Home | Login | Recent Changes | Search | All Pages | Help

DecisionMakingPrinciples

Robert Rubin, the former secretary of the Treasury and former co-chairman of Goldman Sachs, uses the following four principles to guide his decision making:

  1. The only certainty is that there is no certainty

  2. As a consequence of Principle #1, every decision is a matter of weighing the probabilities

  3. Despite uncertainty we must decide and we must act

  4. We need to judge decisions not only on the results, but on how they were made

Source: Rubin's 1999 commencement speech at the University of Pennsylvania

What principles does your organization use to make decisions? How do they comapare to Rubin's?

If your organization doesn't use Principle #4, why do you think/feel it doesn't.

What story would you like to share?

SteveSmith 2006.07.30


One problem with #4, certainly when it comes to individual decisions, is that we never really know how they were made. That's true for most decisions by most organizations. What they rationalize was their process is seldom their process. - JerryWeinberg 2006.07.30
Jerry: What they (individuals and organizations) rationalize was their process is seldom their process.

I agree. I do think, however, that a review triggers a chance for rethinking. The rethinking may not show up in the rationalized process but it may indirectly influence how the next decisions are made.

Principle 2. As a consequence of Principle #1, every decision is a matter of weighing the probabilities

My experience is the the people whose opinions count the most (typically upper management) rarely make decisions, such as ship dates, based on weighing the probabilities. If they do, it's based on probabilities from their own personal desires rather than from the people who are creating the product. Why? Probabilities from the lower ranks might reveal cracks in a carefully constructed story that has financial impact on the company or its stockholders. This story must be protected at all costs unless external forces break it.

For instance, there is a large software company with an operating systems that is two years behind its original schedule. Influential outsiders doubt whether the OS will make it to the market at the beginning of next year, which is the plan of record. I suspect that if at the beginning of this year, you polled the people who are developing the product about the probability it would ship early next year with acceptable quality, you would have heard 10-20%.

What was the advantage to this large software company by making its decision about the ship date the way it did?

What were the disadvantages?

SteveSmith 2006.08.04


The advantage was that it kept the marketing people off the developers' backs for a while.

The disadvantage was that the date was entirely meaningless in terms of what could actually be delivered, and when the reckoning came, it slowed down the development even further. - JerryWeinberg 2006.08.13


Updated: Sunday, August 13, 2006