02514: Bad Bell Curve Usage


My current Audible book while walking is Work Rules by by Laszlo Bock, former SVP for People Operations at Google. So basically it's another book about how Google works, but from an HR perspective - and this makes all the difference. And the past few days I've been covering the chapters on the nitty gritty of people management from a Google perspective.

Part of the discussion was talking about the traditional management model where we use a standard Gaussian bell curve to model employee performance and thus expect there to be a certain percentage of top, average, and bottom performers and how you're expected to coach your middle performers to move up the average level of performance while coaching out your bottom performances, thus actually truncating that tail of the bell curve.

This brought back memories of how this was implemented in a previous company where every department was expected to profile their employees using this distribution model. However I was in a very small department at the time and thus it typically meant that only 1 person could be a top performer while 1-2 were bottom performers in need of action plans for improvement. It was a rather depressing way of doing things since it meant that after I had processed my performance for reviews for everyone, they would then get reviewed to be normalized against this bell curve and thus the scores would have to be adjusted accordingly. It made it feel like my actual appraisal didn't matter as much matching the model did and thus the final appraisals that I had to deliver to my team were not the ones I had originally prepared and felt were a good measure of their performance for the period.

Laszlo seems to agree with that sentiment and points out how different teams follow different performance distributions and a single bell curve doesn't work across the board. In addition, the process of culling the bottom 10% really hurt the company more than helped it as a lot of skilled talent were let go because the company did not create the sort of culture and environment that focused more on development.

While our company has been trying to do things differently in many ways, this book does have me rethinking a lot of my established understanding of people management and I think there some concrete ways we can put some of these ideas into practice. It means resisting some of the instincts honed by previous years in super serious corporate environments, but then again that's what I've repeatedly been doing this whole time ever since I joined the company. And it's a fun ride that I have zero regrets about.

Comments