After I posted a picture and some comments about a first Scrumban simulation on the company's news feed, the replies were overwhelming. A lot of people wanted to do this simulation with their team. If I would have to do the facilitation of all these simulations myself, this would almost become a full time job, which was not my aim and besides that, I didn't want to become the bottleneck for teams wanting to try this Scrumban simulation. So instead I planned a train the trainer session, for a first number of people that replied. But soon 1 session wasn't enough, so I planned a second one. And now I even planned a third and final session. By the time that the 3rd train the trainer sessions is done, approximately 25 people know how to facilitate this simulation.
But back to the session itself. The first train the trainer session was last Thursday. 8 colleagues showed up for the session. After a bit of explanation about the different elements and the rules, 2 teams started. The difference between the teams was quite striking. 1 team spent a lot of time prioritizing the backlog and planning the first iteration. They were more the thinkers, whereas the others were more the do'ers: that team managed to finish 2 iterations in the foreseen time, whereas the thinkers could only finish 1 iteration.
Again, I could capture very valuable feedback from the participants to improve the rules of the simulation. At this moment unplanned work has a priority, but there is no motivation, no driver to make sure that also unplanned work gets done in time. So we need to think about a penalty: you gain business value when you finish a product backlog item, but you can loose business value if you don't finish unplanned work in time. The only thing we can do to calculate this penalty is via the cycle time and lead time: the cost of wait as a penalty. Similarly, we can also lower the business value of a finished backlog item with the cost of wait.
My initial idea was to calculate the cost of delay once an item - be it planned or unplanned - is done and use that as a penalty. But if you only calculate the cost of delay once the item is finished - which is perfectly fine for planned work - you can leave all unplanned work items in the reported stage, without penalty. And that is not correct. So one of the participants did a very good proposal how to deal with this penalty: every day an unplanned work item is not finished, you immediately decrease the value creation according to its priority (e.g. 1 for low priority work, 5 for medium priority work and 10 for high priority work). Thank you Dominique, for the simple but effective solution, which did not require any changes to the cards. I will update the facilitators manual as soon as possible.