Fair Evaluation Isn't Something Numbers Can Create

This is not a uniquely Korean phenomenon. Harvard professor Michael Sandel, who has had a significant influence on global sociology,

takes a critical look at the meaning of "fairness" in his recent book The Tyranny of Merit. 

Among younger people in particular, the notion of fairness shows up as an active embrace of meritocracy — which means the same operating principle as "performance-ism," where greater wealth flows to those who demonstrate higher performance and capability. 

In the corporate world in particular, performance-ism is taken as always right, by executives and employees alike. We take it for granted that top talent who earn high evaluation results through a fair process should receive high rewards, and we hold the belief that receiving differentiated treatment based on individual ability is a fundamental principle of the system. Across society, a paradox of fairness has taken hold in which "differentiation" based on ability is treated as "fairness" itself — and in the workplace, this principle of differentiated rewards is taken even more for granted. So even though "high evaluation = high reward" is accepted as common sense, ironically, evaluation and rewards — as an HR domain — are ultimately pointed at different goals.

The Different Goals of Evaluation and Reward

In a highly competitive society steeped in meritocracy, we tend to equate evaluation with lining people up in a queue. We accept this queue as a natural result, and when problems surface in the process, we tend to attribute them not to the queue itself but to the criteria used for queuing, or to the judges making the call. As a result, at companies that have neglected to evolve their HR systems, it is not unusual to find factors unrelated to actual work performance — like language test scores or internal training course grades — mixed into evaluation.

"How do we produce the clearest possible result?" becomes a more important question than "What should we be evaluating?", and sometimes this clarity of result gets confused with fairness itself.

Overseas, techniques like "Stack Ranking" and "Rank & Yank" — queuing people, assigning grades, rewarding the top differentially, and marking the bottom for exit — spread widely from the 1980s onward. The godfather of this approach, former GE chairman Jack Welch,

ran the "10% Rule," which laid off the bottom 10% of employees every year, preceded by long, intense discussions about talent (Session C). Many companies around the world borrowed this technique, classifying talent by grade and differentiating their treatment accordingly. But many of them overlooked what actually mattered: the criteria for evaluation.

As a result, even the U.S. IT giant Microsoft, well into the 2000s, was still evaluating talent based on quantitative results and widening the gap between high- and low-rated employees — to the point where creative, high-potential employees were being laid off in droves, only for these "marginal" people to be picked up by the rising IT stars of the time — Google, Apple, Amazon — where they went on to create groundbreaking services, in full view of everyone.

Microsoft, having learned a hard lesson, later abandoned all its quantitative and complex evaluation metrics (performance, deadline adherence, product types, customer ratings, and so on) and chose instead to evaluate a single abstract-but-comprehensive essence — Impact — that captures both organizational performance and individual capability. People who delivered meaningful impact to the company and to their collaborators through performance creation, new ideas, or the sharing of methodology received high evaluations, and among them, those whose level of impact

represented "competitive value" were given best-in-market compensation through a "One Microsoft" HR system. This is a key part of Microsoft's successful comeback.

The takeaway from Microsoft's case is that in the end, "what we evaluate" is what matters. The ultimate goal of evaluation should be to evaluate in a direction where the company grows and the individual develops, aligning each person's work, results, and capability with the organization's ultimate development. In that process, evaluation should motivate employees and give their behavior an indicative legitimacy. If, until now, "fairness" in the effort to persuade employees and earn their acceptance has been focused on "presenting objective results that no one can dispute," going forward it must shift toward recognition and belief that an employee's behavior can drive the shared development of both the company and the individual.

Indicators of this kind are typically abstract and qualitative, and often emerge as a composite of many people's "impressions." That means today's queue-based system is inevitably going to change, and companies that are serious about rethinking evaluation are now studying a move to absolute evaluation.

Rewards, on the other hand, carry a goal that can look somewhat different from the individual's and the company's points of view. For individual employees, feeling that high evaluation is tightly linked to high reward produces a greater sense of fairness. But from the company's side, even an employee who gets consistently high evaluations year after year does not necessarily create pressure to keep raising their rewards, as long as comparable capability and skill can easily be found in the market. Conversely, an employee who is too junior or too unskilled to earn a high rating is an employee the company will easily lose if other companies are all chasing that skill profile — applying low rewards in line with a low evaluation will simply hand that person to a competitor. In short, while employees may believe evaluation and reward are linked in a direct lever, at the company level, reward is actually driven more by external market competitiveness than by internal evaluation factors.

The belief that tightening the link between evaluation and reward will strengthen motivation can actually generate side effects. To preserve employees' "belief in fairness" — that the reward they receive as an outcome is the same as the evaluation they received as a cause — companies try to link rewards back to evaluation even when the real reason to adjust rewards lies in external market competitiveness. The result is a cascade of evaluation errors and errors in metric calculation, and transparent communication becomes even harder. More often, you see complex structures in which allowances and add-ons are layered on to compensate a few employees whose market scarcity and hiring difficulty are high — independent of evaluation.

To Turn Meritocracy Into Real Performance-ism

If, until now, we have tried to raise fairness in evaluation through quantified results and simple, clear-cut criteria, going forward, a holistic evaluation of qualitative and abstract competitive value will have to earn trust through the synthesis of "multifaceted evaluation." 

If we can bring together observations from many people, each looking from their own angle, and accumulate those opinions as data that serves as evidence, then this "collective perspective" will ultimately act as the mechanism that raises credibility. For the company, the accumulated opinions about each person also become input for Talent Management, which means we can move away from the massive, long-running deliberations that GE used to run and use more objective, richer observational content in much lighter discussions.

To further raise the sense of fairness in this process, timely and immediate feedback has to become more active. Feedback gathered from a variety of observers in real time creates the foundation for employees to positively accept the final evaluation result. For the company, delivering that feedback also allows for more flexible goal management and more agile capability improvement. To make these touchpoints real, the company needs to be prepared in several ways. 

First, on the system side, the HR structure needs to be built around "job" and "role." To evaluate employees by their true value,

and to let multiple managers and collaborators participate in that evaluation, the job each person owns must be clearly defined, and the roles they are expected to play and are given to play must be clear as well. Only then can colleagues judge whether the "impact" someone has is commensurate with the level of the job and role.

Going further, the rewards that apply to employees can also be built on a system that follows job and role. As discussed earlier, the first determinant of the rewards applied to any given employee is market competitiveness. That market is becoming more segmented by job, and within each job, more differentiated by the level of role someone can take on. If the entire HR system is structured around job and role, it becomes much easier to apply a system that can provide efficient competitiveness in line with market levels. Instead of the headache of determining rewards based on tenure, accumulated job level, and the distribution of evaluation ratings — while simultaneously trying to balance cost efficiency, retention, and the story told to employees — the process can be moved toward something much more market-friendly.

Second, digital transformation is ultimately a prerequisite. Synthesizing, accumulating, and organizing the diverse perspectives of colleagues as data — and using them in discussions — as well as managing work in real time, collecting feedback, and organizing and delivering it: all of this requires digital tools to be viable.

The continuous performance management tools being developed recently all focus on data-ification, analytics, and real-time functionality, and the trend is to deliver them not as complex internal systems for HR operations but as Software-as-a-Service (SaaS) offerings designed to secure employee participation.

Let Fairness Be the Soil of Mutual Flourishing

In the end, The Tyranny of Merit concludes that young people today are so steeped in meritocracy that they hold an illusion: "in a fair social system, if I just demonstrate enough ability, I can gain wealth." Ultimately, though, the social systems we have — exams, academic and career evaluations, reputation and fame — are structured to favor those with advantaged birth backgrounds, and they end up deepening inequality.

Even though the real-world system delivers this tragic message, fortunately, companies are much smaller organizations with a more unified purpose than society as a whole. If we can run evaluation and rewards according to their "true meaning," there is hope that both the organization and the individual can enjoy the effects of motivation and performance improvement.

The key, of course, lies in a flexible, job-market-friendly system and the digital innovation that supports it.