This article made me realize that goal setting is really a specialized case of "beware of the metrics you set". Whatever you decide to measure is always what people focus on, so you need to be darn sure you're measuring the correct things or you end up with huge distortions in people's behavior.

This is something that Joel Spolsky has hammered on repeatedly, regarding employee motivation - http://www.inc.com/magazine/20081001/how-hard-could-it-be-si...

Setting the goal was not the problem here: the problem was the way they chose to achieve that goal. Building better cars might, rather than doing everything to sell crappy ones, would have had a much better chance for success.

They could have decided to produce something completely different, like nokia, that old world-leading manufacturer of rubber galoshes.

That's a matter of setting the right goal. Perhaps "make better cars" would have been appropriate goal, insetad of "make more money".

Making more money is an ok goal, though a bit too general for practical use.

The problem comes from a short-sighted approach to making more money now at the expense of making money later, which is what GM did. If they'd decided to sell more cars by making better ones, it probably would have taken longer to get to that point, but they'd have been in a much better position to compete with the other car makers.

Is this not about setting the wrong goal, as against setting a goal itself?

Yeah, but that doesn't make nearly as interesting a headline. :-)

The paper referenced by the story, from business professors at Northwestern, UPenn, U of Arizona, and HBS, is titled "Goals Gone Wild". (Since both the Boston Globe and the submitter resisted the temptation to use that as the hed, I had to mention it.)

The abstract and full paper (PDF) are available at:

http://hbswk.hbs.edu/item/6114.html