How to manage an employee who knows their job better than you do

The manager and the expert answer two different questions

A manager takes over a team of engineers and discovers that one of them knows the system better than anyone else in the company. In a meeting she says: “I decide, because I’m the boss,” even though she cannot assess the technical arguments. The expert reacts with irritation, and the others stop speaking up. Having the edge in specialist knowledge is not the same as being responsible for the budget, safety, the client and how the team works together. The manager’s job is to bring these perspectives together, not to win a knowledge contest. An employee can have full autonomy in choosing a technical solution within agreed limits, while their manager is responsible for priorities and for choosing between the organisation’s competing needs. A good conversation starts with a sentence like: “You know this system better than I do. I need to understand the possible options, their costs and their risks so that I can make a decision about the project as a whole.” Acknowledging someone’s expertise does not diminish your position. It shows why the position exists in the first place.

Research does not say that a manager’s knowledge of the field is irrelevant

In a study by Artz, Goodall and Oswald, based partly on longitudinal data, a supervisor’s technical competence was strongly associated with employee satisfaction. In the model analysed, it was even the strongest single predictor of job satisfaction. This does not mean that a manager must be able to do every task better than their direct reports. The finding is a reminder, however, that a manager’s complete ignorance of the team’s work makes it harder to assess effort and undermines the credibility of the conversation. A manager can learn the language of the field, the basic constraints of the process and the typical risks without trying to become the best specialist on the team. They should understand which questions to ask and how to assess the evidence. A good step is to review several real cases together: where a failure comes from, how to detect an error and which signals call for escalation. The expert gains a conversation partner who values their knowledge, and the team gains a manager who can make a responsible decision.

Trust does not mean giving up on checking risk

“I trust you, so I won’t check anything” is just as extreme an approach as monitoring every step. In a meta-analysis of 112 independent studies covering 7,763 teams, De Jong and colleagues found a positive relationship between trust within a team and its performance. The strength of the relationship depended, among other things, on task interdependence and differences in skills. This is not an experiment showing how much oversight each project requires. In practice, you can separate the freedom to choose a method from the obligation to set out assumptions, tests and risks. A developer can choose a library within the agreed standards, but changing the core architecture requires consultation if it affects security or several teams. The question “How will we check whether this works?” should not be taken as questioning someone’s professionalism. It is worth explaining that the manager is accountable for the consequences of decisions and therefore needs clear criteria and a visible review process.

An expert’s experience can be valuable and fallible at the same time

An experienced specialist says: “We’ve always done it this way, so it’s bound to work.” A statement like this may reflect genuine practice, but it does not close the discussion. In their analysis of the conditions for expert intuition, Kahneman and Klein identified two important elements: a relatively predictable environment and the opportunity to learn from clear feedback. A sense of certainty does not, on its own, prove that a judgement is accurate. So a manager might ask: “Which similar cases are you basing your recommendation on? What would tell us that this situation is different? How can we test the assumption safely?” It is important that these questions do not sound like an exam designed to humiliate. In new, unstable conditions, even an excellent expert may need data from beyond their own experience. Conversely, a manager who questions every answer on principle wastes the team’s trust and time. What helps here is clear assessment criteria, not a struggle for authority.

Agree three levels of decision and a way of resolving disagreements

Together, list the decisions the expert makes alone, those that require consultation and those that belong to management. The first group might include day-to-day optimisation, the second changes that affect other teams, and the third legal or strategic risk, or a large budget. In each case, agree how key assumptions will be documented. When you disagree, start by naming the criterion: safety, cost, deadline or whether the decision can be reversed. “I see it differently, but I don’t know who’s right” can be replaced with the question of what additional test would settle the matter. If the decision cannot be made to depend on a test, the manager should make clear on what basis they are choosing an option and when they will come back to assess the results. The expert may hold a different view, but should know which decisions they are responsible for. A well-managed team is not one in which the manager knows all the answers. It is one in which everyone knows how decisions are made when knowledge is incomplete.

The strongest authority comes from taking responsibility, not from posing as an expert

When a manager admits, “I don’t understand this mechanism yet – can you explain it to me?”, they may earn more trust than if they improvise a technical explanation. But they must then keep their side of the bargain: get to grips with the data, decide on time and protect the expert from chaotic changes of instruction. Acknowledging the other person’s knowledge does not mean handing over to them all the difficult conversations with the client or responsibility for the budget. Nor should the expert use their information advantage to shut down discussion. It is worth discussing how recommendations should be presented: the conclusion, two alternatives, the main risk and how it will be verified. Once a decision has been made, the manager explains the choice to the team and later comes back to assess the outcome. When everyone knows their role in this process, there is no need to keep proving who is the cleverest. The specialist’s knowledge and the manager’s responsibility can complement each other.

Managing an expert is not about competing with them on knowledge or handing over control completely. A manager should understand the constraints of specialist work, agree the scope of autonomy and make decisions based on transparent criteria. Respect for expertise goes hand in hand with responsibility for the team as a whole.

How Empatyzer helps a manager work with an experienced expert

The new manager of an IT team needs to hold a conversation with a specialist who has known the system for ten years and dismisses most of her questions as “irrelevant”. She describes to Em specific situations, the purpose of the conversation and the limits of her own responsibility. Em helps her start by acknowledging the expert’s knowledge and then calmly agree which decisions he can make himself and how he will present risks affecting the whole project. If relationship tips are available in the organisation, they can help adapt the tone of the conversation to differences in the need for autonomy, control and style of argument. The manager can rehearse a response to the line “You don’t understand it, so you can’t make the decision” without getting drawn into a contest. Em helps her put together questions about tests, alternatives and whether a decision can be reversed. She can then prepare a short summary for the team: what belongs to the expert, what requires consultation and who is responsible for the final decision. Micro-lessons develop the skills of delegation and of holding conversations with people who have different working styles. The practical result is collaboration based on competence rather than on a struggle for position in the hierarchy.