Few workplace topics inspire as many differing opinions and controversy as the subject of performance reviews. Although they are considered standard or even obligatory in many medium and large organizations, there has been markedly high levels of frustration with them, both from employees and from the managers who conduct reviews. This universal disdain would, you might imagine, inspire change, but they remain a constant feature of even the most cutting-edge workplaces. There are legal and organizational considerations that contribute to this, as, in theory, performance reviews allow organizations to reward the most performant employees, address shortcomings, and provide evidence for organizational growth or reduction.
Some critics of performance reviews have been sharing their hopes that artificial intelligence (AI), or its more accurate title of large language models (LLM) can help improve reviews. As a seemingly impartial piece of technology, AI tools can help organizations track performance in ways that reduce or remove bias, providing a more accurate portrait of each employee’s track record. This hope is surprising, as many employees have also registered exhaustion with the widespread adoption of AI in the workplace, but this may speak more to the lack of trust most have for current performance review practices than to trust in AI.
Here are the ways AI might help you improve performance reviews, and some issues it can’t solve:
Why AI?
The sense that AI can help organizations improve their review processes is based on a few assumptions that speak to the core benefits of AI tools. AI is primarily touted as a time-saver, automating repetitive or labor-intensive tasks, and this could be made to include tracking the performance of employees and teams. Additionally, it theoretically could remove many of the common biases that managers often unconsciously lean on when making assessments.
Removing bias
Perhaps the most compelling evidence against performance reviews is the widespread evidence of bias, particularly along gender, racial, and other lines. This issue is so pervasive and destructive to the value and trust in the process that it has led many HR leaders to call for an end to the practice altogether. AI can address this by simply removing it as a consideration and documenting exactly how each assessment is made. By providing a transparent record of its assessment criteria, it can give employees confidence that they are only being assessed as employees, and other factors are automatically ignored.
Taking everything into account
Another common criticism of performance reviews is recency bias, or the favoring of employees who have performed well recently rather than performed well consistently. For example, say you had two employees, one who struggled in the past but completed some major milestones in the last few months and one who performed well over the entire period, you might actually favor the one who showed improvement over the one who was consistent. An AI system which takes performance over a set period of time into account should not have this issue, as it would weigh each month equally.
What AI can’t fix
Despite offering some significant benefits, AI tools are not inherently perfect. A common misconception about technology is that it is inherently unbiased, but anything that runs on human input, including AI, will replicate human flaws and biases.
Tracking performance
A key factor which limits the effectiveness of performance reviews is that “performance” can be highly relative. For roles like sales or development, where performance can be tracked numerically, performance can have a clear meaning. However, roles which involve interfacing with the public, leading teams, or training others can be harder to assess. Another factor is the range of responsibilities. For example, if a member of a three person team has to quit for personal reasons, and the other two members take on their responsibilities, how should that assessment change? Theoretically, an AI tool could be tweaked to account for this, but it still does not deal with the fact that “performance” itself can be difficult to distill.
Restoring trust in performance reviews
The widespread mistrust of this practice is not just a matter of process or technology. It isn’t hard to see that, even if you were to incorporate AI, employees still might have issues with the guidance it gives them, the way it functions, or the way your managers use it. You might be replacing one problem with another — instead of having to continually tweak the performance review process, you have to continually tweak the AI. This is further complicated by the fact that AI has its own issues — the possibility of hallucinations, overly-generalized feedback, and other issues would give employees and managers plenty of reasons to be highly skeptical of AI output.
Should you use AI?
While incorporating AI into performance reviews could offer some significant benefits, it also brings plenty of risks. Before adopting these tools, it can be helpful to reconsider the way you use performance reviews and whether the whole process makes sense. You may find that these kinds of reviews do not fit every role, or that your managers have too much on their plate to do an adequate assessment of all their reports. You could also gather data on how your organization feels about the reviews, offering the chance for anonymous feedback via an internal survey. If there is a generalized trust of the assessment process, AI can help you speed up the more labor-intensive parts while offering greater transparency. However, this doesn’t mean an AI-powered assessment process will be perfect, and you should be ready to make adjustments.
Get an expert’s perspective on your organization’s needs
The 501(c) Services team works to keep up with the latest trends and technological developments in the nonprofit space, and we can help you strategize and utilize these tools to the fullest. By identifying crucial resources you might not have been aware of, we can assist you in furthering your organizational goals and building sustainable processes for team growth and success. If you’d like to learn more about how we work, please get in touch with us.
ABOUT US
For more than 40 years, 501(c) Services has been a leader in offering solutions for unemployment costs, claims management, and HR support to nonprofit organizations. Two of our most popular programs are the 501(c) Agencies Trust and 501(c) HR Services. We understand the importance of compliance and accuracy and are committed to providing our clients with customized plans that fit their needs.
Contact us today to see if your organization could benefit from our services.
Are you already working with us and need assistance with an HR or unemployment issue? Contact us here.
The information contained in this article is not a substitute for legal advice or counsel and has been pulled from multiple sources.
(Image Credit: Canva.com, Some of these images are AI Generated)
