MBO, verified for over half a century, and OKR, chosen by Silicon Valley. We analyze the essential differences between the two frameworks and strategies for adoption by organization.

Two Branches from the Same Root

Goal management is one of the oldest topics in modern management. What should we work toward, and how do we know if we are heading there properly? Management by Objectives (MBO) and Objectives and Key Results (OKR) emerged as systematic responses to these two questions. Interestingly, the two frameworks are not products of completely different camps; in fact, they share a single lineage.

MBO is a concept first presented by Peter Drucker, often called the 'father of modern management,' in his 1954 book *The Practice of Management*. Pointing out the phenomenon where managers get bogged down in daily tasks and miss the organization's broader direction and long-term strategy, Drucker proposed a management philosophy that synchronizes individual goals with organizational objectives. He defined MBO not as a specific technique, but as a 'management philosophy,' viewing goals as determined by situations rather than formulas.

OKR is a form that evolved a step further based on MBO. Andy Grove, CEO of Intel, combined Drucker's MBO with a measurement device called 'Key Results' in the early 1970s. Grove initially called this method 'iMBO' (Intel Management by Objectives), a name that acknowledged it as an Intel-style variation of MBO. He defined OKR as an answer to two simple questions: Where do you want to go (Objectives), and how will you know if you are getting there (Key Results)?

John Doerr is the figure who learned Grove's method at Intel. Working as an Intel sales representative in 1975, he received OKR training from Grove before moving to the venture capital firm Kleiner Perkins. When Kleiner Perkins invested in Google in 1999, Doerr introduced OKR to an early-stage Google with around 40 employees. Larry Page and Sergey Brin adopted it company-wide, and alongside Google's success, OKR spread beyond Silicon Valley to the rest of the world. In short, MBO is like the grandfather of OKR, and OKR is a descendant redesigned to fit the reality of fast-growing tech companies.

Difference in Philosophy: A Tool for Control or a Tool for Alignment?

The most essential difference between the two frameworks lies in 'what goals are used for.' MBO has essentially evolved as a tool to measure and evaluate performance. Superiors and subordinates agree on goals together, define their respective areas of responsibility, and after a certain period, evaluate performance based on whether those goals were achieved. It is a structure where goals immediately become the yardstick for evaluation. Because of this, MBO naturally connects with personnel systems such as compensation and promotions.

On the other hand, OKR is designed for 'alignment and focus' rather than evaluation. One of the core characteristics of OKR emphasized by Doerr is absolute transparency. The entire organization, from the CEO's OKR to a new employee's OKR, must be able to view each other's goals. Members grasp at a glance how their work connects to the company's grand goals and focus on the most critical priorities within that scope. This is also why it is recommended not to tie OKR directly to evaluations. The moment it is bound to evaluations, members will only set safe, achievable goals, undermining the original intent of setting ambitious goals.

This philosophical difference also appears in the level of goals. MBO goals generally take on the character of a promise that 'must be 100% achieved.' Conversely, in OKR, particularly Google-style OKRs, ambitious goals (often called 'stretch goals') are intentionally set, and achieving around 70% is actually viewed as an appropriate level. Consistently hitting 100% is interpreted as a sign that the goals were too easy. The contrast between MBO as control and commitment, and OKR as challenge and alignment, becomes clear here.

Differences in Structure and Operating Cycles

Looking at the components makes the differences even sharper. MBO is simply structured around 'Objectives.' It determines what is to be achieved, converts the degree of achievement into measurable standards, and evaluates it. In contrast, OKR clearly separates 'Objectives' and the 'Key Results' that measure them. Typically, 2 to 4 quantitative key results are placed under a single qualitative objective. For example, if the objective is 'make customers love our service,' the key results are expressed in numbers, such as 'achieve X% repurchase rate' and 'achieve Y points in Net Promoter Score (NPS).' Objectives set the direction, while key results provide a structure that objectively proves whether you are moving in that direction.

The two methods also diverge in their operating cycles. Traditional MBO typically has an annual cycle. The flow of setting goals at the beginning of the year and evaluating them at the end operates in tandem with performance reviews. Because of this, a limitation has been pointed out: it is difficult to flexibly change goals even if the market environment changes drastically in the middle of the year. OKR, on the other hand, is based on a short quarterly cycle. Just as Doerr emphasized the experience of writing new OKRs every quarter during his Intel days and 'having a Polaris-like reference point every quarter,' OKR is designed to respond to changes by resetting goals in rapid cycles. The faster the market moves, the more this short cycle acts as a strength.

The directionality also differs. Classical MBO has a strong top-down character where organizational goals flow from top to bottom. Although Drucker emphasized joint setting by supervisors and subordinates, actual operations showed a prominent tendency for upper-level goals to dictate lower-level goals. OKR aims for a mix of top-down and bottom-up. It values driving a sense of ownership and engagement among members by designing about half of the company's goals to originate from the field.

Strengths and Limitations of Each Approach

MBO's strengths are clarity and stability. Because goals and evaluations are directly tied, lines of responsibility are clear, and it is easy to link with compensation systems, providing direct motivation to members. It remains an effective method in organizations where work is relatively predictable and performance metrics are clear, such as stable manufacturing and sales organizations. However, its limitations are also clear. Because it is directly tied to evaluations, incentives arise for members to set conservative goals that are easy to achieve. The rigidity of annual cycles and the side effect of obsessing over short-term numbers are also frequently cited. Even Drucker himself, in his later years, acknowledged the limitations of MBO in the sense that "it is just a tool, and it only works when you know the target, but 90% of the time people don't know the target."

OKR's strengths are alignment, focus, and agility. Through company-wide transparency, it makes the entire organization look in the same direction and rapidly adjusts priorities on a quarterly cycle. Ambitious goal-setting stimulates innovative performance that goes beyond incremental improvement. This is why it particularly suits fast-growing startups or technology and service organizations subject to frequent changes. Conversely, its limitations are also distinct. The principle of separating it from evaluations can actually cause confusion in reality. If member questions such as "So achieving OKRs has no bearing on evaluations at all?" are not addressed sophisticatedly, the system hollows out. Furthermore, designing good key results requires considerable training, and if the organizational culture is not ready to embrace transparency and autonomy, adoption tends to be superficial.

What Should Our Organization Adopt?

The answer is that "it depends on the state and purpose of the organization." Neither side is inherently superior because they solve different problems. Here are a few decision-making criteria to help guide choices.

First is the volatility of the business environment. If an organization has a relatively stable market and products with annual plans that fit well, MBO's annual cycle operates without burden. Conversely, for organizations where priorities change every quarter and rapid experimentation and pivots are necessary, OKR's short cycle is advantageous.

Second is the maturity of the organizational culture. OKR presupposes transparency and autonomy. It works only when there is trust that members can share each other's goals and that setting ambitious goals will not result in punishment even if they are partially missed. An organization lacking such cultural soil that suddenly adopts an OKR detached from evaluations can easily degrade it into "extra work whose purpose is unknown." In this case, it is more realistic to first solidify the basic framework of goals, responsibility, and compensation using MBO.

Third is the primary purpose of goal operations. If the core task is to accurately measure performance and link it to compensation and promotions, MBO's structure is appropriate. On the other hand, if the task is to align all members to a single priority and elicit innovative challenges, OKR is the better choice.

From a practical standpoint, a 'phased combination' rather than an either-or choice is also an effective strategy. Operating evaluations and compensation under MBO's stable framework while introducing OKR as a separate layer for quarterly strategic priority alignment is one such approach. However, unless the distinct purposes of the two systems are clearly explained to members to avoid growing confusion, maintaining a clear division of roles—such as 'evaluations to MBO, directional alignment to OKR'—is the key.

Ultimately, frameworks are merely tools and do not guarantee performance in themselves. Just as Drucker called MBO a philosophy and Grove condensed OKR into two questions, the core lies in the essential question: "Where is our organization supposed to go right now, and are we honestly measuring whether we are getting there?" Organizations capable of answering this question before choosing a framework are likely to achieve performance regardless of which framework they select.