Skip to content
Masteryboth

Metrics & outcomes

Outcomes over output; the one metric that matters.

The short version

Output is what you shipped; outcome is what changed for the customer and the business. Mature PMs are measured on outcomes. The craft is choosing the one input metric you can actually move that predicts the outcome you care about — and resisting the dashboard sprawl that lets everyone feel busy while no one can say whether it worked.

Why it matters

It is easy to ship features and call it progress. But a roadmap of shipped output with no movement in customer or business outcomes is just expensive motion. The shift from "we launched X" to "X moved Y" is the shift from junior to senior PM.

The discipline is distinguishing the metric types and picking well. Output metrics (features shipped, adoption count) are easy to grow and easy to game. Outcome metrics (retention, satisfaction, revenue, cost saved) are what matter but lag and are noisy. Input metrics are the levers you control today that drive the outcome tomorrow — find the highest-leverage one and obsess over it.

How to choose metrics

01
Separate output from outcome. Adoption count is output; whether adoption changed behavior or value is the outcome. Track both, judge on the latter.
02
Find the controllable input. Pick the leading input metric you can move now that predicts the lagging outcome — that is where a PM creates leverage.
03
Pick one north star. One metric that, if it moves, the work was worth it. Supporting metrics exist, but one leads.
04
Instrument before launch. If you cannot measure whether it worked, you have not finished building it (see Definition of Done).
05
Watch for the gaming. Every metric becomes a target and then a bad metric. Pair it with a guardrail so you do not win the number and lose the customer.

Common mistakes

Celebrating output (launches, feature count) as if it were impact.

A dashboard of 20 metrics where no single one is the decision driver.

Optimizing a metric into a target until it stops reflecting real value (Goodhart's law).

Shipping without instrumentation, then guessing whether it worked.

In a startup

Pick a single north star early (often activation or retention, not raw signups) and orient the whole team around it. Vanity metrics are seductive when you need to feel traction — resist.

In an enterprise

You will swim in metrics and reviews. The skill is finding the highest-leverage input metric in a crowded WBR deck and making the case for it — and not letting a leadership-favorite vanity metric distort the roadmap.

Quick check

Optional — test the lesson. Nothing is gated.

1. What distinguishes an outcome from an output?

2. Why pair a target metric with a guardrail?