Archive for the ‘Decision’ Category

Your company needs to long-range forecast? Don’t depend on a single person!

January 18, 2008

Companies try to forecast the future. For example, a printer company tries to forecast the future demand of printers. Based on their forecasts, they make planning decisions. Thus, decisions are made in the face of uncertainty.

The more power you give to a single individual in the face of complexity and uncertainty, the more likely it is that bad decisions will get made.

Conversely, decisions made by aggregating the collective wisdom of a diverse group of people will outperform even the smartest person most of the time.

— Paraphrasing The Wisdom of Crowds by James Suroweicki

To have good decision-making, give people closest to the problem decision-making power

December 8, 2007

I think that these can safely be considered to be “laws”:

  1. Individuals closest to a problem have the most accurate and complete information about the problem.
  2. The further removed an individual is from a problem — by time, space, rank — the less information and the less accurate information he has about the problem.
  3. Good decision-making requires accurate and complete information.


  • To have good decision-making, give people closest to the problem the power to make decisions. For example, give decision-making power to those people on the factory floor.
  • Avoid centralizing decision-making power to a few individuals at the top of a hierarchy.

Unless people know what the truth is, it’s unlikely they’ll make the right decision

December 7, 2007

“Unless people know what the truth is, it’s unlikely they’ll make the right decision.” [Wisdom of Crowds by James Suroweicki]

If employees see problems and keep it to themselves, it leaves an organization without knowing the truth, and thus unlikely to make the right decisions and stunts organizational learning. For example:

“Those nurses whom doctors and administrators saw as most talented unwittingly caused the same mistakes to happen over and over. These “ideal” nurses quietly adjust to inadequate materials without complaint, silently correct others’ mistakes without confronting error-makers, create the impression that they never fail, and find ways to quietly do the job without questioning flawed practices. These nurses get sterling evaluations, but their silence and ability to disguise and work around problems undermine organizational learning. Rather than these smart silent types, hospitals would serve patients better if they brought in noisy types instead.”

Hard Facts by Jeffrey Pfeffer and Robert I. Sutton

Democracy doesn’t mean endless discussion, it means a wider distribution of decision-making power

December 5, 2007

During the 20th century organizations wanted to include more workers in the decision-making process. That is, organizations wanted to be more democratic.

To accomplish this, organizations formed lots of committees and groups for workers to participate in.

However, the actual decisions were still made at the top.

Democracy doesn’t mean endless discussion. Democracy means a wider distribution of decision-making power.

— Extracted from The Wisdom of Crowds by James Suroweicki

Consensus + Management Hierarchy = Poor Decision-Making

December 2, 2007

The idea that top-down organizations are oppressive and damaging, and that workers should be given more decision-making power is well-known to managers.

To involve as many employees as possible in the decision-making process, management forms lots of teams and committees, comprised of a variety of workers.

Thus, before a CEO makes any decision, the issue makes its way through each layer of management hierarchy. At each level the issue is vetted by a committee.

Each committee resolves the issue by reaching consensus (lowest common denominator).

As the issue bubbles up through the hierarchy the opinions and ideas become more and more watered-down.  By the time it reaches the CEO there is little innovation or diversity left.

Paradoxically, in trying to make the decision-making process as inclusive as possible, companies actually make top executives more — not less — insulated from the real opinions of the workers.

Layers of management, coupled with a “can’t we all get along” (consensus) attitude is a recipe for poor decision-making at the top.

— Extracted from The Wisdom of Crowds by James Suroweicki

Here is a related blog: Consensus versus Collective Decision-Making

Experiment-driven, data-driven approach to web design

August 30, 2007

The book Hard Facts talks about the importance of making decisions based on evidence and data (“experiment-driven, data-driven mind-set”).  At this point in the book the authors are talking about how fully Yahoo! embraces this approach:

“Yahoo! Inc is skilled at running experiments and learning from them, as well as building a culture that emphasizes evidence-based management. Usama Fayyad, chief data officer at Yahoo!, points out that because its home page gets literally millions of hits an hour, the company can design rigorous experiments that yield results in an hour or less — randomly assigning, say, one or two hundred thousand visitors to the experimental group and several million to the control group … Yahoo! conducts experiments and uses the results to enhance company revenues and profits.  Much of this can be done very quickly; sometimes, results can be seen within minutes of tweaking something on the homepage or in Yahoo! Mail. This means there is no reason to spend time discussing which variation to explore or what design opportunities to pursue — it is often cheaper, easier, and faster to simply try all of them and learn what actually works. Yahoo! typically runs 20 or so experiments at any time, manipulating things like color, placement of advertisements, and location of text and buttons. These little experiments can have big effects, like the one run by Nitin Sharma, which showed that simply moving the search box from the side to the center of the home page would produce enough additional ‘click-throughs’ to bring in about $20 million more in advertising revenue a year … Yahoo! has the mind-set that says, Instead of debating which screen design looks best, or which placement of content and which choice of specific content works best, we’re going to try it all and see what works.”

Hard Facts by Jeffrey Pfeffer and Robert I. Sutton