Lots of ink was (figuratively) spilled in the digital media world a few weeks back talking about the implementation of stack rankings at Yahoo. Was this Marissa Mayer’s second fumble when it comes to employee management? You remember the first (well debated) fumble being the work-from-home kerfuffle earlier this year. While the true impact on their workforce and, ultimately, corporate performance certainly cannot be seen immediately, that hasn’t stopped people from discussing it and, generally, panning their decision to implement stack rankings.
And in an interesting twist... the news came the same week it was revealed that Microsoft stopped that same practice. So what's this all about?
So let’s talk about what stack ranking is… at its simplest, it’s what the name implies. You stack every one of your employees.
Under these, like most performance systems, those at the top get the accolades, bonuses, promotions; those at the bottom, well, they see the opposite. This certainly seems reasonable. Determine your top performers and reward them. Perhaps part ways with your poorest performers. Simple, right?
But it’s not just ranking them first to last that makes the Yahoo and Microsoft cases newsworthy, but rather distributing them into groups based on some sort of bell curve (or other pre-set distribution) where a certain percentage must fall into categories from best performers to worst performers. This is the distinguishing factor in this debate - not just rating or ranking performers to create your list, but forcing a distribution.
The result of using a curve is that, regardless of performance, some percentage of employees are always ranked in the lowest tier (and the highest tier, but that's not discussed much). No matter what. Is that the right thing to do? In my mind, it begs questions like:
The most important question, however, is whether or not this type of stack ranking method achieves the purpose for implementing it.
Microsoft implemented a stack ranking practice, but abolished it after it didn’t work well for them. Pure coincidence that it was the same week as Yahoo's announcement. Anecdotal evidence for Microsoft suggested that top performers would avoid working together lest face the possibility that they could be stack ranked into the bottom tier based on having competed against only other high-caliber performers.
Wisdom there says you’d be better off competing with people against whom you believe you can rank higher.
The practice was also forcing people to look at short term gains at the expense of long term goals. As well, it created a more politically-charged environment that was less about doing and more about building relationships to ensure a better ranking irrespective of actual results.
What’s Microsoft doing now? Focusing on teamwork, collaboration, growth and development. And now, it comes without ratings or rankings.
I’m picturing them all holding hands and singing Kumbaya.
Now let’s wait and see the results come in.
Back at Yahoo, employees have already lamented being forced to rank some employees in lower tiers on the scale, even if the person really didn’t perform to what that level describes. They just performed at a level below their peers and were forced downward to fit the curve.
But isn’t that a good distinction to make? Not everyone performs equally, so why shouldn’t we express those differences in some meaningful way and allow a ranking to dictate. Apparently some at Yahoo felt they were forcing acceptable performance into what seem like unacceptable categories based on the distribution.
It seems like part of the logic at Yahoo fits with what also helped drive the change in the work-from-home policy – that there were laggards that needed to be weeded out.
These might have been people that weren’t producing results, were taking advantage of the system, coasting, working on side projects, etc. The stack rank in this case might be appropriate if the objective is to weed these folks out. And maybe with a company of that size and those perceived issues, perhaps that’s the way to go.
I wonder too if another part of the logic provides commentary on Yahoo's managers. Could their managers, all the way down the line, have been doing a better job over time to make sure that forced stack rank was unnecessary? Is that what this is all about?
If managers are not capable or not willing to manage performance, distinguish high performance from poor performance or deal with performance issues, forcing them to use a methodology like the stack rank might be the way to go. With stack ranking, managers are forced to think about and differentiate between employee performances – an admirable goal even if you dislike the implementation.
I certainly don’t have an answer here as to whether or not it’ll work for Yahoo… it might meet some of their objectives, even if it ruffles a few feathers.
Certainly the quantitative part of me likes the idea of rankings, but I can see the clear issues with the forced distribution. Most management systems I’ve seen work toward some common rating scale to evaluate people. And when there’s only so much to go around for bonuses and promotions, you do want to separate, and reward, your best. You also need to document performance issues that might surround your worst in the event you move toward termination.
But can be these done in a better way.
Ultimately, this is where a good performance review process comes in… you know, the one that’s interactive and on-going. The one where managers are giving both positive and critical feedback, and employees are open to receiving it. The one where managers are engaged enough to know (and easily document) what's going on. The one where they’re not just celebrating the high achievers, but also working to improve everyone’s performance for the better.
Is that world out there?