SAFe’s big room planning is a great training exercise

Photo by Jaime Lopes on Unsplash

SAFe is well known for it’s quarterly big room planning exercise, where agile teams, stakeholders and other hangers-on get together to forecast and plan the next few sprints of work. It’s quite an event, and unless you happen to have some very large facilities already you’ll end up in a hotel ballroom or other large hall. You’ll find yourself asking, will we really do this every quarter from now on? How do we make this less painful? Will we actually get good at this? Do we have to run it this way?

What big room planning gets you is a great training forum. The work leading up to the first PI planning has been some serious organizational change: New roles and functions, teams deconstructed and rebuilt, new prioritization and management approaches. What better way to teach all these people en masse how to work in their new roles, teams and priorities than a giant workshop!

While PI planning appears to be the be-all and end-all of planning at first glance, it’s really the culmination. Good planning can’t happen in a day or two. It starts a good month before, with stakeholder consultations, roadmap reviews, feature writing, and of course dear old WSJF. What you get in PI planning is the final forecasting and vetting of those plans with some reasonable due diligence. While everyone learns in the big room, coached along by agile coaches or consultants, one major benefit, and behavior it hopefully ingrains, it’s walking across the room to discuss dependencies and risks with other teams and stakeholders so that they can be unraveled and understood. The silos melt before your eyes in a palpable way.

As time goes on, PI planning beings to morph away from the official recipe to suit your needs. While the pandemic forced us to go virtual with planning, we were discussing it after three or four big room events due to the increasingly costly and complicated logistics as we added more teams. The nature of readouts and plan assessments get refined as well. At the first PI planning the scrum masters took it in terms to read out their velocity and the coaches took a list in their notebook! Fast forward a couple of years and it’s barely discussed — we have that automated with tools and dashboards for each release train. Over time you’ll choose many tweaks to the formula.

Unforunately, after the first 2 or 3 planning sessions, there was no sense that we were on a maturity curve (easy to see now in retrospect) but stuck with this one labor intensive way of planning, and it actually raised many questions about its continued viability. By about the one year mark there was a loss of confidence and much talk about dropping SAFe or moving on to something else.

This also wasn’t helped by a the tenor of the transformation, which tended toward following the book to the letter rather than by the principles of Lean and Agile. For example we had a PO sync that was a fairly pointless status check, we had Business Value ratings that seemed redundant. Again, looking back, I can see why the framework asks for those things, but they didn’t make sense at the time, and rather than getting to the heart of their purpose, they were instituted per official SAFe instructions. Subsequently we have evolved our own, equally good versions.

Big room planning was a mixed bag for UX. We were able to walk up to any team and dive into their plan and ensure they had accounted for design and research in appropriate ways. Ad hoc conversations sprung up around design strategy that would otherwise have been unwieldy alignment meetings. As a team “outside” the system, UX is unfortunately sidelined and needs intentional steps to counter the positioning (or absense of it) in the SAFe system. However, big room events made it possible to dip into plans and contribute to the process with being formally part of proceedings. Moving to virtual we have lost some of this but are working on bringing it back, with easily accessible Miro boards for sprint planning and team drop ins.

So what’s my advice about PI planning after two years / about 8 PIs?

1. Don’t adhere mindlessly to the rule book, figure out why the rituals and rules were put there, make it your own.
2. Put plenty of effort into structuring the run up to PI planning: Roadmap reviews, Feature writing, stakeholder syncs.
3. Embrace the training element of big room planning, recognize that everyone is in the “crawl” phase for 2–3 PIs.
4. Change it up after a couple of PIs to suit your needs.
5. Have a bottle of wine handy for management review the first couple of times! Lots of anxiety about how it’s all working (or not).
6. UX leaders — use big room planning to understand your teams’ engagement model and how to position them for success (In scrum? External dependency?) Evolve this over time too!