Phyllis Korkki of the New York Times wrote a piece this Sunday for her Workstation column called Invasion of the Annual Reviews. It emphasizes the risks and downsides of annual reviews, and she quotes me quite a bit -- I didn't realize how much until the piece came out. But she got it right, as she always does (I have worked with her before, she is very professional and very careful). As Phyllis had only so many words to work with, and she weaves in the perspectives of others including an interesting clarifying statement from a Yahoo! spokesperson asserting they are not doing forced ranking, but rather "“Our system lets employees understand how they are performing relative to expectations (exceeding, achieving or missing), and there are no hard and fast rules.”
I got a couple emails from friends -- one congratulatory and the other that disagreed with me -- suggesting that the piece meant I was opposed to all annual performance reviews, not just bad performance reviews. I confess that I have raised the question of if they should be abolished before and two parts of the piece may have helped to fuel this impression:
This paraphrase from Phyllis:
Professor Sutton is wary of rankings and yearly evaluations in general. Many organizations, he said, would be better off if they provided continuous feedback, with formal evaluations coming into play mainly if a worker is being eyed for promotion or has shown substandard performance.
And the closing paragraph:
“If performance evaluations were a drug, they would not receive F.D.A. approval,” he said, because “they have so many side effects, and so often they fail.”
These are accurate representations of my perspective. But I think it is important to make clear that I am not opposed to all performance evaluations, only bad ones. -- and unfortunately, they are done badly more often then they are done well. So, what are the hallmarks of good performance evaluations? Consider three:
1. Is what happens during that annual conversation and evaluation woven into the fabric of every day life, or as I was quoted in the piece "“this weird form you fill out every year that has nothing to do with everyday life.” So if your boss gives you positive feedback all year, or doesn't give you feedback at all over the course of the year, and then you get bad review, the boss isn't doing his or her job. In organizations that generally do evaluations well, I think of McKinsey and GE, although they do yearly evaluations, there is also an emphasis on teaching and nudging leaders to constantly give their direct reports regular coaching and feedback. There are two tests here.
If you are a boss, do people often seem surprised by the feedback you give them during annual reviews?
Does the review conversation seem uncomfortable and unnatural, something that bears no relationship between the feedback and coaching (or lack of it) that happens throughout the year.
2. How is excellent performance defined and measured? In some firms, even though the goal is to create collaboration and information sharing, "stars" are nonetheless anointed solely on the basis of individual achievements (and for more senior folks, on the basis of their team of department's contribution, not on their contribution to the overall success of the organization). In too many companies, although leaders hope for cooperation, they reward backstabbing, stomping on others on the way to the top, and other flavors of dysfunctional internal competition. So the question of "who is a superstar here" is the one I ask leaders all the time, I want to know "are they the people who are great individual performers AND who help others succeed -- or are the people that ignore and even undermine their colleagues?" This is a theme that Jeff Pfeffer and I wrote about in our books on The Knowing-Doing Gap and Hard Facts, and I revisit in Good Boss, Bad Boss. Note also that both GE and McKinsey are usually very careful to anoint the right kind of stars -- and so are a lot of other organizations I admire including IDEO, P&G, and the Cleveland Clinic.
3. Finally, while I believe strongly in weeding out bad apples and rewarding good behavior, what are the assumptions about the nature of a human organization? Do leaders believe that there will always be a certain percentage of losers who will need to be weeded out and a certain percentage of amazing performers who deserve the lion's share of rewards. This is the assumption that drives many stack ranking and one I don't like and that is contrary to the evidence. It is also contrary to the logic of the quality movement (which weirdly, GE who at least had this system in the past embraced as well). Imagine a manufacturing system or worse yet a hospital where you assumed that year over that there would be a 10% defect rate, 70% of the work would only be OK, and only 20% would be great. Unfortunately, that is the implication of how stack ranking is done in some places, even other wise very well managed places.
Here is how things can play out when this assumption is implemented in a misguided way. One senior executive I know had spent years building a great 12 person team. He hired carefully, he weeded out a few rotten apples, and has the team humming. Then, the company hired a head of HR who had unwavering faith in firing the bottom 10% each year -- he was required to fire one of his people. In essence, following the logic of Deming and other quality gurus, he had built a system with no defective parts -- but was required to throw one away. He refused to do it and quit -- which, as he told them, solved their problem, because now they didn't have to fire a member of his team (which they did anyway, as he "didn't count.")
I hope this clarifies my views. Note that the NYT's piece also talks a bit about Adobe's recent efforts to abolish annual reviews and replace them with frequent check-ins. Huggy Rao and had in-depth conversations with Donna Morris, the brave executive who led this change, and we talk about the details in our forthcoming book Scaling Up Excellence. The upshot is that Donna and her colleagues worked on shifting the focus from the mechanics of annual reviews to the nuances of daily interactions between leaders and their teams.
I am quite interested in the path that Adobe is taking because the result may be that, indeed, there is a better alternative to yearly performance evaluations. Indeed, note that, in the places that do them well, they are woven into the fabric of everyday life -- so perhaps an interesting test of how bad AND how good your annual performance evaluation is "what would happen if we didn't do them?" Oddly, if you are doing them really badly, then I would argue that doing nothing might be better. Just give employees an envelope with their yearly raise and skip that stilted dysfunctional disingenuous yearly conversation. And if you are doing things really well, then perhaps you will find out that you don't need them after all because people are getting such regular feedback and coaching that the formalities are a waster of time in most cases-- think of all the time and money you would save!