Category Archives: Uncategorized

From reducing to absorbing complexity

This reminded me of the work we sometimes did as consultants with frameworks, strategy models and elaborate templates. I felt that we didn’t really capture the complexity, we reduced and simplified it in a way which didn’t really work. I don’t think we really saw and allowed for an open system which constantly emerges dynamically, but aimed more for a controlled setting with plans and execution:

“Boisot felt that a fundamental premise underlying bureaucracies was a belief that knowledge is something that can be simplified, and then easily transferred between parties – an orientation towards reducing complexity. (…)

Boisot recognised that shifting the way companies worked with knowledge presented significant cultural and organisational challenges. He felt that the ability to make the shift from reducing to absorbing complexity would be one of the key determining factors about which companies survive the transition to a knowledge economy and which would become the dinosaurs of a former era.”

The fewer assumptions a model has, the more flexible we are and the less obstructed in our observations. The higher our capacity for learning. And the swifter we can act and change course when needed.

Quote from article: Strategy as Hypothesis


Ideas for improvement – Google Translate


I wonder at what point Google Translate will offer the option of a hyperlink to the original text with the translation.

Sometimes the translation doesn’t make sense and getting back to the original can help – and also help one understand the other language better, maybe even learn…

By effects:
1. free advertising when linked to Google Translate page…
2. some learning on when people want to check the translation which may be an indicator of the quality…

Management miscellania – Common misunderstanding # 1 – good – cheap – fast

Statement: you can’t have good quality, fast and cheap at the same time.

Good and fast, not cheap.
Good and cheap, not fast.
Cheap and fast, not good.

This is a common misunderstanding: it can be all 3.


The better your processes, the less waste you have.
The less waste you have, the faster you can produce.
The less waste you have, the higher quality you can produce.
The less waste you have, the less resources you need.

So this then leads to higher value per unit of cost/time.

If you want to improve, focus on value added activities / cost.
If you focus on cost, it invariably leads to worse quality by higher cost.

This is the reason why so many cost cutting operations go wrong…

Changing behavior – when, how and how much? Comments on an article on the Gates Foundation-Blog (2013)



This sounds like an interesting example. If I understand correctly, the idea is to stimulate farmers to “pay it forward” by giving some of their proceeds to other farmers. My question here is: why? Why not stick to giving them beans, help them to grow them successfully and spread the program?

It seems that the underlying assumption is that since they have received help, they should also help others. They should reciprocate. And the assumption is that it is not enough to redistribute this newly earned wealth within their own social environment. But reciprocation is only valid when some of the beans (assumption: they got beans, they give beans to others) should be redistributed among the other farmers.

Some problems – do you really want them?

1. Redistribution of wealth:
As they may all not be very rich (and even then people underreport…), it seems very natural to want to save something for themselves. Let’s not even start hypothesizing what social and other factors may come into play when you effectively are redistributing wealth to these other farmers (who might be friends, enemies, competitors etc.). Because it seems that this is exactly what you are doing here.

2. Linking proceeds with “paying forward” incentivizes cheating:
The amount of what they donate is linked to the proceeds they generate. And of course they underreport. In a way, the whole idea incentivizes cheating as a behavior.

3. Compliance costs resources:
There is a “compliance problem”, because you have to spend money on resources to control the adequate reporting of the farmers proceeds. So in a next step you could try and look for a simple metric which is independent of the individual success of the harvest, but more general and easily computed. So people don’t have to cheat. But then you still have the problem of administering etc. So essentially the problem doesn’t go away.

A simple heuristic could be “Give back the beans you received”. In bad years that might be too much, in good years too little. The question is more: does the amount of beans they give back really matter that much?

4. “Nudging” makes it more complicated:
What to do then? HarvestPlus then looks for a way to “nudge” them into doing this. In order to “motivate” them to donate one way of doing this is by means of a lottery in advance and a ceremony. So in essence you make a very “simple” thing (donating beans to farmers) more and more complicated. As much as I like behavioral change strategies, is this really the best option?

So starting from simple “linked” assumptions (“paying forward” with “with beans” & “to other farmers”) we end up with something which gets more and more complicated.

Let’s get back to the starting question:

Why wouldn’t we want the farmers to decide themselves what they want to do with their crop? So they can “pay forward” to their own best interest. We might even encourage them to save, so they have some resources when times get rough. There wouldn’t be any compliance issues. No cheating.

If we apply this to ourselves – and walk a mile in their shoes – how would we feel?

Imagine having received a “free” job training (or just your education) in the US, which asks you to donate to others after you have succesfully reentered the job market (depending on your selfreported income). How does that sound?

That puts you into a rather difficult position. Wouldn’t you rather be taxed directly, so the state does the redistribution? Or would you prefer to be asked to contribute in another way? Maybe educate others, so they can make more “educated choices” and modulate the behavior of their most successful peers? Might that not be a more important goal than being “forced” to pay forward to people you didn’t choose?

What do you think? Am I missing something here? What mistakes did I make? I look forward to the discussion!

(This comment was partly inspired by a conversation with Gerd Gigerenzer at TEDx Zurich about educating people to make informed choices. Any mistakes are all mine.)

Hugo Mercier et al. – Explaining the cultural success of bloodletting – alternative factors?

I came across an interesting article on bloodletting trying to explain its success and wondered if there might be alternative factors:

  1. the availability effect
  2. the recency effect
  3. the post hoc ergo propter hoc-fallacy
  4. incentives
  5. activity vs. passivity (sense of agency)
  6. the placebo effect

1. Availability – Frequency:
If one asks in which instances do we see blood, it seems there are a quite few occasions. Maybe even more so in the past than in the present. (Assumption: I don’t have historical data on the prevalence of bleeding as percentage in occurence in illnesses. I assume one might be able to find it).

Bleeding is quite common (in women even more so), observable and often non-lethal. People get better. (Edit: The same with f.e. the common flu, after a lot of sneezing, coughing up phlegm and blowing one’s nose (relieving oneself of unwanted substances in more or less liquid form)) Inversely, death itself and many illnesses do not involve any bleeding…

2 and 3. Recency and Post hoc, ergo propter hoc:
Especially in common cases of inflammation by an object/sting etc., the cure often seems to open the wound and push the puss/splinter etc. out. The marker that seems to say: “you’re good” is, apart from relieved pressure etc., no more puss comes out of the wound, instead of it we push blood out of it. If the inflammation is not contained by the immune reaction of the body, it can lead to a lethal sepsis (blood poisoning). (Edit: I wonder if this was sometimes called “bad blood”?)

dilbert-post hoc ergo propter hoc

So if we now assume, that this “blood comes out of the body” was the last thing which happened, is interpreted as a sign for “you’re going to be ok”, then this fits well into Hugo Mercier’s et al. attractor/reasoning/relevance argument. It also fits well with availability bias, recency bias and post hoc, ergo propter hoc fallacy.

The symptom becomes a cause, since we just seem to know from experience that after blood letting the person gets better (pattern recognition = then and then and then), but we don’t have a better model to explain the phenomenon for it (theory of mind). To a man with a hammer, everything looks like a nail.

4, 5 and 6. Incentives, activity and placebo:
If a little bloodletting would normally not kill a patient, it was a rather safe way of generating revenue. The equivalent of that today might be the prescription of some drugs or treatments.

It also seems that many common diseases in the past were not very well treatable: Dysentery (the “bloody flux”), Ergotism (“St. Anthony’s fire,” “holy fire,” “evil fire,” “devil’s fire,” “saints’ fire”),  Gonorrhea, Influenza, Leprosy (“lepry”), Malaria (“the ague”), Measles, Plague, Puerperal fever (“childbed fever”), Smallpox (the “red plague”), Typhoid fever… (

Treating patients with bloodletting was also a safe way of actually doing something, whereas in many cases the patient could not really be helped. The treatment of bloodletting might even help the placebo-effect.

A. I expect people, who from personal experience had seen many people “bleed to death”, to be less inclined to the practice of bloodletting (patients refusing bloodletting). In this way we might be able to factor out the reputational investment of the doctor.

I guess one can come up with more testable hypotheses to falsify the 6 factors.

The original article is:
Miton, H., et al., Universal cognitive mechanisms explain the cultural success of bloodletting, Evolution and Human Behavior (2015),

The abstract:
“Bloodletting—the practice of letting blood out to cure a patient—was for centuries one of the main therapies in the west. We lay out three potential explanations for bloodletting’s cultural success: that it was efficient, that it was defended by prestigious sources—in particular ancient physicians—, and that cognitive mechanisms made it a particularly attractive practice.

To test these explanations, we first review the anthropological data available in eHRAF. These data reveal that bloodletting is practiced by many unrelated cultures worldwide, where it is performed for different indications and in different ways. This suggests that the success of bloodletting cannot only be explained by its medical efficiency or by the prestige of western physicians. Instead, some universal cognitive mechanisms likely make bloodletting an attractive form of therapy.

We further test this hypothesis using the technique of transmission chains. Three experiments are conducted in the U.S., a culture that does not practice bloodletting. Studies 1 and 2 reveal that stories involving bloodletting survive longer than some other common therapies, and that the most successful variants in the experiments are also the most successful variants worldwide. Study 3 shows how a story about a mundane event—an accidental cut—can turn into a story about bloodletting. This research demonstrates the potential of combining different methodologies—review of anthropological data, experiments, and modeling—to investigate cultural phenomena.”

Peter Drucker Forum 2014 – Interlocking goals, practices and metrics (Systems!)

Steve Denning summarizes it neatly:

It also came out that the goals, practices and metrics of this traditional way of running organizations fitted together as a kind of perfect marriage. That was because when you have a goal of making money for the shareholders and the executives, you cannot inspire people to pursue that goal with any commitment or passion. So you have no choice but to run the organizations with command-and-control. You had to have hierarchical bureaucracy to force the employees to pursue a goal that they didn’t really believe in. So shareholder value and hierarchical bureaucracy fit together in a perfect interlocking relationship, like a hand in a glove If you try to change one aspect, such as better team practices, the other aspects—the goals and metrics—undermine the change. So these organizations are stuck in a state of suboptimal equilibrium. When you throw in the phenomenon of exorbitant executive compensation that is linked to keeping up the stock price, it was hard to see how real change could be possible.

Joseph Stiglitz – Corporate Welfare

“Perfect competition should drive profits to zero, at least theoretically, but we have monopolies and oligopolies making persistently high profits.C.E.O.s enjoy incomes that are on average 295 times that of the typical worker, a much higher ratio than in the past, without any evidence of a proportionate increase in productivity.

If it is not the inexorable laws of economics that have led to America’s great divide, what is it? The straightforward answer: our policies and our politics.”

“So corporate welfare increases as we curtail welfare for the poor. Congress maintains subsidies for rich farmers as we cut back on nutritional support for the needy. Drug companies have been given hundreds of billions of dollars as we limit Medicaid benefits. The banks that brought on the global financial crisis got billions while a pittance went to the homeowners and victims of the same banks’ predatory lending practices.”