What a manager says matters. What they revisit, reward, fund, and panic about matters more.

Every manager has a stated set of priorities — usually documented in a team charter, a strategy doc, or a quarterly planning deck. These are real, but they're incomplete. They describe the official priority landscape. They don't always describe the incentives your manager is operating under: what their boss asks about, what metrics show up in reviews, what gets budget, and what gets punished when it slips.

The skill of managing up includes being able to read the gap between stated priorities and operating pressure — and adjusting your communication accordingly. For ICs, this is especially important: your work may be excellent locally and still miss the altitude your manager is being judged at.

Follow the Pressure, Not the Words

A manager's behavior often makes more sense when you ask: what pressure are they under?

Look for:

  • what their boss asks them about every week
  • which metrics appear in exec reviews
  • which customer, revenue, quality, or hiring problems create sudden urgency
  • which work gets budget, headcount, or calendar time
  • which misses create consequences

This is not cynicism. It is operating context. If your manager is anxious about a board metric, your beautifully reasoned update on a side project may not land unless you connect it to the pressure they are actually carrying.

It also helps with difficult managers. An unclear manager may be reacting to pressure they have not translated. An overloaded manager may only respond to the things that create consequences for them. A conflict-avoidant manager may delay decisions because every path creates stakeholder pain. A political manager may optimize for what their boss will notice. None of this excuses bad management, but it gives you a better map.

How to Build the Model

You won't figure out your manager's real priorities in week one. This is a learning process that takes observation over time.

Ask directly, in a useful frame. Not "what are your priorities" but "what's keeping you up at night these days?" or "if we could only move the needle on one thing this quarter, what would you pick?" The framing matters — it gets past the official answer.

Watch what they bring up unprompted. In 1:1s, in staff meetings, in offsites. Track the topics that recur and the ones that never come up, even when they should.

Notice what changes their attention. When something shifts in their behavior — a new urgency, a new concern — ask yourself what prompted it. Was it a conversation with their manager? A customer issue? A board question? You don't always need to know the answer, but noticing the pattern builds the model.

Test it. The model is only useful if you act on it. If you think their real priority is execution velocity and not feature scope, try leading your next update with execution data instead of feature progress. See if they engage differently. Calibration is iterative.

A useful 1:1 question: “When you talk about this work with your manager, what part do you most need to be able to explain clearly?” That tells you the translation layer your updates should serve.

The Underlying Principle

Your manager is not a monolith to be decoded. They're a person with pressures, accountabilities, incentives, and concerns that are partially visible and partially not. The goal of reading priorities is not manipulation — it's effective partnership. When you know what they are actually being measured against, you can align your work with it, communicate in their language, and be a more useful counterpart.

See clearly. Communicate accordingly.