How to choose a performance rating scale that actually works

Rating scales used in employee performance reviews have a reputation problem – and it isn’t a new one. Rating scales are often seen as outdated, overly rigid, or lacking nuance in how they capture employee contributions. And, to say it lightly, employees don’t tend to love them, either. They feel reductionist because they are just that. Rating scales are designed to be a simplified representation of an employee’s work and they exist for the sake of comparison. Some organizations have done away with ratings for those reasons, but then come back to them in time because they are necessary for making decisions.

At Culture Amp, we don’t believe there is anything inherently wrong with rating scales themselves, but the way they’re designed and deployed can change how employees perceive and buy into them within organizations.

As industry analyst Josh Bersin puts it:

“Let me simply say that after a decade of discussion on the topic, the concepts of ratings themselves are not the issue. Organizations need to make decisions about people…and these decisions themselves are essentially evaluative by nature.

The key today is to use lots of data and feedback to make these decisions; do them in a transparent and fair way; clearly communicate what is valued in the company; and give people visibility into others’ goals and projects.”

Bersin’s quote here is, in media terms, outdated. And yet, this industry expert’s commentary on what is needed to capture the full breadth of an employee’s impact at work still rings true. Managers need to pull data about their direct reports’ performance from several sources across the business. They need to tie that data back in with company goals and team goals. Performance evaluations, whether a rating scale is used or not, serves the ultimate purpose of employee growth and development – so that your people can align what they’re doing at work with what the business needs.

Our latest research on sustainable high performance shows that how you rate performance plays a central role in how employees experience growth, fairness, and confidence in your company’s future. So, ratings impact employees’ broader experience at work too.

In this blog, we’ll explore why ratings are still an essential part of performance management, highlight recent data from our research, and provide guidance on how to build a rating scale that’s fit for the future of work.

Why rate employee performance?

It may seem obvious, but at its core: ratings help companies make decisions. From compensation to promotions to development planning, organizations need a structured way to evaluate performance fairly and consistently. This is especially true for larger enterprise-level organizations.

Ratings also benefit employees. When assigned transparently, ratings function to clarify expectations, signal pathways for growth, and help people understand what “good” looks like in their role. According to our recent research, employees who receive high performance ratings also report stronger experiences with feedback, support, and resilience - all critical ingredients for ongoing development.

Done right, a performance rating scale helps differentiate excellence, drive transparency, and ensure people are rewarded fairly. But “done right” is doing a lot of work here. The conditions under which a company rolls out a rating scaled approach to performance alters how it will be received and trusted by employees.

What makes rating scales hard to get right?

When talking with employees about ratings, we tend to hear about their bad experiences with outdated tools or flawed processes. For example, forced rankings or vague numeric scales can result in poor data and lower morale. Here’s why it happens:

• Poor scale design: Companies like to get creative with performance rating scales. Many scales fail to offer meaningful distinctions between different levels of performance, while others are too subject to interpretation.
• Rater bias: With any performance review format, we see issues with bias. Unchecked biases like leniency (giving everyone a high score) and centrality (rating everyone as average) skew results.
• Lack of training: Not all managers have done performance reviews before and are thrown in without much training. Managers may not know how to differentiate fairly, especially when reviewing soft skills or team contributions. When managers struggle to distinguish between “great” and “truly exceptional,” we see inflated ratings and missed opportunities to recognize real impact.

Employees are 50% points more engaged if they feel their performance is evaluated fairly. On the flip side, they’re also 88% more likely to leave within 12 months if they strongly disagree that their evaluation was unfair.

Knowing that how an employee is rated could be a predictor of satisfaction with and at work, we can make more intentional decisions about what type of rating scale and process to use.

Three ways to design a rating scale that works

1. Ensure validity and spread

If your scale can’t differentiate between good and great, it’s not working.

• Spread refers to how well your scale picks up nuanced performance differences. This is also known as variance, differentiation, or range. Without it, you’ll see ratings cluster at the top. In fact, our data shows that without structured calibration, leniency bias can lead to rating inflation, making it harder to reward top-performing employees appropriately.
• Validity means your ratings should reflect the behaviors and outcomes you are trying to assess. If your rating doesn’t reflect the thing you’re trying to measure, it won’t be taken seriously - neither by employees nor managers.

2. Customize your descriptions

Your rating scale and its descriptions should reflect your organization’s brand, values, business needs and career pathing. The more specific you are in defining each response option via description, the better and more consistent your raters will use the scales. Creating clear, differentiated descriptions becomes even more critical as response options increase (e.g., five options versus three).

Some companies customize in order to reduce various forms of bias. We encourage customization, and the Culture Amp platform allows you to reformat your questions and rating scales to address your organization’s goals and minimize the biases you may have seen in previous performance review cycles.

3. Make the process transparent

There is no question that transparency builds trust. The statement rings particularly true when it comes to performance management. With whatever rating scale you do select, let your people know about it. Or better yet, have leaders communicate what’s being measured in the performance process, why it matters, and how the system works. Moreover, when training managers on how to evaluate their direct reports using the rating scale, also give them guidance on setting expectations with employees ahead of time (like in regularly scheduled 1:1s).

Things you can do now:

• Add more detail to top-end ratings to help managers identify varying levels of excellence
• Run calibration sessions to ensure consistency
• Use detailed, role-relevant descriptions that help managers distinguish real differences in performance.
• Consider eliminating neutral options to reduce centrality bias and encourage decision-making
• Schedule a thorough internal communication plan to roll out any change to a rating scale so employees are well informed about how they will be measured

What we learned from 826,000 reviews

In late 2024, we launched a large-scale research project to better understand how performance ratings relate to employee engagement. To do that, we needed to standardize how companies were rating performance. In the process of standardization, we learned a LOT about rating scales.

For example, we learned that most companies using Culture Amp to measure Performance are using a 4-point scale (52%) or a 5-point scale (31%).

In 2023 we found a similar distribution, so there has been relative stability in scale usage for companies that use Culture Amp. That said, to conduct broader research we still needed to standardize the multitude of scales used by these companies.

We scoped our analysis to valid performance reviews (excluding probation reviews, 1:1s, and practice/demo accounts) conducted between August 31, 2022, and August 31, 2024. Then we focused on rating scales used by at least 2 companies and/or over 400 employees, resulting in a sample of 1517 companies based on company size, representing over 462,000 employees.

To make comparisons meaningful, we standardized any performance rating scale used by over 400 employees or more than 1 company into this 4-point structure:

1. Needs development - these underperforming employees are typically a small proportion of the workforce (median is 4%)
2. Consistently meets expectations - these are solid performers and the plurality are rated in this bucket (median 45%)
3. Often exceeds expectations - these are great performers (median 33%)
4. Sets a new standard - these are truly high performers (median is 8%)

What we learned

Performance rating distributions vary slightly by region.

In North America, for example, there is a higher percentage of employees rated as setting a new standard than in other regions.

These subtle regional differences could inform your rating scale selection. If your company is based in North America and you want a scale that helps to prevent leniency bias, you might consider one that adds more variation across the top part of the scale.

On the other hand, if your company is in APAC, which has the highest percentage of employees in the ‘meeting expectations’ bucket and you’re looking to boost performance, you may consider a rating scale that is more focused on behaviors. More on that below.

We also looked into verbiage and found, unsurprisingly, that the words used in rating scales matters.

There are taboo words when it comes to performance scales

Compared to scales that use more neutral terminology, scales that use the term “average” tend to be top-heavy. That is, we see a higher concentration of ratings clustering at the higher end of the scale.

Rating scales that do use the language of “average” typically look like this:

• 1 = Below average
• 2 = Average
• 3 = Above average
• 4= Exceptional

So why the skew? We assume it happens because managers worry that telling their employees they are average can be demotivating. It could be the result of psychological anchoring, where the presence of the word "average" acts as an anchor, influencing reviewers to avoid selecting it and instead choose buckets that are perceived to be slightly above or below average. The word “average” may create pressure on managers to rate their direct reports more favorably – even when that isn’t warranted.

Our advice: Use “meets expectations” or “consistent performance” instead of ‘average.’ These terms are more grounded, growth-oriented, and easier for managers to use fairly.

Examples of rating scales

A 4-point scale that balances fairness and accountability

• Needs development
• Meets expectations
• Exceeds expectations
• Sets a new standard

This format eliminates the middle ground and helps distinguish between employees making a strong contribution and those at peak performance.

A 5-point scale that promotes differentiation at the top end

• Not meeting expectations
• Meeting expectations
• Sometimes exceeding expectations
• Always exceeding expectations
• Setting a new standard

This encourages managers to differentiate performance among high-performing individuals while simplifying the lower end.

A behavioral observation scale for evaluating competencies that are tough to measure

Use frequency-based language like:

• Rarely observed
• Occasionally observed
• Consistently observed
• Always observed

This approach is helpful for evaluating competencies like communication, collaboration, or inclusive behaviors.

Final thought: design with intent

A performance rating scale is and ought to be considered a big decision for an organization. It’s a signal of what your organization values, how it makes decisions, and how it plans to grow.

Leaders can design the performance evaluation process to foster clarity, equity, and even to propel high performance forward. And it really does start with the scale.

Want more tips on building a performance system your people trust? Check out our full report on The Science of Sustainable High Performance, where we share data-driven insights on building sustainable high performance and empowering your workforce for the future.

Learn about the science behind sustainable high performance

Download report