Lean Ball Game Variation

This is a variation of the Lean Ball Game I’ve used for a number of years (prior to COVID). The purpose of this variation is to highlight how teams deal with varying types of demand. The traditional lean ball game only has one type of ball and you try to optimise the process for that one workflow. However, I’ve found that in knowledge work oftentimes you’ll get different types of demand coming into the system all of the time. The volumes of each type may vary over time, plus you may introduce some work item types and retire others. I’ve found that in order for a system to be continually fit for its purposes, particularly in knowledge work, that adaptation is a necessary feature.

I was thinking about this game recently and how important it is to show participants the value of adaptability and the trade offs for optimising for variability. In manufacturing often variability can be seen as a problem – when you’re repeatedly building widgets a thousand at a time, variability results in defects and other kinds of waste. In knowledge work, on the other hand, there can be a great deal of value from the variability. Oftentimes no two work items are exactly the same – we apply the knowledge of our workers to determine how to service the need.


You’ll need to get a set of different kinds of balls. You will need at least 3 kinds of balls and at least 10 of each type (20 is better). I found a set that was made for golfing practice which works really well because the different balls move through the air differently when they are being passed around (another example of variability).

You’ll need an open space so your team can move around and adapt their system of work.

You usually need at least 30 minutes to play the game. It’s usually a good idea to set aside an hour for a good debrief as well.

Get yourself a whiteboard or some butchers paper to record the results of each round.

Running the game

General Setup

For each of the 3 types of balls, you’ll need to move them around the group in a different way. For rounds where you introduce a new type of ball, you’ll need to start by organising the people into the way to ball will flow. You can do this by different means, for example alphabetical by first name, alphabetical by last name, how long it took you to get here etc.

Each round will start with some time to get organised – first by getting the ordering correct, the second part is allowing the team some time to plan how they want to organise themselves.

You can introduce additional rules as well such as “the balls need to go up and then down between each person” or “dropped balls are considered defects and don’t count against the total, don’t pick them up” – but these are entirely at your discretion.

Then comes the “play time” where the balls will all need to travel around the group from the first person to the last – with the facilitator sending the balls into the group. Usually this is timeboxed (for example 2 or 3 minutes – depends on how many balls you have available).

At the end of the round, make sure you take a moment to review the ball / defect count at the end. Give the team time to do a quick retrospective (eg 2 minutes).

Round 1

In the first round, you’re just trying to get the group used to moving the balls around the group. Only use one type of ball for this round – this will enable them to get a base system of work started.

If you wish you might want to replay this round just to get the group used to how to move the balls around the group and to show how they can improve the process.

Round 2

In the second round we use only the second ball type. The team will need to adapt how they are organised to deal with this different work type.

Round 3

In the third round, we use both the first and second type of ball. The facilitator will need to make sure they are mixing things up when they choose which ball to send into the group. The facilitator will need to remember which person is the first for each type and send the balls in at the appropriate entry point. Also, the exit points for each type of ball will be different – so you’ll need to make sure there is a way to capture balls as the exit the system.

Round 4

The fourth round is where you introduce the third kind of ball. It is only this ball in the fourth round (not the original two types). This will let the group get used to the new work item type and how it flows around the group.

Round 5

This is where you do all 3 balls (work item types) at once in the one round. The team will need to adjust how they adapt to the different types – you may see quite a lot of defects (dropped balls). As the facilitator, once again you’ll need to mix up the order of arrival of the different types of balls, remembering that there are now 3 points of entry and exit for the balls.

If there is still time, you can try this a second time to let the team apply a retrospective and improve the process.

Debrief and learning

The point of this exercise is to show how you can design your systems of work. Some systems are designed to do particular work item types really quickly – others are designed to be adaptable. You can introduce concepts for discussion such as cost of delay – what if type 2 was Expedite and type 3 was Fixed Date – how would this change your system design? Look at how communication occurs in the system – are people talking / signalling the arrival of different types? Are there any bottlenecks / issues with how the system is designed, how could you alter the design?

When playing this game I often find in round 3, groups use their left hand for type 1 and their right hand for type 2. When it comes to round 5 they have to rethink their whole system – which is why 3 types of balls is critical for this exercise. During rounds 3 and 5 you also tend to see more defects – it’s not until a retro that teams consider how to slow things down a little to make sure quality is included.

Knowledge work often has different types of work entering the system at different and sometimes random times. The nature of work is discovery based – thus if you’ve optimised for a particular type – have you optimised in the right way or are you missing some potential value. Perhaps you need to relook at how your system is designed so that you can take advantage of some missed opportunities.