Sunday, May 07, 2006

Performance reviews

I have left one company that has written goals and objectives, which are evaluated yearly and turned into a bonus figure ... for another company that has written goals and objectives, evaluated yearly and turned into a bonus figure. I couldn't be happier about it. Why?

The second company doesn't have forced ranking (FR), aka Forced Distribution, while the first one did. With forced ranking, employees are grouped into pre-sized buckets, which HR usually gives neutral numbers like 5..1 or neutral names like "Superior" to "Requires Improvement". The names hardly matter, since they're functional categories that everyone understands: "highly paid" down to "about to be fired."

There are many complaints about FR (use if you don't want to register)
Those who are opposed to forced ranking suggest that the process may:
  • Be detrimental to morale. We want high-performing workplaces, yet we also want people to enjoy coming to work. Can we have both?

  • Emphasize individual performance at the expense of team performance. Employees will be less willing to "pitch in" and help others. Completing individual assignments becomes more important.

  • Promote competition. If employees are ranked, will there be less cooperation?

  • Invite litigation. An important concern, this will be addressed later in the article.
Many of these objections are valid. In particular, people have pointed to Enron's "rank and yank" policy as an indirect cause of its later flaming demise. Here's a pre-crash article that buys the whole thing, saying that this management style at Enron rewards high-performers, and encourages "everyone to work harder and smarter." It's a truly remarkable booster article to be written about a company that would be valued at pennies per share just a few months later. The terrific documentary Enron: The Smartest Guys in the Room touches on "rank and yank", and it's covered in great detail in the book that the documentary was based on, Smartest Guys in the Room: The Amazing Rise and Scandalous Fall of Enron. I recommend them both highly.

There's nothing wrong with ranking employees, of course, and there's not much wrong with firing employees who aren't working out according to the local standards. What is evil is
  1. requiring that the employees be artificially forced into buckets of arbitrary sizes.
  2. iterating this ranking process repeatedly, especially if it leads to mandatory dismissals.
  3. pretending that the ranking is entirely objective.
For point one, Forced Ranking is pseudo-science. You can grab 100 oranges randomly off a tree, measure them by any standard you like, and they'll fit some normal curve. But if you grab 100 oranges out of a bin at a supermarket, it's no longer a random set. Those oranges have passed several filters: those that were picked, those that had the "right" orange color, those of the right size, etc., etc. Some of those causes are interlinked, but the ultimate conclusion is: they're not random oranges that arrive at the supermarket. If you still don't believe me, check out the orange bins at a higher-end grocery store, then toddle over to the cheapest supermarket in your area.

Your fellow employees are not randomly chosen, either, just like the oranges that finally appear in a supermarket. Their resume reflects years of history of a particular type of behavior. The resume was compared, accurately or not, to a job description. They were picked for the job after a series of interviews and tests. Once hired, they discovered what the job really is — they didn't run screaming out the door. They're not random. There's no reason to assume that they're suitably ranked along a bell curve, especially when FR is usually done in the group of people under a single manager, so groups of 5-20 people.

The second evil aspect in Forced Ranking is the iteration. Many businesses can likely be improved by ranking all the employees and eliminating some fraction of the lowest-ranked people. Once. If you're actually firing people each year who are in the bottom bucket, then the people that appear in the bottom bucket this year were considered adequate the previous year. How does that make sense? Were you ranking them incorrectly last year or this year?

This leads directly to the third evil side of Forced Rankings. Why are these objective rankings? Very, very few jobs can be ranked objectively, unless you're a paid Olympic sprinter. And since you're starting with numbers that are tainted by management and personal bias, how is it possible that these groups will magically align themselves into a bell curve? In some cases, is the manager even worthy to judge the employee?

Forced Ranking is all crap. It should be abandoned as a cruel hoax pretending to be a scientific management theory.

Forced Ranking probably appeals to MBA types who think they can apply quality improvement methods to people the way that they can to making machine washers. It seems highly numerate and "scientific". The difference is that the little metal disks don't feel fear, and the cruelty of FR is that you're forcing an arbitrary portion of your staff to live with fear of losing their job.

(Image of bell curve taken from


Post a Comment

Links to this post:

Create a Link

<< Home

Blog Flux Directory