3 ms·
> Managers gave about 10% of the company’s workers poor reviews The article didn't manage to get a quote on what the guidance/orders for the reviews was. For
by neilv 4y ago
> Managers gave about 10% of the company’s workers poor reviews
The article didn't manage to get a quote on what the guidance/orders for the reviews was.
For example, were managers given a target on how many of their reports were to receive poor reviews?
Or did the evaluation criteria change besides the target distribution?
- UncleEntity 4y agoMaybe they are grading on a bell curve like they did in my youth? 10% get excellent reviews, 10% get poor reviews and everyone else falls somewhere in the middle.
- ksaj 4y agoIt was yet another way to stretch that 80-20 rule's mileage. I remember when the idea of a bell curve was new, and you saw it applied to absolutely everything, even if it was only a superficial (or sometimes completely bogus) fit. I've even seen it applied to the Kinsey Scale. Presumably 10% of the population is totally straight, 10% totally gay, and 80% are varying degrees of total party animals after dark. Or something like that. That might have actually been where I first heard of the Bell Curve, but I've seen it applied in a lot of places since. The worst place I've seen it applied is for DST daylight hours. Bell curves exist, but it's strange that people only see it as a variation of the 80-20 rule.
- tristor 4y agoTechnically a bell curve can only be 80/20 of you slice things several standard deviations from the mean. Anything that truly falls on a curve will not end up being 80/20.
- clnq 4y agoIt’s just called stack ranking now. And it doesn’t make sense. A company can be full of high performers or low performers, but the stack ranking principle itself would make management blind to that.
- belval 4y agoI'm confused, is 10% basically the stack ranking that we see at Microsoft/Amazon anyway? Did Facebook not have a "poor performer" bin when evaluating an organization before?
- amethyst 4y agoAlso keep in mind that this is the first year-long perform cycle, so % of ratings is not directly comparable to previous cycles.