Published by:

  • Leadership – Speeches – Workshops
  • Personal Training – Consulting
  • Leadership – Marketing – Strategy

Dare the Impossible – Achieve the Extraordinary.
Vol. 13, No. 1
(626) 791-8973
Copyright © by William A. Cohen 2016


Drucker Tells Us What to do about Risk*

Jan. 2016

Risk in management is unavoidable. Some of Drucker’s clients did not acknowledge this. They thought that risk could be avoided and just wanted Drucker to tell them how. However, according to Drucker, risk could not be avoided, it could be desirable. In fact, the assumption of risk was a basic function of any successful enterprise.

Even attempting to Lower Risk Could Be the Wrong Move

Drucker saw that in attempting to lower risk, managers and professionals of all types sometimes made assumptions which could lead to disaster. The most used method of risk lowering is to assume either no change in the current situation or that the trend, whatever it is, would continue on into the future. The fallacy of either assumption can be seen almost every day as many stock market investors make one of these two assumptions with eventually poor results. Drucker knew that change is inevitable and so advised his clients.

Picking the Right Risk is Crucial

Drucker learned to analyze the situation and to emphasize taking the right risks. Because taking the wrong risk could be even more disastrous. Drucker’s investigations led to the discovery of a critical factor in the process: while deciding on the right risk: one had to institute controls in the actions involved in taking the risk.  Otherwise, even though it might well be a right risk to take, there might be other considerations and more important risks such that in the end actions might be mismanaged and intended objectives not reached with failing results even though taking a risk was of some sort was fully understood and accepted.

Risk Controls and Their Characteristics

Drucker found that all controls invariably have three basic characteristics:

  1. Risk controls won’t be objective and they won’t be neutral
  2. While they need to focus on the real results, the “real” results aren’t always possible to control
  3. They are needed for both measurable and events which cannot be measured or cannot be managed without difficulty

The Near Impossibility of Objectivity and Neutrality

From 1924–1932 a study of lighting conditions was done at the Hawthorne Works, a Western Electric factory outside of Chicago. One experiment was to examine the effect on productivity of better lighting. It sounds simple enough. The control was based on increasing wattage of the light bulbs weekly and then noting the results in productivity.  It was expected that productivity would increase as lighting got better every week, and sure enough, it did.

However, one week some joker decreased instead of increasing the lighting intensity. Guess what? Productivity improved anyway. It was hardly a miracle or an error in measurement. What had happened was that workers expected the lighting intensity to increase and this motivated them to work harder, more efficiently, and hence more productively. Today this is known today as the “Hawthorne Effect.” It demonstrates that the novelty alone of having research conducted along with the increased attention to the measurement results can cause at least a temporary increase in productivity. And it means that as Drucker said, “Controls are not applied to a falling stone, but to a social situation with living, breathing, human beings who can and will be influenced by the controls themselves.”

Focusing on the Real Results

It is relatively easy to measure effort or efficiency, but much more difficult to measure real results with a control. Drucker explained that it was of no value to have the most efficient engineering department if the department was efficiently designing the wrong products. Drucker also differentiated leadership from management. Others have since similarly adopted his formula:  that management was doing things right – read efficiently – while leadership was doing the right things, that is, the effective things to get the job done. So one could be very efficient as a manager and still fail as a  leader.

Measuring efficiency is usually not so difficult. For example, one can measure the number of times a leader recognizes subordinates for accomplishing good work. That’s considered a sign of good leadership. However “good work” can be done on the wrong, as well as the right things. Maybe salespeople are doing a marvelous job selling the wrong product. Is this “good work?” To complicate controls further, how do we know what are “the right things.” This is much harder to know when so many factors and multiple humans are involved. Since leadership is an art, in observation, its quality may most definitely be in the eyes of the beholder.

Non-measurable Events, Too

Controls are also difficult because some events in an organization, important to risk, simply aren’t measurable, results or no. We already noted that you don’t have true facts about the future. You don’t know what may suddenly happen on the way to the future, either. The ubiquitous slide rule, once on the person of every engineer worthy of the name, disappeared almost overnight when the pocket electronic calculator hit the market in 1972 – 73.

The 7 Control Specifications

Drucker investigated further and determined that all controls must satisfy certain specific specifications, of which there were seven:

  1. They had to be economical – the less effort required to gain control, the better.
  2. They had to be meaningful – in other words they had to be intrinsically significant or symptoms of significant developments
  3. They must be appropriate to the nature of what you are measuring – absenteeism of a yearly average of 10 days per employee sounds acceptable, but what if you have only two employees and one was never absent and all absences due to the other?
  4. Measurements must be congruent to the phenomenon measured. As a writer I’m always interested in book sales. I once read a book by a famous entrepreneur who had written a best seller. He had bought a well-known company and promoted the company’s product frequently on television stating that “he liked the product so much, that he had bought the company.” His advertising promotions were great. However, when I read the book, I found it was at best fair. Yet it became a bestseller and sold 2,000,000 copies, surpassing many better books on entrepreneurship in sales, including Drucker’s Innovation and Entrepreneurship which came out at about the same time. Not to mention several that I had written, the best of which had sold a little less than 100,000 copies.

One day it was revealed that the author had spent almost $2,000,000 of his own money promoting the book. Now book royalties are a lot less than you might think, frequently they are about 15% of the net amount received by the publisher. The net to the publisher may be only roughly 50% of the price of the book.  Since this entrepreneur was not the publisher, I estimated that he had earned about $1,000,000 in royalties for his total book sales. So he personally lost a million dollars in selling his book. He may have been willing to pay a million dollars to have a 2,000,000 copy best seller just for bragging rights. But as a control measurement, book sales alone, the most frequent tool used to measure public demand for a particular book by readers is probably a poor tool without other factors being noted and compared.

Control requirements  5, 6, and 7 are much easier and intuitive. They need to be timely. They are an expensive waste of time if the information received arrives too late to be of use. They need to be simple. As Drucker noted, complicated controls just don’t work. They frequently cause confusion and lead to other errors. Finally they need to be oriented toward action. Controls are not something instituted for academic interest. They are for implementing strategy once actions with the right risks are chosen

The Final Limitation

The final limitation on controls is the organization itself. An organization operates with rules, policies, rewards, punishments, incentives and resources, capital equipment, etc. but its success comes from people and their daily, frequently unquantifiable, actions. The expressions of their actions, such as an increase in salary, may be quantifiable. However, their feelings, motivations, illnesses, drive, ambitions, etc. are not.  As an operational system the organization cannot be accurately quantified.

What It All Means

Risk is essential and as a consultant, and this is what Drucker taught his clients. They key is to pick the right risks and then to control these risks considering the many factors which make this control so difficult to comprehend and use. But knowledge is power, or at least stored power. Selecting the right risks and monitoring the seven important aspects of the risk controls identified by Drucker means effective risk management. One cannot do more, nor should the consultant consider doing less.

*Adapted from Peter Drucker’s Consulting Principles to be published by LID in 2016.

“If you want risk taking, set an example yourself and reward and praise those that do” – Jack Welch, Former CEO of General Electric

Recent Linked Articles by Dr. Cohen, not Published in the Journal of Leadership Applications:


TO CONTACT US: CALL 1(626)350-1500 Ext 102 or 1(626)390-1364 or

Yes, we do give international seminars — The U.S. country code is 01.