Your First 1:1 as a New Engineering Manager

Your first 1:1 as a manager is the highest-leverage half hour of your first month, and most new engineering managers spend it doing the one thing that undermines them: talking. You have just been promoted, you feel you should demonstrate competence, so you arrive with a vision, a plan, and a list of changes. Your report leaves knowing what you think and you leave knowing nothing.

The 30-minute agenda that works

Keep it to four blocks and let the middle two run long if they want to talk.

• Set the frame (2 minutes): "This is your meeting, not my status meeting. I'll never use this slot to ask for a project update - that's what standup is for. If you ever have nothing, we cancel and you get the time back."
• Understand the work (10 minutes): What are you working on? What is slower than it should be? What would you fix if nobody said no?
• Understand the person (10 minutes): How do you like feedback - immediately or considered? What does a good week look like? What do you want to be doing in two years?
• Close the loop (5 minutes): Say what you heard, name the one thing you will act on, and book the next one.

The three questions that get real answers

Generic questions get generic answers. These three consistently surface something useful in a first conversation:

• "What is something the team does that you think is a waste of time?" This gives explicit permission to criticise and tells you what your predecessor could not hear.
• "When was the last time you were genuinely stuck, and what did you do?" Reveals whether people ask for help here, which is the single best proxy for psychological safety.
• "How will you know if I'm doing a bad job?" Uncomfortable, memorable, and it establishes from minute one that feedback runs both ways.

What not to do in the first one

Three failure modes account for most bad first 1:1s:

• Do not promise anything. You will be asked about promotions, a project change, a pay review. The correct answer is "I don't know yet, and I'm not going to guess. Give me two weeks and I'll come back with a real answer." A promise you cannot keep costs more than a delay.
• Do not relitigate the past. You will hear about your predecessor. Listen, do not agree, do not pile on. "That sounds frustrating" is enough.
• Do not fix anything live. The instinct to solve the first problem you hear is the engineer in you. Write it down instead. Solving it in the room teaches people that 1:1s are where you go to get overruled.

The hard case: a report who wanted your job

If someone on your team applied for the role you got, address it in the first 1:1 rather than waiting for it to surface. Not addressing it is read as either obliviousness or cowardice, and both are worse than an awkward two minutes.

You have acknowledged reality, avoided false sympathy, and asked them for something. That last part matters: people re-engage when they are asked to contribute, not when they are reassured.

Set the cadence and never cancel

Weekly for new reports, fortnightly once things are stable, thirty minutes, same slot. The cadence matters less than the reliability - a manager who reschedules 1:1s when things get busy is teaching their team that they are the first thing to be dropped under pressure, which is exactly when people need them.