The real impact of calibration curves
and why modern tech companies are dropping the archaic method to measuring employee performance
Many Engineers across the spectrum from startups to large corporations have likely experienced and endured some form of a performance review, sometimes linked to good things like bonuses, promotions, or raises — or on the flip side, performance improvement plans (PIP), and whether or not you keep your job.
One main ceremonial practice many companies have in common, is the begrudging multi-week journey of the calibration process. Some [companies] of which, in particular, incorporate a relative bell curve rating system. To learn more of what this process looks like, I highly encourage reading the two part series from pragmatic engineer Performance Calibrations at Tech Companies Part 1 and Part 2.
t;dr: it’s a game of inside ally’s, politics, and strategy, and favors business over people.
In this article, I’ve summarized core conclusions pulled from multiple resources on why the bell-curve system is ultimately a dying strategy in software engineering orgs.
According to a survey, 92% of employees are dissatisfied with traditional performance evaluations.
Managers spend an estimated 210 hours each year on performance management, while workers devote around 40 hours every year.
53% of employees say that they are not motivated with the performance measure standards
If we choose to fire today’s bottom 10%, we might be firing part of tomorrow’s top 10%. We just don’t know (treating employee metrics as purely random variables).
With the risk of stating the obvious, firing/punishing low-performers induces a fear-driven performance culture in an organization. Furthermore, it breeds a gamified environment where real creativity and productivity take the back seat while people who can game the metrics start winning.
Fitting people’s performance metrics to a normal distribution introduces a bias into the process such that everyone needs to belong to a preselected group. As managers fight over cut-off thresholds, top performers are often frustrated and eventually leave. What the system often ends up doing is to maximize mediocre performers and good system gamers.
...top performers or poor performers, people [generally] want and need to be treated as individuals. People generally don’t thrive in an environment where they are treated like cattle or (worse,) “resources”!
Following are some reasons why bell curves may not be the right approach for present times
Teamwork doesn’t count: Bell curve ranking system does not work with collaborative teams. A regular workday is not defined by a 9-to-5 regime; even more so since the pandemic.
Inaccurate and Unfair Assessments based on bell curve: When teams exceed performance expectations, it is not possible to measure performance accurately on the bell curve.
The use of the bell curve can be so off the mark, that more resources and money might get spent on recruiting and training new employees instead of working with members of the team individually to refine performance.
Forced Rankings – Bell Curves are very demoralizing for employees: The bell curve method runs the risk of diminishing the top performer’s value while inflating the value of middle performers. The bell curve provides a forced ranking of employees that distinguishes stellar performance from performance that is average or below par. This affects employee morale as the bell curve forces groups top and low performers regardless of their actual performance.
Research shows that this statistical model, while easy to understand, does not accurately reflect the way people perform. As a result, HR departments and business leaders inadvertently create agonizing problems with employee performance and happiness.
Research conducted in 2011 and 2012 by Ernest O’Boyle Jr. and Herman Aguinis (633,263 researchers, entertainers, politicians, and athletes in a total of 198 samples) found that performance in 94 percent of these groups did not follow a “normal” (bell curve) distribution. Rather these groups fall into what is called a "Power Law" distribution.
No one wants to be rated on a five point scale. First, much research shows that reducing a year or six months of work to a single number is degrading. It creates a defensive reaction and doesn't encourage people to improve. Ideally performance evaluation should be "continuous" and focus on "always being able to improve."
Ultra-high performers are incentivized to leave and collaboration may be limited. The bell curve model limits the quantity of people at the top and also reduces incentives to the highest rating. Given the arbitrary five-scale rating and the fact that most people are 2,3,4 rated, most of the money goes to the middle.
Mid level performers are not highly motivated to improve. In the bell curve there are a large number of people rated 2, 3, and 4. These people are either (A) frustrated high performers who want to improve, or (B) mid-level performers who are happy to stay where they are. In a sense the model rewards mediocrity.
Even employees that perform well may be placed in the middle ‘average’ group which isn’t exactly a motivating classification to be given and is particularly bad as they make up the majority of the team. If 70% of the employees are working the best they can and are still considered average by the company, then what is that going to do to their engagement levels? They are going to feel demoralized and lack productivity going forward, which will ultimately result in them leaving the company one way or another.
In some instances, bell-curve ranking can even be discriminatory. A lawsuit in 2017 against Uber by a former engineer, felt the company’s ranking system was discriminating against women and the lower rankings meant lower pay and fewer promotions.
When employees are performing to an equal standard, gender bias research has shown that unconscious bias will determine the final score, regardless of being told not to be biased.
A Harvard Law School study found that women are 1.4 times more likely to receive subjective feedback in their performance reviews that have nothing to do with how well they can do the job.
Over the past few years, some of the most admired companies such as Google, Adobe, and Microsoft have given up the bell-curve system — the system is 20+ years old, and most big organizations started waking up to this form of performance appraisal in the late '90s.
Microsoft dropping its age-old practice of using a relative rating system... The company has been criticized for holding on to a system that is believed to hamper creativity. Google too, dropped its complex matrix, got rid of the mandatory bell curve and went in for a simpler classification system, as of 2013. Adobe shed the curve and opted for a target-achievement based model it calls "check- ins".
"I don't believe the bell curve values people's strengths. It only gives a forced value," says Donna Morris, senior vice president, people & places, Adobe, who led the transition away from relative ratings at the firm. "One's merit should stand for what one actually does, and not against what someone else does." Also, in technology, everything depends on innovation, and you need the brightest of minds to come together, she adds. "It is quite destructive to tell your team that one day they need to come together and innovate, and the next day, they get ranked against each other," she says.
"The bell curve is way past its due date. It treats employees like machines, and their work like factory output," says Elango R, global chief human resources officer at Mphasis. "The only thing that should matter at the end of the year is whether or not an employee has achieved the set goals. Target achievement is the number one criterion in terms of performance," he adds.
Overall; The bell curve causes a great deal of unhappiness, and unnecessary stress both for managers and their team members.
Looking for company’s that are still hiring in tech? Check out Still Hiring where new roles are added daily.