After attending both David J. Anderson’s Kanban Coaching Professional Masterclass (see http://edu.leankanban.com/) as well as the Kanban Leadership Retreat in Portugal a bit later, my understanding of Kanban concepts has deepened and broadened quite a bit.
Today, I’d like to share with you my understanding of the Kanban concept of options, commitment point, committed work and lead time.
In every flow of work, there is one point where work gets actually committed, i.e. there is a high certainty now that it will actually be implemented. In Kanban speak this would be at the first column with a WIP limit . In the example below the column is named “To Do”. Thus, this point is called commitment point.
From a customer’s point of view, this often comprises a more or less explicit promise that their stuff is really being dealt with now, and hopefully quickly. Abandoning work downstream should be the absolute exception. This is why it’s called committed work 😉
Everything before that point (i.e. left of it) is merely an option that can be executed (or cancelled). This area is called upstream. This originates from the metaphor of work flowing like a stream.
Likewise, everything to the right of this point is called downstream, and this is where lead time starts to tick, ending with the completed request. Note that lead time is the single key metric in a Kanban system .
I have tried to summarize this in the picture below.
I really like these terms and how they play together as a sense-making model. If you reflect on your practical experiences, you will probably agree that things tend to be managed and handled quite differently depending on whether it is upstream or downstream. Still, metrics are mixed up and lead to distorted data.
Clearly separating out the two areas (options vs. committed work) has in my view a key advantage: Lead time becomes much more predictable now. This in turn helps the delivery manager to better use this metric for a number of important things – for expectation setting with customers, release planning, improvements, etc.
On the other hand, options can be treated as what they are – possible choices of things that could be delivered. To narrow their number down to the things that are actually being implemented, you can use here a number of techniques from realms such as Product Management or Risk Management. It is very natural (and in most cases required) that a certain portion of things get dropped upstream. Upstream, different metrics would have to be used, such as queue lengths and waiting times in certain stages or even drop-out rates.
The Kanban body of knowledge is constantly evolving, and this is an area which has been explored a lot more lately. I hope this post was a good starter into the subject. Let me know if you have any questions / comments / further insights.
 Cycle time is another metric stemming from Lean in the manufacturing area. In most Kanban systems for creative knowledge work, cycle time is not applicable. In the past, I heard definitions like “Lead time starts when the customer puts their request in. Cycle time is when work at the corresponding item starts until it is delivered to the customer.” This notion is overcome by the concept outlined above.
 WIP = Work In Progress, i.e. the amount of work items that are currently being worked at in a certain state or in the whole system. A WIP limit defines the number of items that are concurrently allowed in a certain area.