"It sounds like we collectively agree that Floyd is not performing at an exceeds level. The achieves rating is more appropriate." the VP says. "Sonia, do you agree?"
Floyd's manager Sonia looked dazed. "I guess?"
The VP nods. "Good. HR, please update Floyd in our system? Ok, moving on."
Sonia looks like she might say something, and then sits back.
I feel sorry for Floyd. I don't know a ton about him, but it's entirely possible that he deserved that exceeds rating. Sonia's incompetence more than anything drove Floyd's rating to be lowered.
I make a mental note to bring this Sonia situation up once the lower level managers leave the room, and we have a chance to discuss them.
Sign up for Scarlet Ink by Dave Anderson
Weekly leadership advice from an ex-Amazon GM and Tech Director. Most articles have substantial previews available for free members. Paid members can read all current and future articles, and are statistically more awesome than the average person.
No spam. Unsubscribe is self-service at anytime.
In this article, I'm going to spend some time talking about the nuts and bolts behind the rating process at companies, and how those ratings are reviewed.
I'll talk through a couple of scenarios, which are amalgamations of specific experiences I had at Amazon and Meta. To be clear (since people love to beat on Amazon for their confrontational culture), this is not just an Amazon experience. While people at Meta (Facebook at the time) were perhaps a bit more polite, they were just as careful to defend their team members, and challenge their peers, as those at Amazon. Everyone's human everywhere.
Finally, I'll discuss what we can learn from understanding the rating process, and what it means for your long-term career success.
Is this applicable at your company?
Unless you're at a tiny company, you most likely have some form of a rating system. It might be explicit ratings (like Exceeds at Amazon, or Meets Most at Meta), or a quantitative (number) system, or perhaps it's a verbal input into a raise / promotion discussion.
The purpose of these systems is simple. Basically, all companies agree that they want to reward their best employees with raises and promotions. Those are the employees with good / great ratings. Those employees will hopefully stick around due to those rewards, and become future leaders in the company.
Companies also agree that they need to coach or remove their worst employees. Those are the employees with poor ratings.
How exactly each company executes on this process is different, but the themes and learnings still apply.
For example, most companies have some way of throttling good reviews. Managers are only human, and most managers want to reward their team members. At least the ones you'd want to work for. Given an unlimited budget, many managers would default to an Oprah style, "My team is spectacular, and they all deserve promotions! You get a promotion, and you, and you! Everyone gets one!"
In small companies, this is held back by a combination of social pressure and potentially an interest in doing the right thing for the company.
Once a company grows, these rewards are buried in bureaucracy, so the social pressure is decreased. It's hard for anyone to track / notice if one manager slips in several raises or promotions.
And at larger companies, doing the right thing for the giant, faceless company feels less important than doing the right thing for your friends on your team.
In practice, this often means a set percentage limit on the numbers at each rating, which creates pressure to lower some ratings if teams have been overly positive about the performance of their team members. I'll walk through some example situations below.
The same applies for poor reviews. Most companies have a way of encouraging some managers to rate their employees as underperforming. Again, managers would rather not do it on their own, so the company needs to figure out some way of pressuring managers to watch for those who aren't the best employees.
In the end, you do your job a certain way. Your manager thinks about how you've performed, and tries to assign a rating. And then corporate bureaucracy can interfere with your manager. Below is one way how it happens.
Not all companies have ratings meetings, and Amazon doesn't always have it either. But at a high level, this is how things happen. Ratings are assigned, some level of managers and skip managers discuss things, to try to fairly allocate ratings and raises and rewards. And the below are two examples of how it goes.
A discussion about Chelsea.
"Neal, tell us why you rated Chelsea as Exceeds." our Director asks.
Neal nods, and shuffles his papers nervously. Neal never looks confident, and today is worse than usual. Neal is a line manager of a software development team. He has a team of 6 engineers, and he's being asked to explain his ratings to a room full of his peers, and leaders up to the Director level.
"Chelsea completed the XYZ migration project, it achieved all our goals." Neal begins. He essentially rambles through a list of deliverables on a variety of projects. And then stops talking. A shopping list of project names isn't the best way to sell someone as being a top performer.
"I still don't know how that makes Chelsea an exceeds Neal," Minnie states. Minnie is a senior manager, and the peer of Neal's manager. Minnie is defensive of the ratings in her organization, but seems to take joy in poking holes in the ratings of other teams. "That all sounds like she was just achieving her goals."
Neal doesn't speak up quickly to counter Minnie's point. He's looking at his papers, hoping for inspiration. "She did quite well on her goals." he belatedly states. He clearly doesn't have the information at his fingertips to be able to refute the achieving vs. exceeds argument on her goals.
Smelling a bit of blood in the water, Wendy, one of the managers in Minnie's org speaks up.
"I was on the XYZ migration project." Wendy says. "I didn't feel Chelsea was highly impactful on the project."
Now our Director speaks up. "I did attend one of the status reviews. I asked Chelsea a question about a specific metric, and she was unable to answer my question. Are you sure she's on top of the details?"
I know that this was probably our Director's only interaction with Chelsea. But Neal still doesn't speak up.
"We should be careful about leaning too heavily on a single anecdote." I suggest to the room. "I'm sure that Neal has more experience with Chelsea's overall performance than we have with our brief interactions."
Our Director nods. We've had the anecdote discussion before. It's risky to let singular anecdotes heavily impact someone's rating.
Neal nods. "Yes, she is usually a great leader. She mostly leads inside our team though, not with other teams. Which is why you wouldn't have seen it enough."
I sigh internally. Neal is digging a hole for Chelsea.
"Ah, I see." said Santos. He's another senior manager. And he doesn't speak up a ton, so people listen. "At Chelsea's level, leading outside of her team is a part of the expectations for exceeds. It sounds to me like she's really an achieves employee."
A good amount of positive muttering at that statement. We're all aware that too many people were rated Exceeds before entering the meeting, so some people will need their ratings to be lowered. Unfortunately, that means that almost everyone in the room is incentivized to make certain those lowered ratings happen outside of their organizations.
"What do you think, Neal?" our Director asks.
Neal looks nervous. "Yes, that makes sense."
"Can you explain to Chelsea how she could improve her performance?" the Director asks. "Regarding how she needs to lead more outside of her team?"
I see this as clear code for, I'm not confident in your skills Neal, and I'm being explicit with you in hopes that you don't let Chelsea down more.
"Yes, I can explain this." Neal says.
The rating is updated, and we move on to the next employee.