How to Choose the Right Performance Review Rating Scale for Your Company

In our 2024 State of People Strategy Report, performance management ranked as a top priority for HR teams for the first time. But performance reviews, a key element in the performance management toolbox, can be tricky. Poor quality feedback can make employees more likely to quit, according to a 2023 survey by Textio. 

An effective performance rating scale can help, by guiding managers to give consistent feedback. But choose the wrong performance rating scale — or fail to train managers to use it correctly — and you’ll end up with skewed ratings, frustrated managers, and disillusioned employees. 

So, we sat down with Lattice’s senior talent strategist, Matt Raskin, to get into the nitty-gritty of choosing the right rating scale for your organization. Below, we dive into the pros and cons of different performance rating systems; whether you should use numbers or words in your ratings; and the mistakes that Raskin has seen too many companies make when it comes to rating employee performance. 

What is a performance review rating scale?

A performance review rating scale is a system for evaluating employee performance by assigning a grade to different elements (such as quality of work, job knowledge, or customer focus, for instance). 

Employee performance rating scales can be numerical (e.g., 1 = Poor, 2 = Meets expectations, 3 = Exceeds expectations) or textual (e.g., each rating is described in a word or phrase). 

The goal of using a performance rating scale is to create a more standardized, systematic approach to the performance review process. The scale guides managers to evaluate each team member using the same criteria. It also ensures that human resources teams and business leaders have the data they need to make informed decisions about employee performance. 

4 Steps to Choosing the Right Performance Review Rating Scale

Ready to start using a rating scale? You might feel a little daunted by the range of choices available to you. Should you use numbers or words? Do you need a Likert scale, or would BARS be more effective? Where do you even start? Here are our tips: 

1. Figure out your end goal.

Raskin has seen too many companies start by looking at what other organizations are doing. “Oftentimes, we’ll see companies say, ‘Well, so-and-so did this, so we should do it too.’” 

Instead, he urged HR teams to start by thinking about their own unique company goals and values, and why they want to introduce a rating scale into their employee evaluation system in the first place. “We should think about why we’re doing it, and how it impacts our organizational realities, our culture, and the intentions behind it,” Raskin explained.

So start by identifying the “why” behind introducing a rating scale. Are employees unsatisfied with your current performance evaluation process? Maybe you hope to increase employee engagement. Perhaps you’d like to standardize your performance reviews more, or you simply want more metrics on employee performance to help with decision-making. 

It’s also key to acknowledge that performance ratings don’t occur in a vacuum. Your choices on how to rate performance will affect how you compensate employees, the values and behaviors you’ll incentivize, and your other people programs. 

Raskin also advised HR professionals to invite input from outside the people function before choosing a rating system. “Make sure you’re talking to other parts of the organization, talking to your leadership team, some of your executives, leaders, managers, to get a sense of what they’re looking for, what they’re experiencing, what would be helpful for them. What have other people seen be successful? Why was it successful?” 

2. Define and communicate your rating criteria.

The specific skills, behaviors, and performance goals that you measure will depend on your overall organizational and team goals. (If you need help figuring those out, check out Lattice’s free ebook Implementing Strategic Goals for Organizational Success.) 

Regardless of which aspects of performance you’re rating, the most important thing is that you all agree on what each rating means. If you rate an employee’s performance as a 3 out of 5, what does that mean? Or, if you use a word-based rating system, what does it mean to say that an employee’s performance is “excellent”? Once you’ve decided on your rating criteria, share it with employees and managers ahead of time.

You’ll also need to provide managers with ample training on how to use the rating system. Otherwise, Raskin warned, you’ll see some managers assigning middle ratings for everyone, or giving everyone on their team nothing but top scores. Bias can cause unintentional harm if managers aren’t made aware of it.

{{rich-highlight-1}}

3. Consider the pros and cons of different rating scales.

At this point, you can start looking into the different rating options (more on those later). Some general words of advice from Raskin here: 

• Keep it simple. Too many HR teams fall into the trap of adding complexity to their rating scale for the sake of complexity, Raskin cautioned. If you decide to add additional complexity (using weighted measurements or moving from 5 levels of performance to 7, for instance), Raskin recommended that you “understand why you’re adding it, and what success looks like in each layer of complexity.”

• Consider how your ratings will impact your development conversations. Ratings aren’t supposed to be a one-and-done grading system. Rather, they’re intended to help improve employee performance as part of an ongoing performance management process. Ratings are a tool that managers need to be able to use to guide employee development conversations. And the rating you choose will impact the tone of those conversations. 

“You want to think about the type of measurements you’re looking for and what kind of conversation they should be driving,” Raskin said. “Does a rating of three drive a better conversation versus ‘You’re meeting our expectations’?” 

4. Test and tweak your rating system.

Before you roll out your rating system, consider how you will evaluate its effectiveness and refine it over time, Raskin advised. “We should have a theory of what we believe will happen. We should have goals. And then we should be able to measure our results against those goals, and solicit feedback on the effectiveness of the system.” 

Once you’ve defined how you’ll evaluate the ratings, you may also want to start with a small pilot program with a single team or small group of employees before rolling your new scale out to the wider organization. 

5 Types of Performance Review Rating Scale, With Examples

Whether you’re looking for a simple approach or something more nuanced, there’s a rating scale for you. Below are five common types of scales, each with its own example to help you determine what might work best for your team.

The Five-Point Performance Rating Scale

A five-point performance rating scale is just that: a rating scale with five levels. It might be numerical (rating employee performance from one to five) or textual (e.g., rating employee performance as “poor,” “fair,” “above average,” “good,” or “excellent”). 

Here’s an example: 

Among Lattice customers, the five-point performance rating scale is the most popular: This type accounted for 49% of all scales for review cycles launched in 2023. According to Raskin, “Five-point scales are great because they’re simple and intuitive. They’re also efficient in terms of setup and structure.” 

So what’s the catch? As Raskin explained, the risk with a five-point rating scale is that it can be unclear, leading to skewed data.

“With numerical ratings, you often see central ratings, where people might lean toward the middle rather than giving a four or a two if they know it will have an impact. This can lead to midlining, where a rating that should be a two becomes a three,” he said. “Additionally, there’s often overinflation, where everyone ends up being rated a four or five. This lack of clarity can result in either centralization or inflation of ratings.”

Likert Scale

If you want to add greater clarity, the Likert scale might be a better option. Likert scales are a type of five-point scale where the rater is asked to respond to a statement with a rating out of five, as in the following example: 

Raskin said the advantages of Likert-scale questions are that they’re “easy to understand, can be flexible, and are a good way to measure attitudes.” For example, if a performance review includes a statement like “This employee is a strong team player,” a Likert scale would allow the reviewer to express their own perceptions through agreement ratings ranging from “strongly agree” to “strongly disagree.” 

The downside is that Likert ratings are “subjective,” Raskin warned. For this reason, managers will need to be trained to watch out for their own biases when using a Likert scale. 

Behaviorally Anchored Rating Scale (BARS)

A BARS is like a five-point scale with more specificity. Instead of being asked to rate each aspect of overall performance in general terms (e.g., “poor,” “excellent,” etc., or a score out of 5), BARS defines each rating point with specific behaviors that illustrate what it means to perform at that level.

For instance, here’s a behaviorally anchored rating scale appraisal of how a customer service employee handles customer complaints.

• 1 (Unsatisfactory): Ignores customer complaints or responds rudely, making the situation worse. 
• 2 (Needs Improvement): Acknowledges customer complaints but does not offer a solution. 
• 3 (Meets Expectations): Listens to customer complaints and provides a standard solution that meets customer needs.
• 4 (Exceeds Expectations): Listens to customer complaints, uses initiative to come up with multiple solutions, and follows up to ensure satisfaction.
• 5 (Outstanding): Anticipates potential complaints and addresses them proactively before they escalate, consistently delighting customers. 

According to Raskin, the main advantage of BARS reviews is that they add greater clarity by including specific behavioral examples for every performance dimension. “They provide more detail on what the behaviors are and what ‘good’ looks like. BARS also tend to be more consistent, and they reinforce behavioral aspects well, which is beneficial for employee development.” 

On the flip side, he added, “They can take a while to develop. Keeping everyone aligned on what a good assessment or BARS looks like can be a lot of work. They can be a big lift up front — you might need to bring in a lot of different people and work through various ideas.” 

Plus, once you’ve set up a BARS, it can be hard to modify it. With a standard five-point or Likert scale, you can simply change the wording of a specific statement or aspect of performance, but with a BARS you’d also need to redefine the desired behaviors. “If your goal is to create a less complex system or approach, implementing BARS might not be the right direction,” said Raskin. 

Graphic Rating Scales

A graphic rating scale is a sliding scale that gives a visual representation of employee performance. Typically, reviewers can select an option along a horizontal line or bar. The advantage of a graphic rating scale is that it’s simple and intuitive to use.

However, graphic rating scales still share the same disadvantages as other rating scales — they can be subject to manager bias and may give a limited view of performance (especially if not coupled with written feedback). 

Custom Scales

Unlike standard rating scales, which might use generic terms like “meets expectations” or “exceeds expectations,” a custom scale is designed to reflect an organization’s unique goals, values, or competencies. If your organization already has concepts around performance and seniority as part of your company culture, you might use that existing wording instead of the more standard terminology. 

For example, you might use a scale like this:

• ‍1 - Unacceptable: Consistently fails to meet critical job requirements.‍
• 2 - Developing: Shows potential but improvement is needed in specific areas.‍
• 3 - Proficient: Meets the job requirements and consistently performs to expected standards.‍
• 4 - Advanced: Often exceeds job requirements, demonstrating a high level of initiative.‍
• 5 - Expert: Consistently exceeds job requirements and is recognized as a leader or expert in the role.

Some organizations choose a custom scale because it makes it easy to draw a line between your company goals or terminology and your performance ratings. Employees may prefer a scale that they find meaningful. However, with custom scales, it may be harder to calibrate ratings across departments, and finding consensus on terminology and levels can be challenging. 

Quantitative Rating Scales vs. Qualitative Rating Scales

Performance rating scales fall into two main categories: quantitative rating scales, which use numbers, and qualitative rating scales, which use words. Both options have pros and cons, so let’s break them down a bit. 

What are quantitative rating scales?

A quantitative rating scale uses numerical rankings for employee performance. For instance, a manager might rate an employee’s job knowledge from 1 to 5. 

Numerical ratings are easy to understand and use, and they provide helpful standardized data. However, they can also be problematic. Lattice customer research from 2021 found that, no matter the number of options available, reviewers presented with numeric scales are less likely to rate individuals with the lowest option. For instance, less than 1% of reviewers gave someone a score of 1 out of 4, but 14% gave someone a “needs improvement” (essentially the same rating, the lowest of four levels.) 

The contrast is most pronounced among rubrics with four and six options, with less than 1% of respondents giving numeric “ones.” By comparison, the same segment was over 10 times more likely to give someone the lowest qualitative rating, like “needs improvement.”

What are qualitative rating scales?

Qualitative rating scales use words to describe the performance rating. These might be the same for every question, as in a standard five-point scale or Likert scale; for example, “poor,” “needs improvement,” “acceptable,” “good,” and “excellent.” 

Or they might vary by question, as in BARS, where the wording of the answer reflects the specific desired behavior required in the context of each performance dimension. 

In general, Raskin is a proponent of qualitative ratings. They “tend to do better in aligning with people’s performance, versus numerical ratings which can be so tricky. What’s a three? What’s a four?” He suggested that, in most cases, a qualitative rating scale, coupled with the option to add explanatory notes, will likely be the most effective for both managers and employees. 

Lattice users agree. In 2021, the share of review questions without numbers increased by 40% among Lattice customers — possibly because HR teams wanted to reduce the anxiety that number-based reviews could cause in an already pandemic-stressed workforce. 

Develop an effective performance rating scale with Lattice.

The right performance rating scale can significantly impact your organization’s ability to provide meaningful feedback. With performance management a top priority for many HR teams in 2024, it’s essential to approach this process mindfully, considering both the benefits and pitfalls of various rating systems.

To make sure you get the benefits of effective performance reviews and help mitigate the possible adverse effects, we recommend you read our detailed guide The Ultimate Guide to Mastering Performance Reviews.