Companies are social systems, and their behavior cannot be fully predicted. As in 'true science', management science needs to get back to the working mode of performing experiments and to use the corpus of knowledge mainly for formulating hypotheses and for interpretation of results. Anything else is presumptuous.

I've published the below blog posting in December 2009 already, and I am really happy to finding some support at HBR online, today in Dan Ariely's columne "Why Businesses don't experiment". Wouldn't it be high time to get going on a different path of learning? Wouldn't it be the moment to start to "learn about organizational learning" instead of searching for the magic wand in systems, proceses and concepts?

So I re-publish my plog posting once more ...

The «law of XYZ» ... - how often are they heard and taught in business schools around the world? Traditional management knowledge is often sold as an equivalent to a scientific law: «Lower the price and you'll sell more», «control more and productivity of the work force will increase», «increase the variable salary and the management team will be more focused» ... Armed with all this «knowledge», executives hit the companies around the world and start applying it, surely at their best abilities. And, the result? Sometimes success, quite often indifference in the results, and quite often disappointment. Why that? Is this «scientifically proven management knowledge» not standing up to expectations?

Yes and no, but rather no. Primarily, companies are social systems and not mechanical machines which work according to algorithms. And, by definition, any social system has its own and individual (and initially unknown) rules and inner workings.

In a recent presentation of Hans A. Wüthrich (co-author of "Musterbrecher") I suddenly realized a dramatic difference between "true science" and "management science". In "true science" (biology, medicine etc.) hypotheses are formulated and, subsequently, verified in experiments. All knowledge is been applied to formulate hypotheses only, and, of course, to understand the results of the experiments. So it's a continuous loop of experiments, verification etc. ... Nothing is firm and fix and rigid. But, in too many management teams the laws are written in stone "that's what our customer wants", "that would demotivate our employees", "they can't do that", "that's too complex" etc....

Why no getting back to a lab situation more often and actively make experiments, out there in the market, in there in the company? By brace and crazy and mainly tackle the unwritten laws of your business - these paradigms which exist in every company, particularly in family owned businesses. Experimenting truly is fun, and it provides incredible insights which are otherwise not affordable. And, yes, it takes the readiness to "let go", to let go of control, of 'knowledge' and of perceived security.

Our corpus of knowledge and experience is an incredibly base for formulating hypotheses and for trying to understand results of experiments, but not for apodictically predicting causality. To predict causality and behavior of complex social systems is close to impossible, luckily.

Get an interesting experiment going still today!