Calibrated Autonomy: The Leadership Skill That You Are Missing

Seniority is not a permanent trait. It is the level of autonomy someone can exercise in a specific skill while sharing responsibility for risk and outcomes.

Blog Image

No matter how many years of experience you have, there is always a question that haunts us as leaders: when should we help, and when should we step back?

We do not want to micromanage people, but we also want to see them make sound decisions. We do not want to be perceived as micromanagers, but we also do not want to gamble with our accountability.

Both concerns can become logic traps. Leadership is not about avoiding involvement at all costs, nor is it about making sure your team succeeds flawlessly without ever needing you. It is about calibrating your involvement to the person, the skill and the risk of the situation.

Seniority is not a permanent trait of a person, but the level of autonomy someone can exercise in a specific skill, under a shared understanding of risk and accountability.

I will give you a simple framework that you can apply every day with your direct reports. It helps you balance risk and autonomy, leaning in and leaning back, and perception and results.

Let’s use Tim, a new member of the team, as a simple example. Tim starts asking you questions about how to do his job. It is not that Tim is not smart. He is calibrating himself against you, your expectations and your leadership style.

The first thing we must strive for is creating a safe space. Whenever Tim asks a question, do not belittle him or give him the impression that his question is stupid. Whether he asks something appallingly basic or extremely intricate, your first reaction should not be to answer, but to question him back.

How would you solve this? What do you think?

You might worry that asking a junior to think first could damage your authority or perceived expertise. In reality, it creates a coaching moment, a learning experience and a chance to calibrate your judgement with theirs.

Listen to the answer. If Tim knows the right answer, validate it immediately. Do not simply say, “Okay, do that.” Explain why it is the right approach and add your own judgement. Your validation should make it clear that you are accepting accountability for the decision, not merely nudging him forward.

Instead of saying, “Okay, good idea. Do that, Tim,” try something more explicit: “Tim, that is exactly how we should proceed. Good work. If you have any other questions, do not hesitate to ask.” Tim gains confidence, and you have clearly accepted accountability for what he is about to do.

However, let’s be frank: Tim will not get it right every time.

If Tim gives you the wrong answer, challenge him constructively: “There are a couple of reasons why that would not work: A, B and C. Let’s think it through. Can you come back with three other ways of doing it?”

Notice that we did not tell Tim that he is wrong or foolish. We explained why the approach would not work and added new information. Tim may not have known those factors or thought to consider them. This is another coaching moment: You are calibrating his judgement, expanding his context and strengthening his critical thinking.

Ask him to come back with more than one proposal. Depending on the problem, two or three options are enough.

We then have two scenarios:

  • He comes back with N incorrect or incomplete proposals.
  • One of his proposals (or more) is actually good.

In the first case, ask Tim questions such as “That’s interesting. What made you think of that?” or “Can you help me understand option B a little better? Tell me about XYZ.” You may already know that the proposals are incorrect, but you should still explore his thought process. Accompany him through his thinking journey so you can see whether he is paying attention and thinking deliberately. This also gives you additional coaching opportunities.

Once you have explored his ideas and confirmed that they will not work, it is time to give him clear direction: “All right, these options will not work due to the reasons we just discussed, but that is perfectly fine. What I want you to do is XYZ.” You can appreciate this is not micromanagement. We are now making a deliberate decision which also creates the space and time for calibration and thought.

If you fall into the immediacy trap and give Tim the right answer from the start, he may complete the task, but he will also learn two things:

  • My manager is there to tell me what to do and how to do it.
  • My job is to do what my manager says.

What may look like empathy can actually be a way of ruining autonomy. You dilute accountability and make your team less scalable. The moment you become unavailable, everything stops, including their growth.

If one of his proposals is good, leave it until the end. Explore the others first and then, just like I said before, validate it. Do not simply send Tim to do the work. Take him through why it is the right approach.

Tim can now complete the task with confidence.

All right! Let’s move away from the hypothetical example and into the framework.

The three levels of Calibrated Autonomy

I have shown you two of the three situations you can expect with this framework. Let’s list them and see why the third one should get you excited.

Level 1 — Needs help framing the problem

  • Situation: Someone comes with a problem, has no idea what to do and asks you about it.
  • Your stance: Do not give them the answer directly. Ask what they think could be done. If they have no idea, ask them to take the time and space to think and come back later. You can timebox this: “Come back to me in an hour.” This prevents an infinite brainstorming session.
  • The result: Whether they find the solution or not, you are helping them calibrate their thought process. You are showing respect for their ideas while making your accountability clear.

Level 2 — Brings possible solutions

  • Situation: Someone comes with a problem and already has possible solutions. They ask you which one to choose.
  • Your stance: They are learning that you expect them to think before involving you. Straightforward problems are solved without your involvement, while genuinely difficult problems reach you with useful context.
  • The result: You explore the options together, weigh the pros and cons, and either choose one or explain why a different direction is needed. The person learns how to evaluate trade-offs rather than wait for instructions, and continues to calibrate their thinking with yours.

Level 3 — Makes a sound decision independently

  • Situation: Someone comes with a problem, considers several solutions and applies the one you would have chosen.
  • Your stance: As you coach this team member through Level 2, make it clear that you expect them to build the confidence to do this independently. You remain accountable for the decisions, but you no longer need to approve every step. Through repeated interactions, you have calibrated your judgement with theirs.
  • The result: They can now make sound decisions without needing your constant involvement. You have built their autonomy without abandoning your own responsibility.

People are Junior and Senior at the same time

These levels do not describe the person. They describe their current autonomy in a specific skill. Someone can be Level 3 at design but Level 1 at writing content, Level 3 at coding but Level 1 at automating quality, and Level 3 at coaching others but Level 1 at understanding business politics.

You do not need to keep a profile dossier for every employee or write down their levels. Do not turn this into a faceless, empathy-less bureaucratic form.

People are always concerned about their performance. In practice, this often means they are concerned about how their manager perceives it. Conflict arises when those views do not overlap, especially when a long-tenured person’s view of their performance is completely misaligned with their manager’s.

The risks of this system

The risk of any system is that it can be gamed.

If you introduce the concept too explicitly (as in, you explain it to them), people may start bringing three proposals every time, even when the problem does not require them. Or they may jump the gun too often, gambling so they can appear to be at Level 3. They could irresponsibly play the game instead of learning and getting things done.

Use this framework as a coaching tool, not as a performance test. If people become too focused on the framework, they may interpret every intervention as a judgement about their performance.

If other managers report to you, they could start hiding facts or problems so that you do not get involved, as they fear you will perceive them as “junior”. Here, the issue is not seniority or skill. It is professionalism, and that is a different conversation. But do not blame them. People feel insecure for many reasons. You could be a fantastic person, and never have ill intentions when helping others, but perhaps you forgot about the layoff round that happened last year, and yet they still remember, so now they are playing it safe. The circle of safety is not something you build yourself only, it is affected by the context. Keep that in mind.

So when I tell you to not introduce people too early into this framework, it is not because people should not understand in detail what’s happening, but because you need to develop your own awareness and make a deliberate and calculated decision about when it is a good time for them to learn it. Your decision must have an intent. Sharing this just to say you are transparent is simply a rationalisation of your own ego, and it is not enabling them to grow.

There are also situations where this framework should not be used. When safety, compliance, critical incidents, irreversible decisions or extreme time pressure are involved, give direct instructions first—and explain the reasoning when the situation allows.

Autonomy is not an excuse to delay action or transfer risk to someone who is not ready to carry it.

What you should do now

Start applying this framework when you coach others. You do not need to explain, you do not need to link this article, you do not need anyone’s permission to start practising this.

  • You will improve the results you seek.
  • Your people will receive help at the right time.
  • You will build a leadership style that knows when to lean in and when to lean back.

The goal is not to remove your involvement, but to make your involvement remove dependence.

No one likes micromanagers, but we all need a hand sometimes. Now you know what to do and when.

Click here for the final revelation

None of this works if you do not build your own competence.

Christian Guzman

Christian Guzman

I am a seasoned technology leader with over 20 years of experience, passionate about building resilient teams and human-centric systems.

More Like This