What is Undone Department?
“Undone department—This department, ideally, does not exist.
Unfortunately, sometimes the teams are not yet able to create a true shippable increment every Sprint. This is reflected by their “Definition of Done” not being equal to “Potentially Shippable.” The difference between them is called Undone Work. Someone needs to do this Undone Work, and a common “solution” is to create separate groups that pick up the “undone work”—the undone department. More on this in the Definition of Done chapter.
Undone departments such as test, QA, architecture, or business analysis groups should never exist in the smaller LeSS framework groups; rather they should be integrated into the teams from the start. On the other hand, we unfortunately still frequently see an operations or production undone department in LeSS adoptions, as they often cross organizational boundaries.
A goal in every LeSS adoption is to remove the undone department. How long will this take? The answer is highly dependent on how fast the organization improves its capability.” – from the 3rd LeSS book (Large Scale Scrum: More with LeSS)
“Bad scaling is one of the biggest ‘agile problems’ of modern days for companies.
Bad scaling is one of the three (the other two are : “agile tools” mania and falling a victim to big consultancies’ industrial model [see/play Dave Snowen’s view here:
Bad scaling comes in the form of trivializing agility at is core, weakening agile roles, plagiarizing and relabeling someone else’s experiments and calling them ‘operating models’, copy-pasting Scrum and Scrum roles into Fractal Geometry that look great on paper.
Are there better ways to work? Probably not, if the ultimate goal is to relabel existing enterprise complexity with fancy agile terminology and then call it “enterprise scaling”. But there could be better ways to work if an ultimate goal is to simplify existing complexity (de-scale), and by doing so, improve your chances to scale agile ways of working (e.g. do Scrum, by more than one team, working for the same Product Owner, on the same product, out of the same backlog).
In this session, Gene will expose some classic pitfalls of bad scaling and will recommend how, more good things could be done with less stuff,…how things could be done better, using Large Scale Scrum (LeSS).”
For so many companies, inability to estimate -> forecast –> budget –> manage expectations of clients/users/management is a big issue.
This is not surprising.Under assumption that everyone understands
Scrum and has first-hands experience with all of the below:
“Our teams cannot accurately estimate work”
“Our teams are not stable and not dedicated. Capacity is unpredictable”
“Not everyone is cross-functional and everyone does their own work. What is the point of estimating together?”
“Our teams get confused with historical Velocity-driven planning vs. Capacity-driven planning. When to use what and why?”
“Our teams have hard dependencies on other teams. Some resources are shared. What can we do?”
“Our estimation and metrics get so-so fudged, as they roll up through multiple organizational layers to the top”
“Our organization is fractal: one PO per one team, each team works in a silo. We cannot “standardize” estimation”
And most of the below:
-“Individual teams are more or less OK, with estimation, but how to we ‘normalize’ estimation across multiple teams?”
-“How do we ‘scale estimation’, so that our sr. management can get a sense of how much work we have?”
-“When is the best time to estimate? How provides estimates?”
-“How can we effectively estimate with many developers not be co-located?”
-“When and where is knowledge about the estimated work being used?”
Let’s talk about how most of these problems can be addressed in Large Scale Scrum.
Instead of implying that LeSS is a silver bullet or blanket solution to everything, lets focus on specific dynamics of LeSS, with respect to:
specifics of LeSS product group design
intimate synergy between teams
roles & responsibilities, artifacts & events
in order to understand WHY estimation and forecasting by LeSS product group (e.g. 50 developers) would be more reliable than the same, done by e.g. 50 developers of a traditionally-designed organization.
Although LeSS is a scaling framework, in reality it requires organizational de-scaling. It is also, organizational design framework that helps to address core elements of organizational design: HR policies, finance/budgeting, vendor management, site strategies – areas that are not too comfortable for many companies to address.
System and Lean thinking drive understanding of LeSS.
LeSS is not about: ‘best practices’, maturity metrics, RAGs, KPIs, tools, and operating models.
LeSS 3 adoption principles are:
1. Deep and Narrow (not broad and shallow)
2. Top – Bottom + Bottom – Up.
3. By Volunteering only
On 05/12, a visionary pragmatist, Diana Larsen is co-founder, Chief Connector and a principal coach, consultant, and mentor at the Agile Fluency® Project spoke to Large Scale Scrum Meetup of NYC. Diana co-authored the books Agile Retrospectives: Making Good Teams Great; Liftoff: Start and Sustain Successful Agile Teams; Five Rules for Accelerated Learning. She co-originated the Agile Fluency® model and co-authored the eBook, The Agile Fluency Model: A Brief Guide to Success with Agile.