The question companies encounter most quickly after introducing OKRs is evaluation and compensation. “Should OKR achievement rates be reflected in HR evaluations?” “If they are not connected to compensation, won’t employees fail to take them seriously?” “Conversely, if they are connected to compensation, won’t stretch goals disappear?” These questions are hard to avoid in OKR operations.
However, it is risky to rush to a single answer. If OKRs are completely separated from evaluation, execution accountability may weaken. Conversely, if OKR achievement rates are inserted into the compensation formula, employees will choose safe goals. This article is not legal or labor advice; it addresses the boundary line for linking evaluation and compensation from the perspective of HR operating design. The issue is not whether to reflect OKRs, but which OKRs should be interpreted through what evidence.
The moment OKRs enter the compensation formula, the nature of the goal changes
What Matters compares OKRs with MBO and explains that OKRs operate on a quarterly cadence and spread with a philosophy separated from compensation, unlike annual management by objectives. What matters here is not the simple rule that “OKRs must never be connected to compensation.” It is that the purpose of OKRs is not to produce evaluation scores, but to align priorities and execute strategy.
When a goal enters the compensation formula, employee behavior changes. The incentive to choose safe goals grows stronger than the incentive to choose stretch goals with a lower probability of achievement. Collaborative goals can turn into disputes over the allocation of individual responsibility. Even if strategy changes during the quarter, goals become difficult to revise, and employees try to lower initial goals to avoid a record of underachievement.
This change is a natural response. In a system where compensation is at stake, people behave according to the system. Therefore, whether to put OKRs into the compensation formula is not simply a matter of HR preference. It is a design choice connected to what the organization wants to encourage among challenge, learning, collaboration, and strategic adjustment.
Distortion occurs when committed OKRs and aspirational OKRs are handled on the same scorecard
Google’s OKR playbook distinguishes committed OKRs from aspirational OKRs. It explains that committed OKRs are close to goals whose achievement has been promised, and that the expected score is 1.0. A score below 1.0 requires explanation. By contrast, aspirational OKRs are closer to goals that stimulate innovation and challenge. They have the nature of attempting greater change before the path is fully confirmed.
If this distinction is ignored in evaluation, distortion occurs. Underachievement of a committed OKR can be a signal to review execution accountability, resource allocation, and failure to adjust priorities. But if the low achievement rate of an aspirational OKR is penalized in the same way, stretch goals disappear. Employees choose only goals favorable to evaluation, and OKRs become a list of safe promises rather than a language for innovation.
The same applies to cross-team OKRs. If goals for which several departments share responsibility are simply allocated as individual achievement rates, defensive responsibility grows stronger than collaboration. Shared goals should be reviewed together with who took which part, what dependencies existed, and what leaders adjusted. The moment all OKRs are handled with one identical scorecard, the strengths of OKRs disappear and only evaluation administration remains.
What can be used in evaluation is interpretable evidence, not achievement rates
It is also difficult to completely ignore OKR results in evaluation. If an employee held and executed an important goal throughout the quarter, the process and the result become important material for the performance conversation. However, what can be used in evaluation is interpretable evidence rather than a simple achievement rate.
The Google playbook says that Key Results should describe outcomes, not activities. This principle is also important in evaluation conversations. “Conducted training three times” is activity evidence. “Within 60 days of assigning new leaders, the rate of 1:1 feedback to team members increased from 40% to 85%” is outcome evidence. What can be referenced in evaluation is closer to the latter.
The explanation that problems with committed OKRs should be escalated quickly is also worth noting. More important than underachievement itself is what judgment and adjustment occurred when signals of underachievement appeared. When a schedule was delayed, it is necessary to check whether the leader adjusted resources, rearranged priorities, and resolved goal conflicts. A performance conversation must read the accountability structure behind the number rather than the achievement-rate number itself.
Korean companies need to design a third way between separation and reflection
In Korean companies, sensitivity around evaluation and compensation is high. The stronger an organization’s experience of connecting goal achievement rates with evaluation grades, the more quickly OKRs are also read as evaluation forms. In such organizations, even if HR declares that “OKRs are unrelated to evaluation,” employees do not easily believe it. Conversely, if HR says that “OKR achievement rates will be reflected in evaluation,” stretch goals quickly decrease.
Therefore, the realistic choice is a third way between complete separation and direct reflection. It is a method of using OKR achievement rates as reference material for performance conversations rather than directly inserting them into compensation formulas. Even then, interpretation rules by goal type are needed. Committed OKRs look at fulfillment of promises and execution accountability. Aspirational OKRs look at attempts, learning, and market or organizational signals. Cross-team OKRs look at collaboration structure and coordination accountability.
HR should leave these principles in writing. It must decide which OKRs are subject to evaluation reference, which OKRs are viewed only as learning records, what evidence will matter more than achievement rates, and whether there is or is not any disadvantage when goals are revised midway. If OKRs are operated in an ambiguous state, employees behave in the most conservative way.
Operating examples can be divided into three. First, committed OKRs are viewed as promised execution results, and reasons for underachievement and coordination accountability are checked in the performance conversation. Second, aspirational OKRs are evaluated by the hypotheses attempted, the market, customer, and organizational signals learned, and the selection criteria for the next quarter rather than by achievement rates. Third, cross-team OKRs are reviewed by looking together at each department’s promised contribution and leaders’ coordination behavior rather than splitting them into individual scores. OKRs must be distinguished this way so they are not flattened into a compensation formula.
What HR must define is not the compensation formula, but the boundary of the performance conversation
When designing the relationship between OKRs and evaluation and compensation, the first thing HR must define is not the formula. More important is the boundary of the performance conversation. HR must distinguish what will be asked in OKR reviews, what will be interpreted in evaluation interviews, and which materials alone will be used in compensation decisions.
First, OKR reviews should address progress and the need for adjustment. Second, performance interviews should look not only at OKR results but also at role expectations, collaboration, capability, and contribution to the organization. Third, compensation decisions should be explained separately within the organization’s compensation philosophy and criteria for role, grade, market value, and performance contribution. Without these boundaries, OKRs become an overloaded system that must explain everything.
Whether OKRs should be reflected in evaluation is not a matter that can end with a simple yes or no. Even if they are reflected, achievement rates should not be converted directly into scores; even if they are separated, execution accountability should not disappear. The core is to design OKRs so that they become evidence for performance conversations without killing challenge and learning. If this balance cannot be created, OKRs shrink into a supplementary item in the evaluation system. If the balance can be created, OKRs can become a performance-management language that works throughout the quarter, not only during evaluation season.
