Points to Ideal Days Solves Cross Team Sizing Inconsistencies

“1 Story Point = 1 Ideal Day (6 hour work day)

Seems quick and easy doesn’t it? Mandate that all of your Agile teams conform to this and your problem is solved!

By doing this, however, you’ve inadvertently stripped the Story Point of its original intent while roadblocking your team from personalizing (and humanizing) the process. You risk dismissing the conversations about complexity, effort and doubt while focusing on the mythical 6hr work day.

Another unintended side effect of tying Story Points to actual hours is that it isn’t long before people make the dangerous link between Story Points and Budget.

So what is the silver bullet to this issue? As is the case with all things Agile, there is no silver bullet! Solving this issue depends on the nature of the Agile teams and their relation to the business within the Enterprise.

Are these teams separate business units within the organization, or do they all contribute to the same product?

If the teams each have their own role within the organization and are only loosely tied to the same business goals, my suggestion is to let them be for the most part. Sizing, and especially Velocity, does not translate well across teams or up the organization. It will be an apples to oranges comparison, and you should keep your eye on delivering business value. As long as your teams are collaborating by sharing their Release Plans, does it really matter if they use a T-Shirt size or a Fibonacci scale as a means to an end?

If the teams do happen to roll up into an overall product, then I suggest that sizing occur from a single Product Backlog before decomposing them into each Team Backlog. With this approach, you can have the overall strategic conversations early. I would much rather have representatives from each team weigh in on a single Product Backlog, than try to make sense of it from the bottom up. This also brings a consistency to the sizing while allowing each team to have flexibility at the Sprint Planning level.

To summarize, tread carefully when trying to apply consistency across Agile teams within the Enterprise. It may not make sense to mandate sizing techniques, especially if it causes more harm than good.

Share and Enjoy:
  • Digg
  • del.icio.us
  • Facebook
  • Google Bookmarks
  • StumbleUpon
  • Twitter
  • HackerNews
  • LinkedIn
  • MySpace
  • Posterous
  • Technorati
  • Yahoo! Bookmarks
  • Reddit

, , ,

  • Hi David,
    I agree. usually the PMO/PPQA wants to standardize these things. Last week, I was in a same situation. I was very appreciative of the fact that they were buying into the concept of estimating in story points. But then I had to put some breaks on and say - we are not there yet ( to come up with standard story points) . My advice was to have a few teams try out this approach of user stories and story points. Let’s get mature in estimating. After a while we will talk again about the need for standardizing and if so how to do it.
  • We are applying the scrum/agile framework outside of just product development and across functional groups within the organization. Your point is right about different teams. Its more important for a team to hone in on their actual velocity and work to improve on their accuracy with better reference stories etc. We did try a few techniques to at least keep teams in the same ballpark at first. We created general reference story criteria to mostly help people understand sizing differences. Part of this all working though is alignment from everyone around continuous improvement and highest efficiency possible.
  • Kevin,

    Thanks for your feedback! Please keep us posted on the challenges you face with agile adoption if possible. In my opinion the agile community needs frequent updates from those in the trenches or it risks becoming an echo chamber.

    -David
blog comments powered by Disqus