Tag: Herbert Simon

Herbert Simon on Why the Principles Of Good Management are not Widely Practiced

Herbert Simon, a Nobel Prize laureate and polymath, offered many contributions to the world in fields such as computer science/artificial intelligence, cognitive psychology, economics, and management.

This brief excerpt, taken from the his remarkable autobiography offers some timeless wisdom.

The principles of good management are simple, even trivial. They are not widely practiced for the same reason that Christianity is not widely practiced. It is not enough to know what the principles are; you must acquire deeply ingrained habits of carrying them out, in the face of all sorts of strong urges to stray onto more comfortable and pleasant paths, to respond without inhibition to provocations, and just to goof off. “

The principles of good management are simple but not easy. They require discipline and focus to maintain at the best of times. When things get tough we are further tested.

One thing that comes to mind for me is focus and the ability to say no. In a corporate setting the hardest thing in the world can sometimes be to say no to a good idea. Steve Jobs focused on saying no. And Warren Buffett commented on it when talking about the difference between successful people and very successful people. It’s easy to say yes. And yet we must stay away from the urges to say yes to everything. You need self discipline.


Still Curious? Simon literally wrote the book on Administrative Behavior.

Herbert Simon on the Distinction Between What is Legal and What We Will Tolerate

You’d break the law. In fact most of us would. How can I say this with near certainty? Because if you were put in a position where the ends justified the means, the means would become acceptable.

The person who steals bread so his starving child can eat is an easy one to sympathize with. While illegal, most of us understand and even tolerate this petty behaviour.

However, would we tolerate bigger transgressions? For the answer to this question we turn to Nobel Prize winning social scientist Herbert Simon. Simon’s contributions to our growing repository of wisdom include why the principles of good management are not often followed, the role of intuition in experts, and why organizational planning is doomed to failure.

In his autobiography Models of My Life, Simon touches on the difference between what a society will tolerate and its laws.

A revolution aims at bringing about fundamental changes in institutions by employing illegal tactics. What is legal and what a society will tolerate are distinct. When there is sympathy for ends, illegal means may become acceptable and the laws against them unenforceable.

Fundamentally we seem to believe that if the means warrant the ends, they will be accepted. Killing to overthrow a dictator, which is obviously against the law, becomes acceptable at a certain point if the dictator is too horrible. The question of when it becomes acceptable, however, while easily distinguishable for the edge cases, becomes grey in the middle.

Simon continues:

If a revolution aims at overthrowing an entire legal system, the role of the illegal action is to arouse an already sympathetic population; to goad the defenders of the legal system to severity that will arouse additional sympathy; to demonstrate strength, hence to reduce fear fo the authorities and to increase fear of the revolutionaries; and finally to seize weapons and strong points. When people no longer believe that the existing laws can be enforced, the first half of the revolution has been won. There remains the task of securing for it the “right” party. This has been the common point of failure for the moderates.

There are also situations where the laws are better than what the government will enforce. Something to think about.

Herbert Simon: On Experts and Intuition

The brain is not a linear processing machine taking a bit of input and systematically processing it. Rather we are endowed with a brain that pattern matches.

Dr. Ralph Greenspan says(1):

In no sense does the brain work like a computer. Computers record, and computers have things stored in specific places that are stable. Our brains do none of that. When the great chess master Gary Kasparov lost to Big Blue everybody said “Aha, this machine can think!” Big Blue was not thinking. Big Blue was simply replaying the entire history of chess. That’s not the way that Gary Kasparov or any human being plays chess. We do pattern recognition. Even though we are capable of logic, our brain does not operate by the principles of logic. It operates by selection of pattern recognition. It’s a dynamic network. It’s not an “if-then” logic machine.

When computers play chess they are effectively looking at all the moves available and working out (n) levels deep. While computers have surpassed us in the ability to play chess, largely through sheer brute force, there is an important point that’s worth noting. For a while we kept up and even outplayed them, despite their dominance in processing power.

Think about it. Computers could do millions, even billions of computations per second. Their computational power was awesome. And for a time it meant nothing.


An excerpt from an awesome article by Nobel Laureate Herbert Simon offers some insight (2):

We have seen that a major component of expertise is the ability to recognize a very large number of specific relevant cues when they are present in any situation, and then to retrieve from memory information about what to do when those particular cues are noticed.

Because of this knowledge and recognition capability, experts can respond to new situations very rapidly- and usually with considerable accuracy. Of course, on further thought, the initial reaction may not be the correct one, but it is correct in a substantial number of cases and is rarely irrelevant. Chess grandmasters, looking at the chessboard, will generally form a hypothesis about the best move within five seconds, and in four out of five cases, this initial hypothesis will be the move they ultimately prefer. Moreover, it can be shown that this ability accounts for a very large proportion of their chess skill. For, if required to play very rapidly, the grandmaster may not maintain a grandmaster level of play but will almost always maintain a master level, even though in rapid play there is time for almost nothing but to react to the first cues that are noticed on the board.

We usually use the word “intuition” – sometimes also “judgment” or even “creativity” – to refer to this ability of experts to respond to situations in their domains of expertise almost instantaneously and relatively accurately. The streetwise slum resident has good intuition about how to react to the situations that are often encountered in a slum environment. The manager has good intuition about how to react to the situations that are often encountered in organizations. Both skills have the same basis in knowledge and recognition capability.

This is why we use a two step process for making effective decisions.

It’s also why the best thing you can do during an interview is to distinguish between the two types of knowledge. Elon Musk has a pretty good system for determining whose lying and whose not.

If you’re interested in learning more about Herbert Simon—and you should be— I recommend reading Models of My Life.

Also see Solution by Recognition and Choice Under Uncertainty.


(1) Seeking Wisdom
(2) How Managers Express their Creativity, Autumn 1986, The McKinsey Quarterly

3 Things Everyone Should Know About the Availability Heuristic

There are 3 things you should know about the availability heuristic:

  1. We often misjudge the frequency and magnitude of events that have happened recently.
  2. This happens, in part, because of the limitations on memory.
  3. We remember things better when they come in a vivid narrative.


There are two biases emanating from the availability heuristic (a.k.a. the availability bias): Ease of recall and retrievability.

Because of the availability bias, our perceptions of risk may be in error and we might worry about the wrong risks. This can have disastrous impacts.

Ease of recall suggests that if something is more easily recalled in memory it must occur with a higher probability.

The availability heuristic distorts our understanding of real risks.

“The attention which we lend to an experience is proportional to its vivid or interesting character; and it is a notorious fact that what interests us most vividly at the time is, other things equal, what we remember best.”

— William James

When we make decisions we tend to be swayed by what we remember. What we remember is influenced by many things including beliefs, expectations, emotions, and feelings as well as things like frequency of exposure.  Media coverage (e.g., Internet, radio, television) makes a big difference. When rare events occur they become very visible to us as they receive heavy coverage by the media. This means we are more likely to recall it, especially in the immediate aftermath of the event. However, recalling an event and estimating its real probability are two different things. If you’re in a car accident, for example, you are likely to rate the odds of getting into another car accident much higher than base rates would indicate.

Retrievability suggests that we are biased in assessments of frequency in part because of our memory structure limitations and our search mechanisms. It’s the way we remember that matters.

The retrievability and ease of recall biases indicate that the availability bias can substantially and unconsciously influence our judgment. We too easily assume that our recollections are representative and true and discount events that are outside of our immediate memory.


In Thinking Fast and Slow, Kahneman writes:

People tend to assess the relative importance of issues by the ease with which they are retrieved from memory—and this is largely determined by the extent of coverage in the media.


Nobel Prize winning Social Scientist and Father of Artificial Intelligence, Herbert Simon, wrote in Models of My life:

I soon learned that one wins awards mainly for winning awards: an example of what Bob Merton calls the Matthew Effect. It is akin also to the phenomenon known in politics as “availability,” or name recognition. Once one becomes sufficiently well known, one’s name surfaces automatically as soon as an award committee assembles.

* * *

According to Harvard professor Max Bazerman

Many life decisions are affected by the vividness of information. Although most people recognize that AIDS is a devastating disease, many individuals ignore clear data about how to avoid contracting AIDS. In the fall of 1991, however, sexual behavior in Dallas was dramatically affected by one vivid piece of data that may or may not have been true. In a chilling interview, a Dallas woman calling herself C.J. claimed she had AIDS and was trying to spread the disease out of revenge against the man who had infected her. After this vivid interview made the local news, attendance at Dallas AIDS seminary increased dramatically. Although C.J.’s possible actions were a legitimate cause for concern, it is clear that most of the health risks related to AIDS are not a result of one woman’s actions. There are many more important reasons to be concerned about AIDS. However, C.J.’s vivid report had a more substantial effect on many people’s behavior than the mountains of data available. The Availability Heuristic describes the inferences we make about even commonness based on the ease with which we can remember instances of that event

While this example of vividness may seem fairly benign, it is not difficult to see how the availability bias could lead managers to make potentially destructive workplace decisions. The following came from the experience of one of our MBA students: As a purchasing agent, he had to select one of several possible suppliers. He chose the firm with whose name was the most familiar to him. He later found out that the salience of the name resulted from recent adverse publicity concerning the firm’s extortion of funds from client companies!

Managers conducting performance appraisals often fall victim to the availability heuristic. Working from memory, vivid instances of an employee’s behavior (either positive or negative) will be most easily recalled from memory, will appear more numerous than commonplace incidents, and will therefore be weighted more heavily in the performance appraisals. The recency of events is also a factor: Managers give more weight to performance during the three months prior to the evaluation than to the previous nine months of the evaluation period because it is more available in memory.

* * *

There are numerous implications for availability bias for investors.

A study by Karlsson, Loewenstein, and Ariely (2008) showed that people are more likely to purchase insurance to protect themselves after a natural disaster they have just experienced than they are to purchase insurance on this type of disaster before it happens.

Bazerman adds:

This pattern may be sensible for some types of risks. After all, the experience of surviving a hurricane may offer solid evidence that your property is more vulnerable to hurricanes than you had thought or that climate change is increasing your vulnerability to hurricanes.

Robyn M. Dawes, in his book Everyday Irrationality, says:

What is a little less obvious is that people can make judgments of the ease with which instances can come to mind without actually recalling specific instances. We know, for example, whether we can recall the presidents of the United States–or rather how well we can recall their names; moreover, we know at which periods of history we are better at recalling them than at which other periods. We can make judgments without actually listing in our minds the names of the specific presidents.

This recall of ease of creating instances is not limited to actual experience, but extends to hypothetical experience as well. For example, subjects are asked to consider how many subcommittees of two people can be formed from a committee of eight, and either the same or other subjects are asked to estimate how many subcommittees of six can be formed from a committee of eight people. It is much easier to think about pairs of people than to think about sets of six people, with the result that the estimate of pairs tends to be much higher than the estimate of subsets of six. In point of logic, however, the number of subsets of two is identical that of six; the formation of a particular subset of two people automatically involves the formation of a particular subset consisting of the remaining six. Because these unique subsets are paired together, there are the same number of each.

This availability to the imagination also creates a particularly striking irrationality, which can be termed with the conjunction fallacy or compound probability fallacy. Often combinations of events or entities are easier to think about than their components, because the combination might make sense whereas the individual component does not. A classic example is that of a hypothetical woman names Linda who is said to have been a social activist majoring in philosophy as a college undergraduate. What is the probability that at age thirty she is a bank teller? Subjects judge the probability as very unlikely. But when asked whether she might be a bank teller active in a feminist movement, subjects judge this combination to be more likely than for her to be a bank teller.

* * *

Retrievability (based on memory structures)

We are better at retrieving words from memory using the word’s initial letter than a random position like 3 (Tversky & Kahneman, 1973).

In 1984 Tverksy and Kahneman demonstrated the retrievability bias again when they asked participants in their study to estimate the frequency of seven-letter words that had the letter “n” in the sixth position. Their participants estimated such words to be less common than seven letter words ending in the more memorable “ing”. This response is incorrect. All seven letter words ending with “ing” also have an “n” in the sixth position. However, it’s easy to recall seven letter words ending with ing. As we demonstrated with Dawes above, this is another example of the conjunction fallacy.

Retail locations are chosen based on search as well, which explains why gas stations and retail stores are often “clumped” together. Consumers learn the location of a product and organize their mind accordingly. While you may not remember the name of all three gas stations on the same corner, your mind tells you that is where to go to find gas. Each station, assuming all else equal, then has a 1/3 shot at your business which is much better than gas stations you don’t visit because their location doesn’t resonate with your minds search. In order to maximize traffic stores must find locations that consumers associate with a product.

* * *

Exposure Effect

People tend to develop a preference for things because they are familiar with them. This is called the exposure effect. According to Titchener (1910) the exposure effect leads people to experience a “glow or warmth, a sense of ownership, a feeling of intimacy.”

The exposure effect applies only to things that are perceived as neutral to positive. If you are repeatedly exposed to something perceived as a negative stimuli it may in fact amplify negative feelings. For example, when someone is playing loud music you tend to have a lot of patience at first. However, as time goes on you get increasingly aggravated as your exposure to the stimuli increases.

The more we are exposed to something the easier it is to recall in our minds. The exposure effect influences us in many ways. Think about brands, stocks, songs, companies, and even the old saying “the devil you know.”

* * *

The Von Restorff Effect

“One of these things doesn’t belong,” can accurately summarize the Von Restorff Effect (also known as the isolation effect and novelty effect). In our minds, things that stand out are more likely to be remembered and recalled because we give increased attention to distinctive items in a set.

For example, if I asked you to remember the following sequence of characters “RTASDT9RTGS” I suspect the most common character remembered would be the “9” because it stands out and thus your mind gives it more attention.

The Von Restorff Effect leads us to Vivid evidence.

* * *

Vivid Evidence

According to William James in the Principles of Psychology:

An impression may be so exciting emotionally as to almost leave a scar upon the cerebral tissues; and thus originates a pathological delusion. For example “A woman attacked by robbers takes all the men whom she sees, even her own son, for brigands bent on killing her. Another woman sees her child run over by a horse; no amount of reasoning, not even the sight of the living child, will persuade her that he is not killed.

M. Taine wrote:

If we compare different sensations, images, or ideas, we find that their aptitudes for revival are not equal. A large number of them are obliterated, and never reappear throughout life; for instance, I drove through Paris a day or two ago, and though I saw plainly some sixty or eighty new faces, I cannot now recall any one of them; some extraordinary circumstance, a fit of delirium, or the excitement of hashish would be necessary to give me a chance at revival. On the other hand, there are sensations with a force of revival which nothing destroys or decreases. Though, as a rule, time weakens and impairs our strongest sensations, these reappear entire and intense, without having lost a particle of their detail, or any degree of their force. M. Breirre de Boismont, having suffered when a child from a disease of the scalp, asserts that ‘after fifty-five years have elapsed he can still feel his hair pulled out under the treatment of the ‘skull-cap.’–For my own part, after thirty years, I remember feature for feature the appearance of the theater to which I was taken for the first time. From the third row of boxes, the body of the theater appeared to me an immense well, red and flaming, swarming with heads; below, on the right, on a narrow floor, two men and a woman entered, went out, and re-entered, made gestures, and seemed to me like lively dwarfs: to my great surprise one of these dwarfs fell on his knees, kissed the lady’s hand, then hid behind a screen: the other, who was coming in, seemed angry, and raised his arm. I was then seven, I could understand nothing of what was going on; but the well of crimson velvet was so crowded, and bright, that after a quarter of an hour i was, as it were, intoxicated, and fell asleep.

Every one of us may find similar recollections in his memory, and may distinguish them in a common character. The primitive impression has been accompanied by an extraordinary degree of attention, either as being horrible or delightful, or as being new, surprising, and out of proportion to the ordinary run of life; this it is we express by saying that we have been strongly impressed; that we were absorbed, that we could not think of anything else; that our other sensations were effaced; that we were pursued all the next day by the resulting image; that it beset us, that we could not drive it away; that all distractions were feeble beside it. It is by force of this disproportion that impressions of childhood are so persistent; the mind being quite fresh, ordinary objects and events are surprising…

Whatever may be the kind of attention, voluntary or involuntary, it always acts alike; the image of an object or event is capable of revival, and of complete revival, in proportion to the degree of attention with which we have considered the object or event. We put this rule into practice at every moment in ordinary life.

An example from Freeman Dyson:

A striking example of availability bias is the fact that sharks save the lives of swimmers. Careful analysis of deaths in the ocean near San Diego shows that on average, the death of each swimmer killed by a shark saves the lives of ten others. Every time a swimmer is killed, the number of deaths by drowning goes down for a few years and then returns to the normal level. The effect occurs because reports of death by shark attack are remembered more vividly than reports of drownings.

Availability Bias is a Mental Model in the Farnam Street Mental Model Index

The Inevitable Failure of Organizational Planning

A beautiful excerpt from Herbert Simon’s Strategy and Organizational Evolution:

Anticipating the future means detecting, preferably prospectively,novel features in the environment that may affect the firm significantly in the future, and determining at what point in time attention should be focused on them and energy devoted to dealing with them. The available management time and attention is never sufficient to deal with all the contingencies that may arise; relative priorities for attention, planning and action need to be revised continually. A major function of strategic planning is to conserve scarce managerial, engineering and science attention for the things that matter. …

Large organizations are plagued by a kind of Gresham’s Law: the pressures of everyday activities and of crises drives out planning. Short-term concerns create priorities and deadlines that absorb managerial attention and energy at the expense of long-range concerns. The obvious remedy for this universal problem is to create special organizational units whose sole responsibility is to handle various facets of the strategic planning activity. With such a definition of its base function, such a unit has no excuse to be diverted.


There are two reasons why creating specialized planning units does not automatically repeal Gresham’s Law of Planning. First, if the planners are sagacious and effective, they are sure to experience increasing demands on their time for advisory services to top management and to other divisions across the organization. They will have to be very strong-minded indeed to ward off these interruptions of their mainstream activities, and not to find themselves increasingly entangled in short-run deadlines.


Second, the more tightly a planning group is sealed off from the day-to-day affairs of an organization, the more difficult it becomes for its plans to influence company operations. Participation of many organizational members in the strategic planning process is the surest way of securing the dissemination of ideas that is the basis for implementation; and isolation of the planning activity from the rest of the organization greatly complicates the process of dissemination, among other reasons because those who have not participated in the process will find it hard both to understand and to accept its product.


The problem of isolation of the planning unit arises not only in general strategic planning but in the development and introduction of new products as well. One of the advantages that Japanese manufacturing firms have exploited in their international competition is the speed with which ideas for new products or product improvements are converted into actual production lines. Observation has shown that this speed is possible because manufacturing engineers and even sales engineers participate in the design process almost from the beginning. This enables design engineers to take manufacturing constraints and convenience into account in the designs. Conversely, it involves the manufacturing staff in the product from the outset, secures their commitment to it, and enables them to begin planning the manufacturing operations while design is still going on. At the same time, of course, this procedure does shift some of the attention of design engineers to manufacturing problems and perhaps creates a danger that they will be diverted from their basic responsibility for product innovation.


As in most matters of organization, what is called for, in order to secure a proper attention to strategic planning but at the same time maintaining strong communication links between planning and operating units, is balance, and top management attention to maintaining that balance. On the whole, it is probably unwise to allow specialists to make their whole careers in planning, while others spend their carers in line responsibilities. Some rotation, even at some short-run expense to expertise, can do a great deal to disseminate the products of strategic planning, while keeping planning units in touch with the realities of the world of operations.


Institutionalizing intelligence activities
How can a firm organize so that it will scan the horizons with sufficient vigor, identifying potential problems and potential opportunities? It is no accident that the eyes and ears are located on the surface of the body and not in its interior. Intelligence requires continual contact with the relevant environments, and in the case of business firms two of the most relevant and important environments are the end-use (customer) environment and the science and technology (research and development) environment. Other parts of the firm should not be excluded from the search for information, but these are perhaps the two most important in it.

The marketing function is not simply a function of selling and distributing products to customers. It is equally a function of acquiring, through contact with the end-use environment, information about the future of the firm’s markets and of markets into which it might enter. Salesmen and sales engineers may play an important role in this intelligence activity, but only to the extent that it is an explicit part of their function, they are trained to do so and they are linked effectively in communication with top management, planning and design units. Specialized units may also provide various kinds of intelligence—products of customer polls, for example. I shall not attempt to describe in detail how one organizes intelligence about the end-use environment, but simply call attention to its importance.


Research and development organizations are commonly thought of as aimed at inventing and developing new products, but this is only a small part of the function they should perform. Just as sales organizations are windows on the world of customers, so R&D organizations are (or should be) windows on both the world of nature and the scientific and engineering communities that are engaged in examining nature. A scientist or R&D engineer is not merely an inventor but also a channel of communication between the company and the discipline in which he or she works.


Scientific disciplines operate as huge blackboards (collections of journals and books, professional meetings) on which the discoveries of all of the participants are recorded. Only a tiny fraction of the knowledge that any one scientist holds, including very new knowledge, was produced by his or her research or the research of the home laboratory. Most of it was read off the blackboard. Thus, the knowledge that a company R&D department produces should be not only the product of its own laboratory effort, but also the whole body of relevant knowledge that it obtains from the blackboards belonging to is professional domains.


Too great a preoccupation with the patentable products invented by the R&D department will obscure its broader intelligence function, and restrict its contribution to the strategic planning effort. The NIH (‘not invented here’) syndrome is endemic in R&D organizations that do not understand that their intelligence responsibilities extend far beyond their laboratory research programs.


Companies in industries where there is a rapid turnover of products (clothing and pharmaceuticals are salient examples) usually take conscious pains to organize their intelligence activities. If they did not, they would not survive long. They are prime examples of organizations where success at a particular time will be short-lived unless it is followed continually by new successes. To the extent that they can patent new products or develop brand-name loyalties, they may be able to buffer to some extent the volatility of their environments. But basically they live, not by making isolated innovations, but by organizing to produce a steady stream of innovations. Without strongly developed intelligence capabilities, they are unlikely to be successful at this.


Strategic planning is aimed at dealing with the enormous uncertainty and constant change that modern organizations find in the environments to which they must adapt. A market ‘niche’ is typically a transient thing. Statistics of firm growth show that special firm advantages typically have half-lives measured in a few years rather than in decades or generations. The task of strategic planning is to assure a stream of new ideas that will allow the organization to continue to adapt to its uncertain outside world.


In my remarks, I have tried to view the strategic planning function in its broader setting of the whole decision-making process in an organization. Making choices and evaluating them are simply the final stages in the decision process, and seldom the most important stages. Before choices are made, the occasions for choice must be identified, effort must be focused on problems or opportunities and possible courses of action must be designed. Classical decision theory has relatively little to say about these crucial initial stages of decision but students of strategic planning have become increasingly aware that problem identification and alternative generation are crucial components of strategy. Cognitive science and artificial intelligence have learned a great deal about these processes in recent years, and are becoming important sources of ideas for planning theory and practice.


Strategic planning will not happen by itself, or even if we simply set up organization units formally charged with doing it. The planning effort will be effective only to the extent that it permeates the entire organization and only if its products are disseminated effectively. A central idea (call it a ‘mission’ or a ‘company goal’ or ‘basic principles’), embedded in many heads where it is evoked on the occasion of decisions, is more crucial than an elaborate written list of things that are somehow supposed to happen. Organizational identification, essential to the implementation of strategic plans, is a good deal more than simple loyalty. Identification implies absorption of strategic plans into the minds of organization members where they can have direct effect upon the entire decision-making process, starting with the identification of problems, continuing with the design of alternative courses of action, and leading ultimately to effective implementation.

If you’re interested in learning more about Herbert Simon, I recommend reading Models of My Life. Also, if you liked this article you’ll like Solution by Recognition and Choice Under Uncertainty.

Herbert Simon: Scientific Research has much in Common with Successful Stock-Market Investment

This is a fascinating expert from How Managers Express Their Creativity” by Herbert Simon that deals with information flow within grounds and organizations. Simon compares successful scientific research with successful stock-market investing. Both scientists and investors are looking for mis-priced bets

In this respect, successful scientific research has much in common with successful stock-market investment. Information is only valuable if others do not have it or do not believe it strongly enough to act on it. The investor pits his knowledge, beliefs and guesses against the knowledge, beliefs and guesses of others.

In neither domain—science or the stock market—is the professional looking for a “fair bet.” On the contrary, he or she is looking for a situation where superior knowledge—knowledge not yet available to others—can be made, with some reasonable assurance, to pay off. Sometimes that superior knowledge comes from persistence in acquiring more “chunks” than most others have. Sometimes it comes from the accidents that have already been mentioned. But whatever its source, it seldom completely eliminates the element of risk. Investors and scientists require a “contrarian” streak that gives them the confidence to pit their knowledge and judgment against the common wisdom of their colleagues.

If you’re interested in learning more about Herbert Simon, I recommend reading Models of My Life. I’ve also written about Simon before (here, and here).