I feel so good, I just got my bonus. It’s nice of the man, the company to give me something for doing a good job. Plus, I got an awesome performance review — I can see it there on the whiteboard. I am the data point at the very right end of the bell curve, just like when mom gave me something for being a good boy. I scored especially high in the hearts category. I mean after all I achieved all my goals. Performance looks really good this year. Oh, there we go, another treat. My name is Max and I’m a black labrador.

The Board Room Version: Bonuses, KPIs and performance reviews are based on behaviorism — a school of thought within psychology which has been overtaken by research. It only works for simple, repetitive tasks — not for knowledge work. It uses rewards which Kohn says easily become punishment, destroy teamwork, ignore root causes and discourage risk taking. Most importantly rewards destroy intrinsic motivation for work. Herzberg shows that hygiene factors like salary, bonuses and free coffee can only demotivate if they are missing. Work itself is the only real source for motivation. Deci & Ryan and Csikszentmihalyi explain that Autonomy, Competence, Connectedness and Flow are the factors that drive us. True motivation comes from well designed jobs, not from treats. Passive employees are not personal failure but the result of absurd jobs.

Yeah, yeah we are all dogs or rats, very funny. We are all animals in a corporate lab. The thing is, it’s really not funny at all. Some of it comes down to misunderstood science, some to people not keeping up, and some to plain ethics. 

We usually don’t do recurring cameos in this show, but I have to bring back the gunslinging smoking corner manager for this one. You know the one making decisions out of intuition which is not intuition at all. Now, he needs to find out how to manage his people. But he has learned something since we last saw him. This time he turns to data, he goes all science on us. And he has discovered something called behaviorism. He likes it. It is something he can observe, something measurable. Like the minus 10% he gets in his team’s engagement score. 

And, no kidding, it all started with animals. Edward Thorndike became famous for putting cats into puzzle boxes. The cats had to perform some kind of action — pull a loop, press a lever, operate a latch — to escape and get food. Initially they would just do random things like scratching and pushing until they accidentally did the right thing. When the cat had to do it again, it became faster with each repeated trial. Thorndike then called that the Law of Effect. Behavior followed by good stuff happens increasingly often. Behavior followed by bad stuff happens less. 

John B. Watson, the father of what came to be known as behaviorism, put it into an even more generalized formula: Humans and animals — there is rarely a clear distinction between them in behaviorism — are highly moldable through environmental conditioning. Watson’s formula was simple: Man and animal only differ in the kind of behaviors they show, not in the principles. 

Taylor was not a behaviourist psychologist, but scientific management shared the same dream: You just need to standardize work and design the environment to a point where the worker can do only the right thing, as fast and as efficient as possible. 

Law of Effect, Scientific Management — if this makes you think of physics applied to people, you are absolutely right. Although this is only classical physics. Quantum physics was conveniently forgotten, but we’ll get to that.

But first we have to introduce the pigeons and Mr. behaviorism himself, B. F. Skinner. He did not use cats, but pigeons and sometimes rats — I always thought these were alike. Well the setup was similar to Thorndike, but Skinner looked a bit closer. He developed operant conditioning, finding out more about reinforcement schedules, timing, rates of response and the influence of different kinds of consequences. Based on his studies he started to predict the animals’ behavior then finds ways to control the behavior. 

And all of a sudden we don’t just have escape behavior, we have a learning concept. Thorndike’s cats learned how to escape. Skinner’s pigeons and rats learned to repeat behaviours under controlled reinforcement schedules. 

For Skinner it was irrelevant if his subjects were animals or humans. He applied the same rules for both and studied both and to be fair, his results confirmed his view. Skinner then had his vision where he could control all living organisms. This is of course a fully deterministic world view. If all beings are determined by external influence there is no free will. And even the objection that usually stops determinists — that this would also mean that the scientists themselves were fully determined by external influence — did not bother Skinner. Even when writing his own autobiography he admitted to that. “I did not direct my life. I didn’t design it. I never made decisions. Things always came up and made them for me. That’s what life is.”

What a strange thing, this behaviorism, you might think now. Actually it’s not, strange I mean. It is one of the most common assumptions about how psychology works. Nowadays it is even referred to as Pop Behaviorism by psychology researchers. Modern scientists have actually long moved on. The old absolute version of behaviourism no longer dominates psychology. Only in cognitive behavioural therapy as part of clinical psychology is a refined version of behaviourism still living on. What didn’t survive is the assumption that external rewards shape complex human behaviour.

 But its simplified popular version still runs most people management. Never named, never questioned, just practiced as common sense. The vision of total control never quite worked in the lab, but it found a much more comfortable home in the work place. 

There were always problems with behaviorism, but these problems were just pushed aside and forgotten. See it isn’t even clear who trained whom in these experiments. Was it the scientist or the rat? Psychologist and Skinner critic Alfie Kohn once wrote a lab report from the point of view of the rat, describing how the rat had successfully trained the scientist to deliver treats whenever the rat pressed a button.

This is more serious than you think and we see it to this day in management. Because behaviorism is the basis of reward based performance management. You deliver on a KPI and you get a bonus. You have just been trained to do the right thing through an external condition. You are the rat. Well, ok, as long as you get your bonus. But the picture isn’t that clear. 

Since Deming we know that what we measure and therefore manage in performance management is not overall performance. We can only manage that one number we are looking at. And not even that very accurately because rules get bent, definitions shifted, exceptions made. Eventually the manager can only measure the compliance to the measurement system. You will not improve the performance of the team or company you will only improve that KPI’s appearance. And that will then result in you getting the treat, oh sorry, the bonus. 

Now, who is training whom? The one asking for KPIs to look good or the one making the KPIs look good so he can get something. And that’s the moment when quantum physics makes its entrance. At quantum scales, measurement is not a passive observation; it is a physical interaction that can change the system being measured. Ha, I always knew management is nothing but quantum physics. Oh wait, does that mean I have to read Planck, Einstein, Bohr, Heisenberg and Schrödinger. Well at least one of them also had a cat.

Kidding aside, see the dogs and cats and rats and pigeons, wonderful beings that they are, could not object. People can. And did. And were ignored. The most fundamental problem of behaviorism is its uncontrolled application to humans and human activity. Mainly to management practices. And it comes in all forms: Bonuses, performance reviews, merit ratings, stack ranking, competency frameworks, gold stars and KPIs. Let’s first look at the question how bad a job behaviorism really does.

The problem starts with bonuses and what their intention is versus how they are perceived. The intention of a bonus is as a potential reward to make you want to perform better. But beyond the measuring problem already discussed, there is a deeper issue: it is a category error. Bonuses are perceived as payment for work. They appear in your contract right after your wage. That makes them system requirements. That means they are compensation for the time and effort put into the company’s value-adding process. Motivation operates on another level, not the systemic one but the psychological one. 

This category error undermines the bonus from the start. Furthermore the bonus gets absorbed into the expected compensation and loses its value as a reward completely. You see this very clearly when discussing salary with employees. Wages and bonuses are treated the same way. It’s a payment for your effort, not as a pat on the back. For the company this creates a trap: you built something as a motivator that never functioned as one, but now you cannot remove it because it is expected as compensation. The company has to pay but doesn’t achieve anything with it. 

Kohn, in his book “Punished by Rewards” pointed out five further problems with the whole reward system. Should you ever read him be aware that he is a bit of an enfant terrible in the psychology scene, but hey, you read me, so you should be used to that. His points are on the other hand excellent analysis. 

First he points out that rewards become punishment immediately, because withholding a promised reward is nothing but a punishment. The carrot becomes the stick. Have you ever not received an expected bonus. That feeling is not motivating. That is punishment. Then there is the problem of rewards leading to competition, it can often become that and this is worsening relationships within organizations. Now this can be avoided but it is a danger none the less. The next one is always true: Rewards ignore reason and it reflects back on Deming’s points. We are managing the appearance of a performance, we will not look for the real issues. Because the issues are not the relevant thing anymore, only the number that we are judged upon is important. Similar to that is that reward systems do discourage risk taking. We are less likely to try out stuff, experiment and therefore innovate because there is the fear of not getting that reward. 

Now there are authors that say that all of these problems can somehow be fixed with a very elaborate bonus system. One that does not punish in any way, nor encourage competition, always makes you look for the real issues and encourage experimentation. Well, tough but ok. 

The last problem that Kohn sees is however not even fixable by the most miraculous reward system ever created. It is the fact that rewarding an activity takes away one’s desire to do the activity. What? Yes, and this has been empirically studied by many researchers and is known to be true. See, what happens is that the initial motivation why you did the activity gets replaced by the reward. Rewards always lose their attraction over time — and when they do, or when they disappear entirely, the activity loses its reason. This one cannot be engineered away. It is built into the reward itself. 

One study that demonstrated this was the Kefir experiment done by Leann Birch and colleagues. Young children were divided into three groups and offered the yoghurt drink kefir. One group was just handed it, the second group received praise when drinking it and the third group were offered free movie tickets if they drank it. Who do you think drank more of the liquid yoghurt. As Skinner would have predicted it was the group who got the reward. Wait no surprising twist here? Be patient the experiment is not finished. Birch was not interested in the initial effect of rewards, she was interested in what rewarding did to long term behavior. Because a week afterwards which group do you think still liked the stuff? Exactly the one that did not receive any praise or reward. Those who had been rewarded were far less likely to drink it once the rewards stopped. The initial reward and praise had ruined the wonderful sweet gulpy stuff. 

This finding has been reconfirmed in many other studies. There are even jokes about this effect — there was an old man … no I’m going to spare you this one. How long a reward works depends on what it is, but most fade within days, not months. What is the timeframe of your bonus? Even then rewards only work well for people who are in a dependent population. Meaning there already has to be a situation where the rewarded person is controlled by the reward giver. As in employee and manager, student and teacher, child and parents. Well that doesn’t sound like management technique more like a hostage situation.

These studies also showed that reward systems only work well for simple, easily repeatable tasks — drinking kefir for example. Reward systems can enhance quantity but not quality of work. So rewards can make employees work more but not better. We will get to the why later. Rewards, however flawed, would only be useful for simple frontline tasks. But who gets the big bonuses in your company? The guy on the shop floor? 

All studies point towards the same thing, reward systems don’t work for their intended purpose. Are there things that do work? As a side note, some of these same studies tested other organizational measures like training or goal setting without reward attached. These always outperform the reward systems. It is quite clear and Deming put it plainly: bonuses do not work for psychological reasons — they undermine the value of work. For economic reasons — they are perceived as compensation, not motivation. For systemic reasons — they destroy cooperation, create fear, corrupt numbers. Even Alan Blinder, a Princeton economist who edited a major productivity research anthology, found that how employees are treated outperforms how they are paid.

But Deming didn’t stop with bonuses. In his view evaluations and grading are just as bad. They try to do the same thing but on even shakier ground. Same source, same problem. Simply put, it’s the ‘you are a good boy’ I do with my dog, and while it works for Max because he truly is a good boy, in most cases these things are numerical nonsense because they force distribution. 

Most evaluation systems force you to choose which attributes a person excels in and which need improvement — because we don’t say bad, right. You cannot choose all good because that is not reality. We all know that. Or do we? Some other evaluation systems even do the same for groups of people which is even worse. Only 5% of your team can receive top scores. Most have to be somewhere in the middle. Again this is clear, right. We know the principle. We even know how it looks as a graph. It’s normal distribution, the bell curve. A lot of data points fall in the middle, only very few at the bottom and at the top. Math proves it, apparently. 

OK, I need to breathe deeply here in order not to scream out loud. First, it’s statistics, not — ah, whatever: IT IS NORMAL DISTRIBUTION as in naturally occurring, not played with. Companies are not the same as the total populations of humans. At least I hope, otherwise I would fire the whole HR department. You have a selection process when you hire, then you train people, these are interventions applied to the population. If your workforce still fits normal distribution after it, forced ranking is proof of your failure dressed as a performance management tool. It’s like when the Total Perspective Vortex forces me into a distribution whether I belong there or not. I am here. Insignificant. Regardless of how many Norwegian fjords I have designed. And yes, my name is Slartibartfast from now on. 

Look, whatever money you throw at people it’s not going to motivate them to work better. And we should know that since the 1950s. This is what Herzberg found after years of research for “The Motivation to Work”. He found out that older studies on motivational factors in the work environment did not provide any conclusive results. They did not make any sense basically. Problem was that people — totally understandably — thought that all factors had to be measured on one single scale. Deeply demotivating to super motivating stuff, more of the bad reduces motivation, more of the good increases it. Simple, like math. It just didn’t work out. 

So he looked at the data and did more studies himself and found that there are actually two separate scales. Certain factors can only demotivate you. Others can only motivate you. That’s the Two Factors Theory. And all the stuff that companies normally do, like wages, bonuses, free coffee, office setup, working conditions are, so-called hygiene factors. They are really important in the sense that if they are missing employees will get demotivated. So companies need to manage them. But they will not earn you a gold star as an employer. The good stuff — your motivators — are a completely different story. They have nothing to do with extrinsic stuff. 

And we should know that. Yet we keep throwing useless stuff at people. All it would take is reading a couple of books on psychology. We even ask people the wrong questions when trying to find out what motivates them and what satisfies and what delights them. If I have to tick one more endless 1-to-10 questionnaires on my engagement as an employee, I will rip that digital form apart. Does anyone actually think you’ll get usable feedback from that. Those are numbers that you get. OK, so my trust index is bad. OK, now what should I do? The next time just ask people when they felt really good about work and when they felt really bad. Critical Incident it’s called, used forever. Gives you usable qualitative feedback. Oh, but we have a 5000 employees, we cannot process all those answers. Man, keep up with the times. This is exactly what large language models are made for, to process language for you, categorize it for you, analyze it. Herzberg’s method has a real weakness: attribution bias. Okay, but even if his two boxes are too neat, that still doesn’t save the bonus logic.

And don’t get me started with employee competency frameworks that some consultant frankensteined together from some YouTube video. If somebody misuses Pestalozzi’s Head, Hand and Heart one more time, I’ll send my wife to rip you a new one. It’s a model for how to teach children not an assessment of people’s competencies. I am the only one who feels like being in the Sirius Cybernetics Corporation — the cheerful company that builds things nobody needs and that drive everyone crazy? Why are we so enthusiastic about being useless. 

Science misunderstood, misapplied, outdated. Herzberg once said that it takes about forty years for scientific insights to become popular. Well, his findings are 70 years old by now and have still not arrived in the board room. Other areas have moved faster: the field of animal training itself has left behaviorism behind. It used to be the big thing for training animals in the second half of the last century. But since then that field has also moved on. Good animal trainers still use reinforcement, of course. But they also read the animal’s emotional state, stress signals and specific behaviours. They work with what is actually going on in those little cute heads. Management is still where dog training was fifty years ago.

And science may have not totally abolished behaviorism — but it has moved on to a more nuanced view of what makes us tick. Or better yet, it found something on the other side and that changes quite a lot about how we think about managing people. 

Herzberg was the beginning and Edward Deci and Richard Ryan provided the detailed mechanism. Actually there are several modern views on the subject of motivation. They all have to do with intrinsic motivation at their core. Similar ways — but different names and scope and traditions — to describe basically the same thing. Gosh, these scientists really need a marketing consultant. 

Let’s look at two. First is the one from Deci & Ryan, Self-Determination Theory. Their research identified three basic psychological needs. Through empirical studies they found that humans need Autonomy — a feeling of choice — Competence — a sense of effectiveness — and Connectedness — a feeling of belonging. So a craftsman standing at the workbench, working at something difficult that results in something meaningful for his customers. He will feel intrinsically motivated because he chose that work. He is both challenged and enabled by it. And he works in direct relation with the people it serves. Deci & Ryan made intrinsic motivation into something that could be measured. 

Now what happens if he needs to meet a quota given by his boss or the company tells him to only make one small part and somebody else will assemble it and another will sell it to the customer. Each one chips away at Autonomy, Competence or Connectedness. These are controlling factors and they lower intrinsic motivation. Even if he gets more money for it or a nice bonus if he meets his quota. The craftsman is still standing at the same bench. But something has left the room. We saw that already by the example of the Kefir experiment. Reward does not enhance motivation, it actually destroys it.

Let’s take out the microscope and look at work even closer. What happens when we do a task. The second intrinsic motivation framework we are looking at is concerned with exactly this.  Mihaly Csikszentmihalyi — yeah I know I also cannot pronounce that correctly but the guy is worth remembering — looked at what happens when tasks flow — and that is also the title of his 1990 book “Flow: The Psychology of Optimal Experience”. 

We all know that tasks can feel like something we just can’t find our way into, like we are watching ourselves doing it not experience it first hand. Like listening to music when the acoustics are really bad and you hear distorted sounds, the beat is slightly off and nothing quite resonates. In those moments you are definitely not in the Flow. 

On the contrary it should be like “being completely involved in an activity for its own sake. The ego falls away. Time flies.” Hopefully you have all experienced that. It is truly bliss. I have this a lot when writing, but also when I do trainings, even when I go running on a trail. Because as Csikszentmihalyi writes the conditions are the right ones: I have a clear goal, get immediate feedback — hard with writing but you learn how to do that —, deep concentration — yes possible even in a busy class room —, a sense of control. Most importantly he found that there needs to be a good balance between the challenge — how difficult a task is— and the skills you have. Then the activity itself becomes satisfying. Groovy! 

But then the bonus enters the room and all of a sudden the ego is back in together with the wants. Time stops flying and starts to be measured. The activity is no longer the most important thing, but the reward is. 

Ah, this whole intrinsic thing sounds kind of neat. How could that work in an organization? Herzberg did the whole analysis for us, he actually used the critical incident method of asking specific questions to workers and found what the good stories and bad stories in their work life really were. The bad stories were always about the Hygiene factors — the surrounding of work, the pay, the free coffee, the bonus. The good stories were the Motivators and they were always relating to the activity of the work itself, the content of work. 

So all the cool stuff management does, can only cause dissatisfaction if missing, but the work itself is the only source of motivation. I, as Slartibartfast, am only motivated because designing Norwegian Fjords is just a wonderful thing, the mice just have to pay me fairly and give me clear goal and hopefully keep the planet from getting destroyed a second time. The funniest thing is not my addiction to Douglas Adams — it’s that organizations invest most heavily in hygiene while the thing that actually matters, the design of work itself, gets the least attention.

And it’s not even that hard. It’s called job enrichment — well vertical job enrichment to be exact. It means workers take more ownership and responsibility, that their skills and judgment visibly impact the work, and that the results carry meaningful connection. That does not mean more tasks but doing their task with more Autonomy, using their Competence in an environment of Connectedness. 

This is really important because if you don’t do that, if you constantly control and micromanage people, you will get people who mirror exactly that. Idleness, indifference and irresponsibility are healthy responses to absurd work. The passive employee is not a flawed person, it is a made person. Made by the organization. And you end up with a staff full of Marvins, super AI robots who are depressed and will eventually not even open doors for you. And that would be really ineffective. 

One step further down the rabbit hole — Philosophy. John Dewey was a psychologist and philosopher who set out to reform the educational system in the first half of the last century. He wasn’t one of the ‘give them all freedom, all experience is good’ kind of people. Actually he criticized both conservative and progressive education. See he said that both a totally controlled education and an unguided education can result in the same result because not all experience is educational. Experience can train stupidity. 

What is needed are not rigid structures nor absolute freedom, but an intelligent structure that enables meaningful experience. You already see where this is going. Of course, you are all intelligent people. This principle of intelligent structure or guided experience is exactly what is needed for the work itself. Autonomy, Competence and Connectedness does not mean some weird hippy commune. It means that work is designed to provide positive experiences. And this might be the hardest thing to do, but hey, we get paid for something, right. And maybe designing meaningful jobs is exactly enough of a challenge to match your skills and send you off flowing.

Of course there might be leaders who think — because nobody says this out loud anymore — that people should just work, just produce. They don’t need to be motivated or have meaningful experiences. They are only there to be part of our big machines. 

In response I have to bring out the heavy guns: Immanuel Kant — I guess you all know the name — the god of ethics and reason. He wrote some pretty intelligent stuff. And you better not try to counter him otherwise the hammer of reason will flatten you. 

The thing is people are not things. Kant would say that you cannot treat them merely as a means because a person is a rational being, capable of reason, moral judgement and self—legislation. A rational person can ask: What should I do? What rules could I honestly accept as valid not just for me but for everyone? That’s the categorical imperative. That’s why a human is more than part of your machine, not just a resource in your value-adding process. Humans can give themselves moral law, therefore they have autonomy, therefore they have dignity. Their value is not bound to their usefulness or efficiency or productivity or market value. They are a ‘Zweck an sich,’ not just a means. They can help you create something with your big machines but you can never reduce them to just being a part of it. Humans are not tools.

Look, in the end it all comes down to this: You cannot treat people as if they were pigeons. Leave aside Kant if you want — it’s a crime but you can try. Managing people like that is just plain ineffective. If companies want more than just a herd of rats they need to stop trying to condition employees with treats. Organizations would benefit much more if they would enable people to do their best by creating meaningful jobs.

I guess there is more to explore. 

Is your job meaningful? Discuss here: LinkedIn

Don’t just take it from me, here is some good stuff to read:

Alfie Kohn, Punished by Rewards, 1993

Frederick Herzberg, Bernard Mausner and Barbara Bloch Snyderman, The Motivation to Work, 1959

Frederick Herzberg, One More Time: How Do You Motivate Employees?, Harvard Business Review, 1968

John Dewey, Experience and Education, 1938

Immanuel Kant, Grundlegung zur Metaphysik der Sitten, 1785

Edward L. Deci and Richard M. Ryan, Intrinsic Motivation and Self-Determination in Human Behavior, 1985

Mihaly Csikszentmihalyi, Flow: The Psychology of Optimal Experience, 1990

W. Edwards Deming, The New Economics, 1993

Douglas Adams, The Hitchhiker’s Guide to the Galaxy, 1979