Whether a business model is built on gigabytes, interest rates or the latest innovations in aluminum siding, every company ultimately depends on its people — some more than others. Businesses of any size have stars that drive productivity and get results, but look beyond those high achievers — the break room might be one place to check — and you’ll find others who drag the company down with shoddy performance.

The ultimate success or failure of a company often comes down to the quality of employees. As Jack Welch, former chairman of General Electric, once said, “the team with the best players wins.” But as CEOs and managers try to set up winning companies, they face a surprisingly difficult task: sorting the good employees from the bad ones. Baseball pitchers have earned run averages and quarterbacks have touchdowns, but the value of a given coder or salesperson can be much harder to define. Companies spend millions of dollars and burn countless hours conducting performance reviews and devising checklists to assess their employees, and business scholars have studied the issue with great urgency and intensity.

The results so far? By all available evidence, formal attempts to rate employees don’t seem to meaningfully improve employee performance or give companies any sort of competitive advantage, says Elaine Pulakos, a management expert and CEO of PDRI, a management consulting company based in Arlington, Virginia. “They end up being extremely costly and have no impact on productivity,” she says. Pulakos discussed the science of employee evaluation in a 2018 issue of the Annual Review of Organizational Psychology and Organizational Behavior.

Cartoon of a child being read to in bed. The child is saying: “I think the Little Engine was probably worried about his performance reviews.”

CREDIT: MARTHA GRADISHER / CARTOONSTOCK.COM

Despite many efforts, no one has been able to come up with a rating system that can reliably discern which companies are blessed with a deep bench of high performers and which brim with mediocrity. You certainly can’t tell simply by looking at the bottom line. Pulakos cites a 2012 report that gathered more than 23,000 employee ratings from 40 companies and found no sign that ratings had any effect on profits or losses. “Performance ratings have no relation to organizational performance whatsoever,” she says.

Out of all of the methods used to rate and grade employees, the dreaded annual or semiannual performance reviews are especially unhelpful and potentially harmful, Pulakos says. “They’re really toxic and people hate them,” she says. “You’re creating artificial steps just to check a box.” Pulakos points to brain imaging research positing that even high-performing employees automatically go into a defense mode during performance reviews, turning a supposedly productive meeting into a fight-or-flight scenario.

Formal annual performance reviews can be extremely damaging to a company’s culture, says Herman Aguinis, the Avram Tucker Distinguished Scholar and professor of management at George Washington University in Washington, DC. “It’s a soul-crushing enterprise,” he says. “The employee doesn’t know what they’re supposed to do, and the manager doesn’t see any value in it. They’re only doing it because human resources told them to.”

All too often, Aguinis says, formal performance reviews become a self-serving exercise in politics, not a realistic examination of an employee’s strengths and weaknesses. “Some managers will give biased ratings on purpose,” he says. “I have personally seen a supervisor giving a bad employee a good rating just so that employee could get promoted out of his unit.”

Cartoon of a man sitting in an office getting a performance review. His manager is saying: “Let’s just go through your two areas of strength and your three hundred and twenty two areas for development.” The caption reads: “Gary began to think that his performance management meeting could have gone better.”

CREDIT: MIKE SEDDON / CARTOONSTOCK.COM

Still, some HR experts continue to see some value in annual performance reviews. In a February post on her popular Evil HR Lady blog, Suzanne Lucas says “annual performance reviews aren’t all bad. Formal ratings provide a macro-view of performance and engagement levels across the company. If the results of any group (department, experience level, etc.) stick out — it can indicate a bright spot or potential problem worth looking into.”

The growing body of research questioning the value of performance reviews has encouraged many companies to rethink their approach. Dell, Microsoft, IBM and other big business names such as the Gap, Accenture and General Electric have ditched the process, a move at times fanfared in press releases and headlines. But a 2018 survey by the research firm WorldatWork found that 80 percent of companies still used formal performance reviews. “Behavior change in organizations is really hard,” Pulakos says.

Businesses that abandon formal performance reviews still have to keep tabs on employees, Aguinis says: “Companies that say they are getting rid of ratings are still using ratings. They just have different labels.” For one thing, managers must have some rationale for assigning promotions and raises. If there’s no data on performance, the process of handing out promotions and raises can turn chaotic. In some cases, companies could be vulnerable to lawsuits if they don’t have a way to justify decisions.

To really understand the value of their employees, Aguinis says, managers should double down on the practice of everyday management. That means checking in on employees every day and giving them real-time feedback on things they’re doing well and areas where they can improve. “When performance is a conversation, when it’s not something that happens just once a year, the measurement becomes very easy and straightforward with no surprises,” he says. He adds that it’s important to gather input from many different people within the system — peers as well as supervisors. “The best source of data is often not the manager,” he says.

When rating employees, it’s best to keep things simple, says Seymour Adler, a talent and rewards partner at Aon, a management and HR consulting firm headquartered in London. He ruefully remembers a mistake early in his career, when he was part of a team that came up with a 40-point scale to rate employees. “That’s an over-engineered solution in my view,” he says.

A cartoon in which an ant is getting its performance review in a subterranean nest with plant roots poking through the ceiling and a tunnel leading off somewhere. The ant conducting the appraisal sits at a desk under a sign that reads “Ant H.R.,” and is saying to the subordinate: “Your coworkers complain that you don’t carry your share of crumbs.”

CREDIT: TERRY LABAN. THIS CARTOON WAS ORIGINALLY PUBLISHED IN REWORK

Rating employees solely on objective measures such as sales numbers, absentee days, or customer calls may seem like a winning strategy, but those data points can be wildly misleading, Adler says. A salesperson with the most sales may have a better territory or better luck than others, not more talent or drive. “Objective measures may seem straightforward, but you have to think about all the factors that are beyond an employee’s control,” he says.

Daily evaluation and feedback may sound like an onerous task, but Adler says there’s an important loophole: Most employees do just fine without constant scrutiny. “When I work with companies, I encourage them to get away from ratings and start managing by exception,” he says, meaning that the exceptional employees need the most attention. Out of 100 employees, there might be three or four who are struggling so mightily that they need an intervention or a career change. At the other end, there might be five or so excellent employees who should get special treatment because they drive the company’s success. A 2012 study by Aguinis and coauthor Ernest O’Boyle Jr. found that the top 1 percent of workers account for 10 percent of a company’s productivity. The hardworking, competent but unexceptional workers in between the extremes — Adler calls them “the Mighty Middle” — are going to make about the same contribution to a company’s bottom line regardless of how much time they spend in performance reviews.

Some companies have taken appreciation of superstar employees to extremes. In his 2015 book Work Rules!, former Google executive Laszlo Bock reveals that the company routinely pays high-performing employees five or six times as much as other employees at the same level, maybe even more. He also cites such instances as one worker’s receiving a $1 million stock bonus while another received just $10,000.

Of course, Google is an industry outlier in many ways. Pulakos notes that the company lives on data, and it has methods for rating and ranking employees that just wouldn’t work anywhere else. That’s one of the big lessons of modern business scholarship: Every company has to figure out its own approach to getting the most out of its employees. “You have to evaluate your own strategic goals,” she says. “What works for Google is not going to work the same way for anyone that is not Google.”

In the world of business, there aren’t many universal truths. Just one, really: Annual performance reviews are the worst.