Red Bead Experiment (Introduction)

Before I link the Red Bead Experiment to this blog, let me give some background information. Dr. Deming used the Red Bead experiment to demonstrate a number of points about management, namely variation, systems and ranking of employees.

By variation, Dr. Deming means that the time that it takes to do any task is not exactly predictable. It is possible to calculate an average with an upper limit and a lower limit. Imagine that you are in a tele-marketing department. You have 5 employees whose job is to call people and ask them 30 questions about your product. You realise that on average each call is 10 minutes. The upper limit is 15 minutes and the lower limit is 5 minutes.

By system, Deming means that people are working within a system. As employees, they are powerless to change the system. Only management can change the system by changing the rules. Continuing with the example of a tele-marketing department, if it takes on average 10 minutes to ask customers 30 questions; we can imagine that a person calls 30 people per day (on average). However, let say that the manager decides to change the rules. Instead of asking 30 questions, employees only have to ask 10 questions. Let's say that because of this new rule, the average duration of a call goes from 10 minutes to 5 minutes. As a consequence, the number of calls that employees make in one day jumps from 30 to 60. In effect, the productivity of the department has doubled. Clearly, employees are not working two times harder. Their output simply depends on the constraint of the system that they are working with.

Deming raises an important issues. If people are working in a system that has variation, it is not logical to rank them. The performance of the employee does not depend on their effort alone but the effort of the employee within a system. For example - sticking with the tele-marketing department - if I have 5 employees (A, B, C, D and E). If they all average calls of 10 minutes and if nobody has calls less than 5 minutes and nobody has calls of more than 15 minutes, then they are all performing within the norms determined by the management. If on Monday, employee A makes more phone calls than the rest, that does not mean that employee A was working harder than the rest. It simply means that on Monday, A was lucky. He had a few customers who answered the questions more quickly than normal. But on Tuesday, the same employee might make fewer calls than anybody else. Why? because he might have had a few customers that spent more time answering the questions than normal.

Deming argued that:
a) looking at individual performance does not matter as long as they are within the norms set by management
b) to reward some employees for "working hard" demotivates the rest becasue they know that the 'extra' performance was due to luck. They perceive this to be unfair as they worked just as hard as the person who got the reward.
c) to improve productivity, management has to change something in the system. Employees, by themselves, are powerless to change the rules.
d) many managers make the mistake of ranking employees because they have a very poor understanding of statistics. Once you understand statistics, think in terms of averages, upper limits and lower limits, your ability to manage people improves substantially.

Lastly, one warning. There are several links and - all together - the videos are quite long. The first video is OK but it gets a bit boring after a while. The last video - where the audience is told the lessons behind the experiment - is very important and very interesting. You need to be patient, warch the boring bits in order to appreciate the most important bit at the end.