Today’s accounting information systems have greatly reduced the amount of time managers need to gather employee data in order to analyze performance. In fact, many information systems are sophisticated enough to provide performance feedback to workers with reduced or even no management intervention.

We see this in several forms, including workplace gamification, in which performance rankings are updated in real time, and in performance management systems that display dashboards tracking completion of goals. Some companies take a different approach and offer more frequent feedback via a software system rather than through face-to-face communications. In either case, the adoption of automated feedback methods can reduce or eliminate direct manager-employee interactions.

These systems save managers vast amounts of time and energy typically used to compile the information needed for feedback. In addition, companies can gain efficiency because frequent feedback is timely and allows for nimbler operations. After all, what employee wants to wait until the annual performance review to find out how things are going?

Is more automation always better? Certainly not! After all, ATMs were in use as early as the 1960s, but they haven’t completely replaced bank tellers. There are obvious communication benefits to hearing someone’s voice, looking into their eyes, and reading their body language. Nevertheless, we believe that one potential advantage of automated feedback compared to traditional feedback systems is that the use of automated feedback provides companies with the opportunity to more fully control the feedback message—specifically, whether or not managers use emotional language when providing performance feedback to their employees.


Emotional language is wording that expresses an emotional reaction of the person providing feedback. It can occur intentionally or unintentionally, in writing or in person. For instance, managers may deliberately use emotional language in order to motivate those whom they lead; managers of sports teams do this all the time. Yet managers may also unintentionally use emotional language that can be detrimental to team members’ performance, such as in high-pressure, long-hour situations where deadlines are tight and nerves are on edge. In such scenarios, even seemingly innocent comments meant in jest (“Hey, slacker, speed it up over there!”) may be taken the wrong way by an overworked employee.

Thankfully, technology affords managers the opportunity to avoid emotional powder kegs. Performance feedback delivered to employees solely by a company’s accounting information system can be structured in such a way that unintentional emotional language is removed. For example, performance data can be provided directly by the information system using algorithms and predefined fields. This could then be accompanied by a system feature that allows optional managerial comments on the performance, if desired. This modern-age feature of workplace performance management systems begs the question: To what extent should managers avoid emotional language, positive or negative, when giving feedback? Can it inspire improved performance, or will it be detrimental?

To help understand this issue, we conducted a series of experiments to examine the effect of managers’ use of emotional language when providing performance feedback in written/electronic form, with the aim of better understanding how variations in noninteractive, system-produced performance feedback influence future employee performance. We hope that our research can aid companies in deciding how, and in what form, automated feedback should be adopted.


Why does a manager’s use of emotional language matter in performance feedback situations? Simply put, emotional language is a signal that influences how individuals process and react to feedback. Think of it this way: When employees receive feedback, they may consider (1) the implications of the feedback in relation to the amount or type of effort they need to exert or (2) the implication the feedback makes about them as a person—their ability and intelligence, for example.

Research refers to these two potential interpretations of feedback as attention to the task and attention to the self. The extent to which employees think about the task or themselves depends in part on signals or cues in the feedback message. Emotional language is one such cue that may cause employees to think more about themselves when receiving feedback (see Figure 1).

When employees are more focused on the implications of the feedback for the work they perform, they’re more likely to think about their motivation and effort. Thus, employees who receive negative performance feedback may infer that they need to work harder, whereas employees who receive positive performance feedback may infer that they’re already working hard enough—or even that they can reduce their effort.

In contrast, when employees are more focused on what the feedback implies about them as a person, they tend to be more concerned with tangential issues—such as self-esteem, control, and impression management—that distract from the primary purpose of the feedback, which is performance improvement. For example, when a manager uses emotional language while giving feedback (“I’m really disappointed in your work this quarter”), employees are more likely to experience their own emotional reactions to the feedback, such as dejection or anger. To deal with these feelings, employees may focus on whether the manager’s reaction is fair or whether they even care about what the manager thinks about their work. In other words, managers’ emotional language can muddy the feedback waters and prevent employees from focusing on what matters most (“What should I do to improve?”).


To help shed light on whether and when managers should incorporate or seek to reduce emotional language when providing feedback, we examined two important feedback settings. In the first setting, the performance feedback was unexpected—participants didn’t know they would be receiving performance feedback. Typical of most work settings, managers often provide unscheduled feedback to employees to help improve performance. For example, a manager may notice that an accounting system implementation is going well (or terribly) or that certain investments in infrastructure are performing well (or poorly), and take a minute or two to compliment or express concern to employees.

In the second feedback setting, the study participants expected the timing and format of the feedback. This setting more closely approximates the formal feedback processes that companies utilize, such as during annual performance reviews or project review deadlines.

We recruited several hundred business majors to participate in the first two experiments. We asked them to take the role of a worker and paid them a fixed wage ($15) to do a simple task that mirrored the testing of circuit boards. The students first completed a five-minute practice round to familiarize themselves with the task. Then, they began Round 1 production, which lasted 10 minutes. Afterward, we provided them with a paper-based feedback report that contained a summary of their performance and indicated how it compared to that of another anonymous participant. Half of the students learned that they performed better than someone else, and the other half learned that they performed worse.

To test how emotional language from the manager influences future performance, we also varied the language in the performance feedback reports. Specifically, half of the students who learned that they performed better than someone else were also told that the manager was “pleased” with their performance, whereas the other half were not. Likewise, half of those who learned they performed worse than someone else were also told that the manager was “disappointed” in their performance, and the other half were not.

After reviewing the performance feedback report, students completed Round 2 for an additional 10 minutes. We then looked at the change in participants’ production from Round 1 to Round 2 to see how the type of feedback (positive vs. negative) and the emotional language (if present) impacted the students’ performance.

We also wanted to know whether the characteristics of the feedback recipients themselves would influence how they responded to the manager’s emotional language. We chose to focus on the personality trait of psychological entitlement, which likely sensitizes employees to expressions of emotion. Psychological entitlement can be understood as a “stable and pervasive sense that one deserves more and is entitled to more than others,” as described by noted social psychologist W. Keith Campbell and his colleagues. (For more, see W. Keith Campbell, Angelica M. Bonacci, Jeremy Shelton, Julie J. Exline, and Brad J. Bushman, “Psychological Entitlement: Interpersonal Consequences and Validation of a Self-Report Measure,” Journal of Personality Assessment, August 2004.)

In general, entitled individuals expect high levels of rewards and preferential treatment. They’re highly sensitive to the opinions of others, desiring approval and praise. Our expectations were that individuals high in psychological entitlement would be relatively more attuned to managers’ emotional language. Additionally, psychologically entitled individuals tend to avoid detailed and effortful thinking, making their response to feedback more emotional than analytical. To incorporate this characteristic into our experiment, we measured participants’ psychological entitlement using the Psychological Entitlement Scale developed and described by Campbell, et al. The statements used to measure psychological entitlement are listed in “How Psychologically Entitled Are You?”

Finally, using the same task and design as the studies that included business students, we ran an online experiment recruiting workers through Amazon Mechanical Turk, an online crowdsourcing marketplace, to test whether managers can counteract employees’ reactions to emotional language unintentionally communicated during feedback. Specifically, we added an additional statement to the feedback message stating that the feedback only reflected the employee’s performance on the task and didn’t reflect on her or him as a person.


What did our experiments teach us? We found that participants who received negative feedback that also included the emotional language of the manager’s disappointment performed worse than participants who simply received negative feedback without such language. In contrast, for those participants who received positive feedback, the inclusion of emotional language that indicated satisfaction with the participant’s performance led to no difference in future performance compared to those who simply received positive feedback (see Figure 2).

We also found an interesting pattern of results based on participants’ level of psychological entitlement. Of those participants who received unexpected negative feedback, we found that the manager’s emotional language that expressed disappointment caused even worse performance for more-entitled individuals. On the other hand, more-entitled participants significantly improved their performance when their positive feedback reports included the manager’s emotional language of being pleased. Thus, based on the feedback recipient’s level of entitlement, the inclusion of the manager’s emotional language led to an amplification, good or bad, of responses to performance feedback.

Interestingly, we didn’t find the same pattern of results for psychologically entitled individuals when feedback is expected. It may be that clear communication of how and when feedback is delivered allows employees to mentally prepare themselves to receive feedback and not to take it personally.

Lastly, there’s some hope for managers who unintentionally use emotional language when providing feedback. We found that a statement from a manager depersonalizing the feedback (stating that it applies to employees’ work on the task and doesn’t reflect on them as a person) counteracts the adverse effects of negative emotional language.


Our findings have several practical implications. First and foremost, the way managers provide feedback is perhaps as important as the feedback itself. Our results suggest that emotional language acts as a signal that diverts employees’ attention away from the implications of the feedback for their work and toward the implications of the feedback for them as a person. As a result, we find that negative feedback is more effective when delivered without emotional language from the manager.

Our results suggest that decoupling the manager’s emotional language from the delivery of negative feedback makes for better employee performance down the road. These insights are useful for managers because negative feedback is often an effective mechanism to help drive improvement, and coupling emotions with the feedback is a common and natural human experience.

Consider the following example: Soraya is a newly promoted controller for a small private business and, for the first time, faces the challenge of providing feedback to Chad—an underperforming staff accountant. Chad is bright, but he seems to lack motivation. How can Soraya deliver the feedback to best motivate Chad to improve?

If you’re a manager like Soraya, faced with an employee who is underperforming, you may be inclined to tell Chad how upset you are with his performance, thinking that this approach may inspire Chad to improve. Our results, however, suggest this tactic has the opposite effect. You’re likely better off delivering the feedback in a matter-of-fact way, showing Chad how his performance compares with expectations while avoiding any emotional reaction. Doing so communicates the need for improvement without demotivating employees or causing them to seek retaliation. Overall, restraint and discipline are key components when providing negative feedback.

On the other hand, it makes perfect sense to include positive emotional language for employees who are more likely to respond emotionally to positive feedback. As noted, our experiments suggest that this is particularly important for unscheduled instances of feedback. Thus, in addition to communicating performance relative to expectations, figurative “pats on the back” for good performance can reinforce employee behavior and spur additional improvements in the future. For some examples of how certain feedback—positive and negative—might be worded, see “Keys for Emotional Feedback.”

Our survey results are particularly relevant given the increasing prevalence of feedback delivery via company information systems. To the extent that these new feedback methods reduce or eliminate direct manager-employee interactions, they also give companies (and managers) the opportunity to more fully control the flow of emotional messages when providing feedback.

As you can see, it isn’t always easy to give constructive feedback that motivates employees without it having a potentially negative emotional effect on them. Whether you’re a manager who provides performance feedback—or an employee who’s looking to gain valuable insight from it—we hope that our findings can help you more effectively give and receive feedback in order to reach a higher rung on the corporate ladder.

About the Authors