For years, I worked as a Team Lead, and I got used to being involved in almost every important technical decision.
I knew the codebase. I understood the architecture. I had opinions about how things should be built, and I was comfortable pushing for my solutions. Sometimes, I pushed harder than I should have.
That involvement gave me a sense of control. And because the projects I worked on were successful, I had a convincing reason to keep operating that way.
From my perspective, the results supported the approach.
Then… I started leading multiple teams.
Initially, I carried the same habits into the bigger role. I still wanted to understand the details, influence technical decisions, and make sure we were choosing the right solutions. The scope had changed, but my understanding of how to contribute had not.
Gradually, it became impossible to sustain.
I could no longer be fully immersed in every team’s work. There were too many decisions, too many conversations, and too much context to keep in my head. I was still participating with the confidence of someone who knew the details, even when I no longer knew enough of them.
Some of my decisions were poor…
That was uncomfortable to admit. I had spent years building credibility through technical judgment. Now, the way I was trying to preserve that credibility was exposing the limits of my understanding.
I could feel myself losing control, and I was afraid of what stepping back would mean. If I stopped being the person with the answers, would people still see me as credible? What exactly would make me valuable?
Eventually, I made a deliberate decision to change how I led. I wanted to build my contribution around empowering other people to lead, make decisions, and take responsibility.
That decision gave me a direction. Making it real required a much deeper change than simply delegating more tasks.
I had to question the story my previous success had taught me.
When projects go well while you are heavily involved, it is easy to conclude that your involvement is what made them successful.
Sometimes it contributed significantly. But successful delivery does not tell you whether every intervention was necessary, whether another approach would have worked, or how much more independently the team could have operated.
I had real experience and useful technical judgment. I also had a habit of treating those strengths as a reason to remain central.
Leading several teams exposed the limitation. My attention had become a shared dependency. The more decisions I tried to influence, the less context I had for each one.
And that created a difficult contradiction: I was trying to protect quality by staying involved, while my limited understanding could make some decisions worse.
If you recognize this pattern, start by examining where your authority exceeds your current context.
Look at a few recent decisions you influenced. What did you know firsthand? What were you assuming? Did the person closest to the work have information you had not understood? Did your input improve the decision enough to justify the delay or redirection it caused?
Technical experience remains valuable. It does not give any of us unlimited visibility.
The fear of losing credibility made the change personal.
For a long time, technical competence had been a dependable source of credibility for me. I understood the problem, had a solution, and could explain why I thought it was right.
Moving away from that meant changing something deeper than my calendar.
A week spent aligning priorities, developing a lead, or resolving a disagreement between teams produces a different kind of evidence. You may have contributed significantly without being able to point to a solution that was yours.
That can feel unsettling when your identity is built around being the person who knows.
Herminia Ibarra’s research on professional identity offers a useful perspective here. In a qualitative study of professionals moving into more senior roles, she described people adapting by observing role models, experimenting with possible professional identities, and evaluating those experiments through feedback and personal standards.
My reading of this for engineering leadership is that you have to practice the new contribution before it feels natural.
Waiting until you feel completely comfortable letting somebody else decide can keep you in the same pattern indefinitely.
Choose one recurring situation where you normally establish your value through the answer. Let your lead formulate the recommendation. Help them examine the tradeoffs. Leave the decision with them when it falls within their authority.
Then look for a different kind of evidence: the quality of their reasoning, the clarity of the decision, and their ability to handle a similar situation next time.
You are building a new basis for credibility. It takes deliberate attention to recognize it.
I also had to look honestly at what control was giving me.
I cared about the projects. I wanted us to make good decisions. Those intentions were real.
But insisting on my solutions also gave me something personally: confidence that things were happening in a way I understood and could influence.
This is where ego enters the conversation. I use that word to mean our attachment to a particular image of ourselves… the capable expert, the reliable problem solver, the person others need.
You can want strong, independent leads and still feel uncomfortable when their independence makes your own expertise less central.
That discomfort can appear as another design review, another request for detail, or another argument for why your approach is better.
A question I would encourage any leader to ask is:
Am I protecting an important outcome, or am I struggling to accept that this can work without my solution?
Imagine your lead proposes an approach you would not have chosen. It meets the requirements, respects the constraints, and carries an acceptable risk.
Can you support it?
If your real standard is “the decision must match the one I would make,” you have given people very little room to lead.
The practical work is learning to distinguish a material concern from a preference that carries extra weight because of your position.
Empowering people needs to become an explicit agreement.
Once I decided to change, “empowering others” had to mean something more concrete than being less involved.
Telling people to take ownership is easy. Making it clear what they own… and behaving consistently when they exercise that authority… is harder.
For each recurring decision, establish five things:
Who owns the decision.
What outcome they are responsible for.
Which constraints they must respect.
What requires escalation.
When and how progress will be reviewed.
Consider a Tech Lead responsible for a service migration. They might own the implementation approach, with agreed expectations for reliability, security, and staffing. They escalate if the plan changes another team’s commitments or introduces risk outside their authority. You review progress at defined milestones.
Inside that agreement, they have room to decide.
Some decisions can be made independently. Others require consultation. A smaller set may require your approval. Make those distinctions explicit, because people should not have to discover their authority by accidentally crossing an invisible line.
A 2018 meta-analysis by Gavin Slemp and colleagues, covering 72 studies and 32,870 participants, found that perceived leader autonomy support was positively associated with autonomous motivation, well-being, and positive work behaviors. The findings describe associations; they do not establish that a particular delegation practice will cause better results.
The implication I draw is that ownership needs meaningful influence. Giving someone accountability while retaining all consequential decisions leaves them responsible for work they cannot fully shape.
The agreement also has to survive disagreement. If you reopen every decision you would have made differently, people will learn to wait for your approval regardless of what you said.
My technical background needed a different purpose.
The experience that helped me as a Team Lead still had value. I needed to become more selective about where and how to apply it.
A leader of several teams needs enough technical understanding to assess risk, challenge weak assumptions, and connect engineering choices to business consequences.
The depth required depends on the decision.
If several teams are migrating a shared platform, you need to understand the major dependencies, reliability risks, capacity demands, and conditions that would require changing the plan. A questionable assumption that threatens the entire migration may deserve detailed investigation.
That does not automatically make you the owner of every implementation choice.
Before going deeper, ask:
What decision in my role will this additional detail help me make?
If you can answer clearly, investigate. If the answer is mostly that you feel uncomfortable without knowing everything, recognize the limit you are trying to avoid.
You also need to adapt your involvement to the person and the risk. A new lead working in an unfamiliar area may need closer support. An experienced lead handling a reversible decision within clear boundaries usually needs less.
Explain why you are getting involved, what you need to understand, and what will allow you to step back. Otherwise, temporary support can turn into permanent supervision.
The work becomes developing judgment in other people.
For me, this is the substance behind the decision to move into leadership through empowering others.
You need people who can assess situations, make tradeoffs, and act responsibly when you are elsewhere.
When a lead brings you a decision, understand their reasoning before contributing your own answer. What are they optimizing for? What alternatives did they consider? Which assumptions matter most? Who else is affected? What would change their recommendation?
The conversation should reveal what support they need.
Sometimes they lack technical knowledge. Teach it or involve someone who can. Sometimes they lack business context. Share it. Sometimes they already know what to do but are unsure whether you will support them.
Do not make them guess which of those problems you believe they have.
And do not turn every conversation into a coaching exercise. If you know an important fact, say it. If the decision requires your authority, make it. Questions become frustrating when they replace help you could provide directly.
Development also needs to be specific. One lead may need to connect technical tradeoffs to business priorities. Another may need to handle disagreement. An EM may need to address underperformance instead of repeatedly absorbing unfinished work.
Pick a real situation, agree on what they will do, and review it afterward.
Discuss the reasoning as well as the outcome. A good decision can produce a disappointing result under uncertainty. A weak process can get lucky. People develop better judgment when you can examine both without confusing them.
If clear expectations, appropriate support, and repeated feedback do not produce improvement, address the performance or role-fit issue. Quietly doing the person’s job around them will keep the dependency in place.
Giving people ownership also changes how you respond to problems.
This is where the commitment gets tested.
It is relatively easy to let someone own a decision when everything is progressing smoothly. When something goes wrong, the temptation to reclaim control becomes much stronger.
For routine escalations, ask for their assessment, their recommendation, and the specific help they need. They may need a priority decision, additional capacity, or your intervention in a conflict with another team.
Providing that help does not have to transfer the whole problem to you.
Urgent issues are different. A production incident, security concern, or serious people issue should be raised quickly, even when the person has no solution to propose.
Amy Edmondson’s study of 51 manufacturing teams found that psychological safety was associated with learning behavior. Her work describes a shared belief that the team is safe for interpersonal risk taking. Although the study was not about engineering management, it offers a useful foundation for understanding whether people can admit uncertainty and surface mistakes.
My application is simple: ownership must remain compatible with asking for help.
If people interpret empowerment as “you are on your own,” you may receive fewer escalations while becoming less informed about the risks.
When you do take over temporarily, explain the reason and make the return of ownership explicit. Otherwise, a difficult week can undo months of work toward independence.
Other people need to see that your leads have authority.
Your behavior in front of stakeholders matters as much as what you say in a one-to-one.
If someone bypasses a lead and gets a decision directly from you, you have taught them a faster route. The lead is then left responsible for delivering a commitment they did not make.
Reconnect the stakeholder with the decision owner. Provide missing context, clarify urgency, and help resolve anything outside that person’s authority.
Be equally careful with informal technical suggestions. An engineer may hear a passing idea from their manager’s manager as an instruction. State whether you are offering a possibility, asking for an investigation, or making a decision.
You should still talk directly with engineers and keep an independent understanding of team health. People also need safe routes for concerns about their own manager.
The point is that authority should be clear enough for your leads to act, while support and escalation remain accessible.
You need to use your attention differently once you stop owning every local decision.
This is an essential part of the transition I wanted to make.
Leading multiple teams creates responsibilities that are easy to neglect while you are absorbed in each team’s implementation details.
Two teams can make reasonable decisions independently and still produce a poor overall result. They may optimize for conflicting priorities, depend on the same scarce expertise, or assume that somebody else owns a critical integration.
Your work includes making those conflicts visible and resolving them with the relevant leaders and business partners.
You need to align commitments with capacity, clarify priorities, resolve ownership gaps, and develop the people responsible for delivery. Sometimes that means stopping work or challenging a commitment that should never have been made.
Discuss this mandate with your own manager. If the organization still expects you to approve every design while leading multiple teams, the expectations need to change too.
You cannot sustainably add this work on top of an unchanged Team Lead role.
Look at your calendar and ask whether it reflects the contribution you intend to make. Your one-to-ones should include leadership development. Meetings across teams should resolve shared decisions and obstacles. Routine status reporting should leave space for those conversations.
The practical result should be teams with clearer direction and leaders who can act effectively within it.
If you are in this transition, start with one visible change.
You do not need to hand over everything at once. You need a credible way to practice a different approach.
For the first week, record the decisions and escalations reaching you. Identify which require your authority and which arrive through habit, unclear ownership, missing context, or a development need.
In the second week, choose one recurring decision category and agree on its boundaries with the relevant lead. Give them the context and support needed to own it. Choose something meaningful with a downside you can manage.
In the third week, pay attention to your response when you disagree or feel the urge to take over. Name the actual risk. Provide the help required. Respect the authority you agreed to give.
Use some of the attention you recover to address a problem between teams that you have been postponing.
In the fourth week, review the experiment together. Were decisions timely and sound? Did concerns surface early? Was your involvement useful? What remained unclear? Adjust the agreement based on the evidence.
Look at independence alongside delivery, reliability, and the visibility of risk. Fewer escalations alone can be misleading. The aim is stronger judgment and less unnecessary dependence, with problems still reaching the right people early.
The decision I made was to stop treating my own technical involvement as the foundation of how I led.
That meant accepting that I could not be fully informed everywhere, that my solutions were not always the best ones, and that protecting my credibility by staying central could make me less effective.
For years, being deeply involved had helped me feel in control. Leading multiple teams forced me to confront the limits of that approach.
The question I would leave you with is the one at the center of this transition:
If your teams could make good decisions without you, would you feel successful… or would you feel less important?
If you recognized yourself and need help with this transition, feel free to reach out. I coach senior engineering leaders on exactly this, in my Get Out of the Loop program we’ll address the real situations from your work and have measurable results. It’s time for you to reach the next level of leadership.

