Demonstrate The Value Of WIP Limits In 5 Minutes

As experts in Lean principles we talk about the benefits of managing WIP constantly, but I think often it is challenging to get people to buy into the concept. It is somewhat counterintuitive to many at first glance especially given that many tout their ability to multitask as a strength. “You’re telling me we will get more done by doing less???”. I have found that concepts like this tend to be much more relevant to folks when you provide an interactive way for them to experience it rather than just talking about it. Simple simulations are a great way of providing them that experience.

I came across this WIP limit simulation a few years ago on TastyCupcakes.org. It doesn’t seem to be there any longer so unfortunately I cannot provide credit to the author. It is a very simple, but effective simulation you can do in about 5 minutes. I use it anytime the opportunity presents itself and have had very good results with it. The only things you will need are:

• Whiteboard or piece of paper
• 3 different colored markers or pens
• A stopwatch (just use your smartphone)
• A volunteer to do the simulation

On the whiteboard or paper draw up the 3 columns with the following headings. There will be two rounds to the simulation so if you are using paper draw the 3 columns on the front and back:

Each columns represents some type of deliverable. Call them whatever you like stories, epics, initiatives, etc. In order to deliver each of them your volunteer must fill in the letters/numbers of the respective columns using the appropriate colored pen. The end result will look like this:

Round 1:

• As quickly as they can your volunteer will fill in the grid going across the 3 columns then down until they have filled in the entire grid
• They should have all three pens in their hand with caps on
• Start your stop watch as soon as they begin
• As they fully complete each of the three columns record the amount of time that has passed up to that point
• When they are done share the time it took to complete each of the three columns

Round 2:

• As quickly as they can your volunteer will fill in the grid going all the way down a column and then move on to the next column until they have filled in the entire grid
• Just like round 1 you should record the time when they finish each column and then share the times with all the participants

I think it is pretty obvious that the the times will be drastically different and that it shows the benefits of limiting WIP to deliver value sooner. What may not be as obvious just reading this is how visible this makes the impact of context switching.

The overwhelming majority of the time I have done this you could see the context switching at work. People will struggle to choose the right colored pen, write the correct number/letter, lose track of what row they are on, drop the pens, get frustrated, start swearing, etc. It makes what is normally hidden painfully visible to everyone watching as well as the person doing it. My message to people afterwards is this is not a trick. This kind of thing is happening everyday when you are working on many things at once. The only difference here is it was more visible because we have a very condensed cycle and set of tasks.

Sometimes I will add in additional elements like associating a decaying value with each column to demonstrate how this plays into the cost of delay or charge a time penalty when they make errors to account for the cost of low quality.

People will sometimes push back on the simulation and claim the lost time isn’t from context switching but because of having to take the cap on and off the pens. My response is you are partially correct. There is some context switching going on in the person’s head in taking the caps on and off, but some of it is just the additional time required to physically do the switch…also known as overhead. Not terribly dissimilar from having to attend multiple grooming meetings each week because you are working on three different teams. High WIP results in both context switching costs as well as additional overhead.

In any case I always allow people to try the simulation in different ways such as not having to take the caps on and off, using the same color for each column, having different people in the room do the simulation, or using different symbol sets in the columns. The numbers may improve somewhat, but there is still a stark contrast between doing it one row at a time and one column at time. This also demonstrates the benefit of using short cycles and continuous feedback to experiment and incrementally improve, so pushback is something I am actually hoping to get from the person/group.

I have found this to be a simple and effective way of helping people to get an initial understanding of the benefits of reducing WIP. It costs nothing so I would encourage you to give it a try. I would love to hear your feedback along with other techniques you have found to be effective!