The First 90 Days as an Engineering Manager

2026-06-18 · 10 min read · Tools & Productivity

The jump from senior engineer to engineering manager is one of the hardest transitions in tech. Not because the work is harder — but because the definition of "work" changes completely.

As an engineer, your value was in what you built. As an EM, your value is in what your team builds. That shift is obvious intellectually. Internalising it takes months.

Here's a week-by-week guide for navigating the first 90 days without burning trust, shipping the wrong things, or trying to stay an engineer while being a manager.

The Biggest Mistake New EMs Make

Before the playbook: the single biggest mistake new EMs make is trying to prove their value by doing too much too fast.

You restructure the team. You rewrite the processes. You jump into code reviews because you miss coding and it feels useful. You have strong opinions about the architecture and you share them constantly.

All of this signals insecurity, not competence. It undermines the team's autonomy and makes people feel like they're being managed by someone who doesn't trust them.

The first 90 days are not for doing. They're for listening, learning, and building relationships. Everything else comes later.

Days 1–14: Listen, Don't Fix

Your only job in the first two weeks is to understand the current state as accurately as possible.

Run 1:1s with everyone on your team. Not a status meeting — a genuine getting-to-know-you conversation. Ask:

Take notes. Don't share what people tell you with each other. This is information gathering, and the safety of the conversation depends on confidentiality.

Meet your stakeholders. Product managers, design leads, other EM peers, your own manager. Ask each of them: "What does success look like for this team from your perspective? What's not working right now that you'd want me to focus on?"

Read everything. Past incident reports, project postmortems, architecture docs, PRs from the last 3 months. You're building context that will take months to develop naturally — accelerate it.

Observe before commenting. Sit in on team meetings without leading them. Watch how decisions get made, how conflict is handled, how people interact. You're mapping the team's culture.

What you're NOT doing: making changes, expressing strong opinions publicly, or reorganising anything.

Days 15–30: Build Trust, One Person at a Time

By the end of week two you should have a rough mental model of:

Now start investing in individual relationships.

Keep 1:1 cadence consistent. Weekly 30-minute 1:1s with every direct report, every week. No skipping, no rescheduling unless absolutely necessary. This is how trust is built — through showing up consistently.

Start returning value immediately in small ways. Remove a bureaucratic blocker. Get someone unblocked from a dependency. Advocate for something your team asked for. Small wins compound.

Be transparent about your own learning curve. "I'm still getting context on this" is fine to say. What's not fine is pretending you know things you don't, or making decisions without enough information. Engineers have excellent BS detectors.

Don't take sides in existing conflicts. There will be people who try to bring you into existing interpersonal tensions from day one. Listen, don't judge, and don't commit to a position until you understand the full picture.

Days 31–60: Start Shaping

By the end of month one you should have enough context to start having informed opinions. Now you can begin to act — carefully.

Identify the one thing that's blocking team velocity most. It might be a broken deployment process, unclear product requirements, a recurring class of incidents, or a single interpersonal conflict. Pick the highest-leverage thing and start working on it.

Get explicit about how you want to work together. Share your own working style: how you like to be communicated with, what decisions you want to be looped in on vs what you want the team to own, how you'll give feedback. Invite the same from your team.

Start setting clearer expectations. Not rigidly — but people need to know what good looks like. Be explicit about quality standards, communication expectations, and how you'll evaluate performance.

Do your first real 1:1 check-ins. After a month of listening, you now have enough context to give genuine feedback. Share one thing you've noticed each person does well, and one thing you'd like to see them grow in. Keep it light at this stage — you're building the feedback muscle, not delivering a performance review.

Days 61–90: Lead

By month three, you should be operating as a manager, not just observing as one.

Set team goals. Based on what you've learned, work with your team to define what you're optimising for this quarter. Connect it to the company's objectives. Make it visible.

Run retrospectives on team health. How is the team doing on collaboration, communication, and pace? What's one thing you could improve as a team this quarter?

Build your EM habits. The practices that will make you a great manager aren't the dramatic interventions — they're the consistent small ones. Weekly 1:1s. Timely feedback. Keeping track of people's growth goals. Removing blockers fast. Do these reliably and you'll be ahead of 80% of EMs.

Evaluate your own performance. Ask yourself: is the team shipping more smoothly than when I started? Are people clearer on expectations? Have I built the trust necessary to have hard conversations? If the answer to any of these is no, figure out why.

What Success Looks Like

At the 90-day mark, success as a new EM doesn't look like a big structural change you drove. It looks like:

The EMs who struggle in their first 90 days are usually the ones who confused activity for progress. The ones who thrive are the ones who listened first and earned the right to lead.

You don't get trust by announcing that you're trustworthy. You get it by showing up the same way, every week, until people stop wondering what you're going to do next.