Bell Curve Performance Ratings: Why HR Experts Are Moving On

When companies run performance reviews, they produce a bell curve of manager ratings. 

This bell curve, known as a “normal distribution”, shows most employees as average, or “Meeting expectations.” A smaller number of employees fall to the left or right. They may get a “Needs Improvement” rating, making them candidates to be exited from the company. Others may be “Exceeding expectations.” They’re more likely to get promotions and rewards.

Bell curves are what you get when you measure additive variables. These are variables that don’t affect one another, like height, or IQ. In the early days of performance reviews, a bell curve made a lot of sense. Work was solitary and repetitive.

But today’s work is different. It happens in networks, using tools like Slack, Teams, and Zoom. It’s collaborative and creative. Does a bell curve still make sense?

Talent follows a power law (And Why 9-Box Grids Capture It Better)

There’s a secret about talent management that HR doesn’t understand, but academic researchers do: talent follows a power law.

Why is this? Because work today isn’t solitary or repetitive, it’s creative and team-driven. There’s a lot of variability in how employees plan a marketing campaign or design a new software feature.

That means we’re not measuring additive variables anymore, but multiplicative ones. That’s why we have “10X” software engineers. Or account executives who 3X their quotas and find time to lift the rest of the team up too.

You can test this for yourself. Think of 8-10 people who you work with. My bet is that there are one or two of them who are so amazing, you wish you could clone them. That’s the power law of talent.

We’ve been thinking about talent the wrong way

Manager-driven performance reviews aren’t built for this new world of work. One reason is that 10X employees create a lot of their value by helping the people around them. Managers often lack the visibility to see this.

But here’s the bigger problem: the bell curve measures manager ratings, not actual performance.

By squashing the power law into a bell curve, we hide the true impact that top and bottom performers make. And we make many of our talent decisions with this misconception at the core.

That includes compensation, which under-rewards top performers and over-rewards bottom performers. Interestingly, in jobs where performance is both measured and transparent, compensation tends to follow a power law. That’s why a superstar like Stephen Curry makes approximately $59.6 million in the 2025-26 season (Basketball Reference, 2026), compared to the median NBA player who makes around $2 million.

How to Replace Bell Curves with Power Law Distribution in Your Organization

Understanding the problem is one thing. But how do you actually move away from bell curves and implement a performance management system that reflects power law distributions?

Step 1: Stop Forcing a Distribution

The first step is to acknowledge that your organization doesn't need—and probably doesn't want—a forced distribution. If your true data shows that 25% of your team are exceptional performers, that's okay. A forced distribution prevents you from seeing and rewarding this reality. Instead of using bell curves, let performance data tell its own story.

Step 2: Use Multiple Data Sources

Manager ratings alone won't reveal power law distributions. You need multiple perspectives: peer feedback, objective metrics, project outcomes, and network analysis. When employees collaborate across teams, their impact becomes visible through organizational network analysis (ONA). This shows who the connectors and amplifiers are—the people who make others more effective. These are often your power law performers.

Step 3: Run Proper Calibration Sessions

Calibration sessions should focus on consistency and fairness, not on fitting a predetermined curve. Bring managers together to discuss their ratings, compare notes on what "exceeding expectations" means in practice, and ensure accountability. The goal is to reduce bias and hidden assumptions, not to force a mathematical distribution.

Step 4: Reward Top Performers Like Superstars

Once you see your power law distribution, act on it. Your top 10-25% of performers drive disproportionate value. Compensation, promotions, flexibility, and opportunities should reflect this. You can't keep top talent if you compress their rewards to fit a bell curve. Use clear performance review conversations and language to communicate how top performers are valued. Treat your 10X employees like the superstars they are.

Step 5: Address Low Performers Directly (or Refer to Performance Improvement Plans)

A power law distribution often shows a clearer picture of bottom performers too. Rather than creating a "needs improvement" category to fill a curve quota, use this data to have real conversations. Some people need development; others may not be the right fit. Either way, you're making decisions based on data, not on how many slots you allocated to the bottom tier.

The companies winning the talent game have already made this shift. They're not constrained by bell curves. They see their talent for what it really is—a power law—and they're building compensation, promotion, and development strategies around that reality. It's not just fairer; it's more effective.

Maybe it’s time to start treating our 10X employees like the superstars that they are.

Ready to move beyond bell curves? See how Confirm's performance management approach helps you identify and develop talent based on actual performance patterns, not forced distributions.