The power of single item flow

We recently ran a lean training exercise at the office: the coin game. The coin game demonstrates the power of single item flow, in a fun and practical way!

Caroline
Process Manager & Project Manager Translations
Posted on
Reading time 5 minutes

We recently ran a lean training exercise at the office: the coin game. I just completed a Lean Green Belt certification, and this was one of the exercises we did on day one of the course. The coin game demonstrates the power of single item flow, in a fun and practical way! Given we've had a number of new colleagues join us over the last few months, running the game at the office was a great learning opportunity for them, as well as for our more experienced colleagues 😉.

The game

The first step was to assign 10 roles. You need four employees, each representing a workstation in a production line. Each employee has a manager. We also had a client, and a CEO. The four employees sat around a table. Their managers stood behind them with stopwatches. The client was waiting at the end of the production line with a stopwatch. The CEO was overseeing, also armed with a stopwatch.

The concept of the game is for the employees to flip batches of coins and pass them around the table through each workstation as fast as possible.

The following times are measured:

  • Each manager times how long it takes their employee to complete their task from start to finish at their workstation.
  • The client times how long it takes from when employee one begins, to when they receive their first flipped batch of coins at the end of the production line.
  • The CEO times from the first flip to the last batch being delivered to the client.

First round: 1 batch of 20 coins

A batch of 20 coins starts with employee one, the first in the production line. All coins start with the same side facing up. The idea is that all 20 coins must be flipped using just one hand. We used the non-dominant hands to make it more interesting! Employee one has to flip all 20 coins before passing the batch to employee two to repeat the task. Once employee two had flipped all 20 coins, they passed them all to employee three to do the same, who in turn passed the batch of 20 to employee four to flip. How long do you think it would take for all coins to reach the client?

Second round: 5 batches of 4 coins

The same game, but this time one batch is just four coins. This means that once an employee has flipped four coins, that batch can be passed to the next in line to begin. This is done five times to complete the task. How do you think that might affect the total cycle time?

Third round: single item flow

For the final round, one coin equals one batch. This means that each time a coin is flipped, it can immediately be passed to the next in line. How would that impact the timings, and by how much? Go on, take a guess!

Results

All the times that were recorded were written up on a whiteboard at the end of each round. We named the four workstations: design, build, test, and deploy. This reflects our process of delivering features pretty well! At the end of each round, we also checked for defects. That is any coins that were delivered facing the wrong way up. This is what the results (timings in seconds) looked like at the end:

Design Build Test Deploy First piece Total Defects
1 batch of 20 coins 19 21 20 18 88 91 0
5 batches of 4 coins 24 27 27 26 18 42 0
Single item flow 23 23 25 27 6 32 0

Learnings

It’s pretty apparent that smaller batches yield a significantly faster total cycle time and an incredible reduction in delivery time to the client. "We knew that single item flow is better, but the huge time difference was still surprising!" said Jeroen, Founder/CEO/CSO.

We knew that single item flow is better, but the huge time difference was still surprising

We also saw that there was less waiting time for the employees each round. By the last round, it was one fluid movement - the employees started pulling instead of waiting for a push from the workstation before.

Although individual times per workstation were lower at the start and had increased by the end, the overall first piece and total time were still far better in the later rounds.

In terms of quality, we counted zero defects in all rounds. The quality didn’t degrade, although throughput went up massively. Leading us to conclude that optimizing single stations doesn’t necessarily optimize the whole.

"It was an eye-opening experience. We all know that working together makes us complete tasks and achieve goals faster, but I was surprised to see how much the time decreased, especially for the client. We need to make our tasks smaller to finish the whole faster," said Daniela, Implementation Consultant. This was a very cool insight from one of our newest team members!

Discussions were sparked amongst participants who questioned how they could apply the concept to their own work, what they were doing and what could be done differently. Owlsome!

Conclusion

In the space of 30 minutes, new colleagues and more experienced staff experienced the power of single item flow in a fun setting that felt like a game. Simple and effective! In terms of how we work internally and deliver our product to our clients, it’s a no-brainer: working in a single item flow is best for us and our clients!