The W. Edwards Deming Institute Blog

Dr. Deming Called for the Elimination of The Annual Performance Appraisal


In Out of the Crisis, page 101, Dr. Deming states the following as one of the seven deadly diseases:

Evaluation of performance, merit rating, or annual review… The idea of a merit rating is alluring. the sound of the words captivates the imagination: pay for what you get; get what you pay for; motivate people to do their best, for their own good. The effect is exactly the opposite of what the words promise.

Dr. Deming understood the appeal of evaluation of performance. He just judged what actually went on in the world and saw that the appeal was not matched by success in practice. Dr. Deming understood the importance of subjecting theories to evaluation within the system those theories were being applied to adopt practices. The most common was of testing beliefs was to use the PDSA cycle to experiment and determine if beliefs were confirmed.

In the introduction to the Team Handbook Dr. Deming wrote:

The fact is that the system that people work in and the interaction with people may account for 90 or 95 percent of performance.

In such a situation you then would have to measure most importantly the interaction with others and the system to evaluate someone. Most often there is either no importance placed on interactions, or their is a very minor (especially compared to the importance) placed on interactions.

From page 296 of the Leader’s Handbook:

Someone in the audience asked Dr. Deming: “if we eliminate performance appraisals, as you suggest, what do we do instead?” Dr. Deming’s reply: “Whatever Peter Scholtes says.”

In Total Quality or Performance Appraisal: Choose One, Peter Scholtes says:

Improvement efforts should focus on systems, processes, and methods, not on individual workers. Those efforts that focus on improving the attentiveness, carefulness, speed, etc., of individual workers — without changing the systems, processes, and methods — constitute a low-yield strategy with negligible short-term results

Conventional problem-solving would ask such questions as: Whose area is this? Who is supposed to replace worn gaskets? We don’t ask “why,” we ask “who.” We don’t look for causes in the system, we look for culprits in the work force. Performance appraisal is a “who-based” approach to problem-solving.

This topic is often one people have trouble accepting at first. And it is one many people have strong opinions about. Please share your comments below.

Categorised as: psychology, respect for people, systems thinking, theory of knowledge


  1. Sean Stickle says:

    “The fact is that the system that people work in and the interaction with people may account for 90 or 95 percent of performance.”

    This is a strong claim with fairly specific numbers attached to it. What evidence do we have that this is true in practice, and that it is true in multiple industries and businesses? What does 90% of my performance as an IT manager even mean — what is being measured here?

  2. John Hunter says:

    I do not know of any specific evidence. Dr. Deming states it this way on page 315 of Out of the Crisis:

    I should estimate that in my experience most troubles and most possibilities for improvement add up to the proportions something like this: 94% belongs to the system (responsibility of management), 6% special.

    as quoted in

    I do not believe the statement “people may account for 90 or 95 percent of performance…” was based on a set of data but instead believe it was based on Dr. Deming’s decades of experience. And he increased the percentage over time – as he learned more.

    I think he was trying to give additional weight to saying something like the system accounts for most… (or something similar). I do think using the figures is not ideal as it leads to the next question raised is what the operational definitions used to collect this data are. But that is my opinion, Deming did use the figures and maybe there is a better explanation that I am not aware of. The quote from Out of the Crisis I think provides evidence he used similar statements when he clearly was not basing the statement on a set of data. I welcome others thoughts on this topic.

    I do think looking at the total of Deming’s work that the point he is trying to make is that looking to blame people is not a good strategy for improvement. The person’s individual impact on results is small in comparison to the system within which they work. So, focus on improving the system; which will achieve better results than searching for what people did wrong.

  3. John Dowd says:

    Relative percentages will vary from situation to situation.

    What is being measured is the amount of variation in output is due to individual differences and efforts and the amount that is due to factors over which individuals have no control. Juran estimated this to be 80%. Deming generally used higher numbers. The point is that this amount is not zero.

    Now we have the problem of trying to solve an equation with two unknowns and only one equation (let A equal to total of individual contribution and B equal to the total of system contribution). As Deming would say, “Even an eight grader can see the problem.” The individual contribution cannot be separated from the system contribution. That is, you cannot solve for two unknowns with only one equation.

    A ‘simplifying’ assumption is to say, “Well if all other things are equal….” , then individual contribution can be ascertained. But, not really. First, like many simplifying assumptions, this one does not hold. “All other things” are never equal. Further, there is the matter of interactions between the individual and the system within which he or she works. This reality adds a third factor to our equation making it even more impossible to separate the individual contribution from the system.

    Of course, the amount of this interaction will vary from situation to situation and from individual to individual. But what should be clear by now is the impossibility of separating those factors in output due to individual differences and those do to the underlying system itself (of which, as we have said, the individuals are a part.

    Hope this helps.

  4. [...] I tend to side with W. Edwards Deming and point 3 of ‘The Seven Deadly Diseases of Management”: [...]

  5. [...] it is important to note Dr. Deming’s argument that no team or individual is to blame for such deficiencies, and that we can improve lead times by reducing batch sizes, it is the separation of authority [...]

  6. [...] Dr. Deming’s quote from the introduction to the Team Handbook [...]

  7. Steven says:

    Look at the Red Bead Experiment. It has been repeated several times and demonstrates that the system – how work is done influences outcome more than individual ability.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>