وقتی یک ایراد چندبار تکرار می‌شود، فرض‌های پنهان را آشکار کنید

When a similar issue returns across several situations, the common response is to attribute it to an individual, their attention to detail, or the team’s willingness to cooperate. That response may create a short-term sense of control, but it usually does not change the cause of recurrence. If outcomes remain unstable despite repeated reminders, another possibility should be examined: people may be making decisions from different assumptions about what correct work means.

This difference does not necessarily mean there is open conflict. One person may define success as delivering quickly, while another defines it as completing the necessary review. One function may see current capacity as a fixed constraint, while another assumes it can be recovered. Each party may be making an understandable decision within its own logic, while the combination of those decisions produces an inconsistent result.

For that reason, repeated messages such as “be more careful” or “coordinate better” usually have limited effect. They do not translate the desired behavior into operational terms. They do not clarify which choice should take priority when speed conflicts with control, quality with capacity, or immediate satisfaction with a deliverable commitment.

In this situation, a manager’s role is not limited to receiving an issue report or assigning blame. The more important element is the observable response after hearing the report. Through the questions asked, the sequence of review, and the way a decision is recorded, the manager signals whether the issue is an individual error or evidence of ambiguity in work design.

That response also affects the quality of later reports. If the first response to bad news is blame or defense of an earlier decision, people may raise information later or provide only part of the context. If the initial response focuses on understanding the decision conditions, warning signs and emerging issues are more likely to be raised earlier.

To move the discussion from personality to work design, a manager can open a short decision record for every recurring issue. Its purpose is not to create extensive documentation. It is to require the involved parties to state assumptions that usually remain unspoken in everyday conversation. The record should be completed before a final solution is presented, not afterward.

The first question is: what does correct work look like to each party? The answer should not end with broad terms such as “the best service” or “high quality.” Each party should describe an observable output. In a hypothetical example, one person may define correct work as sending a response on the same day; another may define it as sending a response whose promises have already been checked for feasibility.

The second question is: which constraint are we assuming is fixed? In many disagreements, people do not disagree about the existence of a constraint; they disagree about whether it can change. Are capacity, time, budget, information, technical approval, or access to a decision-maker actually fixed? Or are they treated as fixed only until a defined path for changing them becomes available?

The third question is: what risk have we accepted? Risk should not be expressed only through general language such as caution or boldness. It should be clear which undesirable consequence is considered acceptable in exchange for which benefit. One team may accept a limited delay to reduce the likelihood of a costly later correction; another team may make the opposite choice.

The fourth question is: who has the right to make the final decision? Many recurring issues arise not from the absence of a decision, but from the presence of several implicit decision-makers. If a specialist, supervisor, control function, and commercial manager each believe they can stop or continue work at a particular point, the process is effectively carried out under shifting rules.

Answering these four questions should not become a meeting for defending positions. The manager should first ask each party to state its view without immediate correction by the other side. The answers can then be placed next to one another, with attention given to the differences: the definition of the output, the constraint assumption, the accepted risk, and the decision authority.

At this stage, the goal is not to identify the right person. It is to reveal where the decision logics diverge. Everyone may refer to the same instruction while drawing different operational conclusions from it. If the difference in interpretation is not recorded, the same disagreement will likely reappear in the next situation using different words.

Separating facts from assumptions is an essential part of this discussion. Facts may include a recorded order status, a stated deadline, confirmed capacity, or the text of an instruction. An assumption is the interpretation used to act on those facts, such as expecting a later approval to arrive on time or assuming that another party will accept a delay.

When facts and assumptions are mixed in one conversation, the discussion can easily become accusatory. A person who considered their assumption obvious may believe that someone else ignored reality. Yet the core issue may be that the organization never specified which assumption had to be confirmed before action could be taken.

A manager’s language must also align with this method. Questions such as “who allowed this?” or “why did you not notice earlier?” may be necessary in an accountability review, but when they are the first response, they push people toward self-defense. A more precise question is: “At the time of the decision, what did you treat as fixed, and what did you treat as changeable?”

After comparing the assumptions, the manager should record a small but clear operational decision. The record may include the issue status, the agreed definition of the output, the valid constraint, the accepted risk, the decision owner, the next action, and the conditions for review. Recording the decision should not replace action; it should directly establish who will do what, within which boundary, in the next similar situation.

Not every issue requires escalation to senior management. However, the organization should know which differences in assumptions deserve escalation. If a decision can create an external commitment, alter an approved level of risk, or force several functions into inconsistent action, it should not remain in an informal conversation. The escalation boundary should be defined by the effect of the decision, not by who speaks most loudly.

This framework also matters in data-driven and AI-enabled work. A system may generate a response or recommendation that appears technically acceptable while teams have not agreed on the standard for accepting it. One function may prioritize response speed, while another prioritizes traceability, human control, or data constraints.

In such situations, the review should not focus only on the system’s output. It should be clear who accepted the recommendation, what control existed before execution, which risk was accepted, and where the authority to stop was located. These questions return accountability from an ambiguous label such as “the system” to specific human and organizational decisions.

Consider a hypothetical example in which a team repeatedly sends a proposal to a customer that later proves difficult to deliver. Sales may assume that response speed is necessary to preserve the opportunity. Operations may assume that no commitment should be communicated before capacity is confirmed. The issue will not be resolved by a general instruction to coordinate more; a shared definition of the initial response, the capacity constraint, the risk of making a promise, and the final decision owner must be established.

After the new decision is implemented, the manager should close the learning loop. Closing the loop means showing the involved people which assumption changed, which rule was revised, and what different behavior is expected next time. If reports are only collected and their outcome remains unclear, reporting will gradually be treated as additional work with little effect.

Progress does not mean eliminating every issue entirely. A more meaningful sign is that, when a problem occurs, the conversation moves sooner from blame to assumptions and decisions. It is also possible to review whether reports include decision context, whether the authority reference is clear, and whether similar decisions are made with similar logic.

Getting started does not require a complex process. A team can review only recurring issues in its operational review meeting using the same four questions. One person can be responsible for recording the decision, and another for following up on the review conditions. If the answers are vague, the issue is not yet ready to be solved; if the answers differ, the true location of the work-design problem has become visible.

The practical principle is simple: before judging people, their decision assumptions must be made visible, comparable, and open to revision. A recurring issue is often not merely an invitation to repeat a reminder. It can be a signal that work boundaries, risk levels, and decision authority need redesign.