Like a broken clock that's right twice a day
Last month, I wrote a nerd processor called Three things I was completely wrong about. In the piece, I shared three times that my guesses about data were absolutely wrong:
- #1: I was wrong about remote work
- #2: I was wrong about who says "I told you so"
- #3: I was wrong about small talk
One thing I was right about, however, is that nerd processor readers would like reading about my mistakes. You shared the piece in record numbers.
So for this week, I decided to share another thing I was completely wrong about: performance reviews.
This one is painful.
Performance review takedown
Writing for Fortune in 2014, I published The abrasiveness trap: High-achieving men and women are described differently in reviews. I shared quantitative data that demonstrated that men and women are described differently in performance assessments even when they get the same overall scores. Women are more likely to get personality feedback and they are described more harshly overall, even when they're top performers.
Instantly viral, the piece started a revolution. Within 48 hours of publishing, I had received thousands of emails about it. In the years to come, one academic researcher after another replicated the findings. Performance management as we knew it had a major bias problem.
The article also catapulted my career forward. I wrote it shortly before starting Textio, and it was key in our early growth. Because of the piece, I met the world's most influential HR leaders, which paid off big time for Textio. I gained a reputation for being the person to take down performance reviews with real data.
Fast forward a few years. I was the CEO of a growing company. People loved Textio! Before long, we had enough employees that we needed to figure out how to make systematic compensation decisions. We had a limited budget. We wanted a system that was both fair and effective to drive company performance.
Uh oh. You know what's coming.
A performance review by any other name
Because I understood the issues with conventional performance management systems, I resisted implementing them. When we were small and I could see everyone's work personally, this worked okay. But as we grew and tried everything other than traditional performance reviews, it all failed.
We tried promoting people based on learning. We didn't want to be the kind of company where people competed with each other to do high-profile work. We thought we could mitigate this by promoting people based on the skills they had demonstrated rather than the achievements they'd racked up.
Maybe in utopia, that would have worked. In the real company we were building, though, a lot of people stopped focusing on achieving things. Productivity slowed way down.
We tried having short career check-ins constantly. Having one high-stakes career conversation per year as the only time an employee receives feedback is terrible. Everyone stresses out. It is common for employees to end up surprised by what they hear.
You know what doesn't help the situation, though? "Replacing" the high-stakes convo with a bunch of smaller "performance check-ins."
When you have constant check-in discussions, every conversation becomes trivial. In fact, without an annual review as a forcing function, some people just skip the meaningful engagement entirely. Ask me how I know.
The reality is, managers need to give feedback in everyday 1-1s, AND you still need the 1-2x a year intentional pause to mutually take stock of how it's going.
We tried training managers.
Look, school is awesome. People can learn a lot from great teachers. I love school.
And also I've never met a great manager who got that way because they took a management training class.
We tried to teach managers how to have performance conversations many ways, many times. I designed and ran manager workshops myself. Our HR team tried too. We worked with external experts. In the end, very little of the training stuck.
In every single case, employee feedback was the same. They didn't understand how their performance was assessed. They didn't understand how to get promoted. The managers who were already excellent stayed excellent; those who were struggling continued to struggle.
While you can teach people to remember a couple of tactics, I discovered this fundamental truth: The only way to get managers to review people effectively is to build a system where they have to review people effectively, because 1) performance tools encode the expected rubrics and behaviors and 2) managers' own reviews depend on their adherence to the system.
The bottom line: I've had many humbling experiences as a leader, but few as painful as this one. We tried so many alternatives. None of them worked.
In the end, do you know what worked best? Performance reviews. We used tools (like Textio) that encoded the manager behaviors we expected, and then audited the output for quality and equity.
The latest AI tool for performance management isn't coming to save you. If you want a team that is both productive and fair, you're going to end up with performance reviews.
Kieran
My Viral Data Stories 101 course is half-price for nerd processor subscribers for five more days. Offer ends Aug 31!
My latest data stories | Build like a founder | nerdprocessor.com