Yes, I’m late to the game on this (in a number of ways). I was on the first of many planes home last week when it was announced that Microsoft was killing its employee performance stack ranking system. The announcement sent shockwaves through the tech, HR, and business news channels.
Stack ranking is simple in theory: Employees are assigned ratings, and those ratings have to fall within a predefined bell curve. Most companies use what is known as the 20/70/10 model:
The idea is that most performers–the 70% in the middle–fall around the mean; solid, if not exceptional. The truly exceptional performers are the 20% to the right–and are incented accordingly. The left 10% are those at risk–whose work performance is poor enough that they should be on a performance improvement plan (or, in the days of Jack Welch’s GE, fired).
Stack ranking at MS was a key way of being able to differentiate the level of reward for performers of different levels once MS’s stock price leveled off about a decade ago. Unlike a smaller, pre-IPO startup company–or a post-IPO company still in rapid growth mode–a mature company, with a stock price that isn’t moving upward, cannot incent employees using stock options.
Unfortunately, this practice has been singled out by many employees as being damaging to Microsoft’s ability to consistently innovate and deliver game-changing products and services. Vanity Fair’s piece on Microsoft stated that every current/former employee interviewed for the article named the stack ranking process as a key driver as to why MS had lost its way. Stack ranking created a culture of individuals focused on championing their achievements over the the outcomes of their team–and with Microsoft attempting to pivot to a more integrated services model, that cannot continue.
Living in MS’s backyard, I know that MS employees and managers have gamed the stack ranking system. Employees that know they are a high performer in one org don’t take positions in a new org–where their knowledge & skills might be crucial–because they are too well positioned within the stack rank of their current team. Conversely, I’ve heard of managers who actively recruit a low performer from another team to help push their incumbents up in the stack rank.
Is stack ranking all bad? Marissa Mayer has instituted a fast-track quarterly performance review process @ Yahoo and is using stack ranking to weed out employees identified as low performers. She is using the process to quickly change the performance culture. Thus far they have eliminated 600 employees–so she is seeing results. Is it the right way to achieve rapid change? That remains to be seen.