This week was mostly conversations — new scope, new people, a lot of catch-ups. When you step into a bigger job, the temptation is to start deciding. The better move is quieter. Four things I was reminded of.

First listen

1. In a new mandate, your first job is to listen, not to fix.

Most of my calendar this week was one-on-ones and intro meetings. I went in with topics, but the goal wasn’t to land opinions. It was to understand how the work actually flows and where people feel friction before I change anything.

The failure mode when you inherit scope is shipping changes on week-one confidence. You break trust and you usually break something that was load-bearing for a reason you couldn’t see yet. Takeaway: treat the first few weeks of any expanded role as a listening tour. Ask how things work and why, take notes, and earn the right to redesign before you do.

Make yourself unecessary

2. The best thing a senior person can do is make themselves unnecessary.

I spent real time this week mapping where the most capable people actually spend theirs — the product-ownership and coordination work that had quietly collected on a few individuals — and started handing it off, with named volunteers lined up to take each piece. It feels counterintuitive to pull your strongest contributor off the thing they’re best at. But when the work only moves if one person touches it, you don’t have leverage. You have a bottleneck and a single point of failure wearing a cape.

The transition only works if it’s deliberate: name the pieces, name who takes each, and talk through the handoff rather than assuming it. Takeaway: if your best people are irreplaceable on the details, that’s a problem to solve, not a comfort to enjoy. Distribute the ownership, let others carry it, and coach from the side. You’re building future leaders, not preserving your own indispensability.

Precise enough to make a decision

3. Ask for the back-of-the-napkin number before the precise one.

Sizing some new work this week, I asked for a low-confidence, rough estimate — not a high confidence well thought out plan. A day-one rough number tells you whether something is a week or a quarter, which is usually the only fact the next decision needs. A precise estimate costs a week of engineering time to produce and often changes nothing.

Takeaway: match estimation effort to the decision in front of you. Get the order-of-magnitude first; buy precision only once the direction is set. Ship, learn, iterate applies to planning too.

Vanity metrics

4. Be suspicious of the metric that flatters you.

I sat through a walkthrough of an engineering-metrics tool run against my own team’s data. Plenty of dashboards will happily tell you your team is busy. The question I keep coming back to: what decision would this number change? If a metric only makes me feel productive but wouldn’t alter what I do next, it’s decoration.  A useful one that jumped out for me is that once PR size is > 200 lines of code, the biggest it gets the higher chance it will take very long to review and also the higher chance of it introducing quality regressions.  It is only a signal, but it is one paying close attention to.

Takeaway: before adopting any measurement, name the action it will drive. Tie it to customer value and outcomes, not activity. Volume and velocity charts are the easiest vanity to fall for because they’re always going up and to the right.

Putting it all together

A bigger job rewards restraint more than it rewards motion. Listen before you redesign. Give your best work away so others grow. Spend estimation effort where the decision actually turns. And measure the things that change a decision, not the things that make the slide look good.