SAFe to LeSS; why I switched Frameworks

As I write this I'm planning on transitioning a SAFe Agile Release Train into the Large-Scale Scrum (LeSS) planning approach and away from Program Increment (PI) Planning. This may sound like a big deal, switching frameworks, but it's not really; frameworks are all based on the same Agile principles and values. Despite switching the planning approach, the SAFe principle that I'm comfortable we are still aligned to is

SAFe principle #7; Apply cadence, synchronise with cross-domain planning.

We're just changing the cadence for a while; now I hear you all yelling...

"why the hell would Niall recommend this... it is hard enough launching trains and getting people used to PIs"

I thought you might find the answer interesting. Read on to learn...

There are two reasons I recommended a framework switch for our planning approach:

1. We scaled our build teams too quickly and could not feed them with enough features to create multi-sprint committed plans

Having ramped quickly there was a sudden urgent need to get scope ready for the teams to build. The problem was that due to the complexity of the proposed scope there were many key design, business and architectural decisions that needed to be made; this slowed down flow into the teams. We started having to bring in scope to fill up the PI that could not be committed to (as it had not yet been shaped into feature/stories). We had a supply/demand imbalance.

2. Business is in a very reactive phase with operational pressure to pivot on short notice

With our customer's strategic investment themes being in flux, a longer planning horizon was unrealistic. Combine this with the lack of ready features and a single-sprint planning horizon looked more suitable than a four-sprint increment plan.

SAFe PI Planning as an Experiment

This switch to LeSS is really the result of feedback obtained from running the PI Planning event as an "experiment". The train will keep on moving but plan in smaller increments until demand/supply are rebalanced. We learned a lot as a team of Agile teams from running a PI planning event and once we build a healthy backlog will probably switch back to a SAFe PI Planning approach (maybe).

If this post interests you, check out these: