Breaks a North Star into its drivers, and those drivers into input metrics teams can actually move — creating a direct line of sight from daily work all the way up to the top-level outcome.
▸ Try the interactive toolA metric tree — also called a KPI tree or metric decomposition — breaks a North Star Metric into its constituent drivers, and those drivers into actionable input metrics. It creates a direct line of sight from the daily work of a team at the leaves all the way up to the top-level outcome at the root.
The North Star sets direction but is too high-level to act on — no one moves 'weekly active users completing the core action' directly. The metric tree decomposes it: the North Star splits into a few drivers (e.g. new users × activation rate × retention), each of which splits into input metrics a team can actually influence. The result is that any team can find their input metric on the tree and see exactly how moving it rolls up to the company's outcome — which turns an abstract North Star into a set of concrete, ownable levers.
Root: the North Star (the outcome).
↓ Drivers: the few factors that mathematically compose it.
↓ Input metrics: the actionable levers teams own.
Every leaf has a line of sight to the root.
A North Star with no decomposition is inspiring but inert — teams can't act on it, so it floats above the daily work without connecting to it.
| The shortcut | What it costs | What it gives you instead |
|---|---|---|
| North Star with no tree | Inspiring but un-actionable; teams can't move it directly. | The tree decomposes it into levers teams can actually pull. |
| No line of sight | Teams can't see how their work connects to the outcome. | The tree links every input metric to the North Star. |
| Conflicting local metrics | Teams optimise inputs that don't roll up coherently. | A shared tree ensures local work composes into the outcome. |
| Guessing at drivers | Unclear which inputs actually move the North Star. | Decomposition makes the causal/mathematical structure explicit. |
Start with the single value metric (Tool 04) as the top of the tree. Everything below must roll up into it.
Break the North Star into the handful of factors that compose it — often mathematically (e.g. North Star = new users × activation × retention). Keep it to a few; a tree with twenty branches guides no one.
Decompose each driver into the specific metrics teams can directly influence. These leaves are where daily work happens — they must be genuinely movable.
Check that moving each leaf metric actually rolls up to move the North Star. A leaf with no real path to the root is a vanity metric in disguise.
Give teams ownership of specific input metrics, so each knows exactly which lever is theirs and how it contributes to the whole. This is what turns the tree into aligned action.
A team had a well-chosen North Star but found it frustratingly abstract — no one could say what to do to move it, so it became a poster rather than a guide. Daily work felt disconnected from the company's headline metric.
Building a metric tree fixed it. They decomposed the North Star into its drivers (acquisition, activation, retention), then broke each into input metrics specific teams could own — onboarding completion, feature adoption, and so on. Suddenly every team could find their leaf on the tree and see the line of sight to the top: improving onboarding completion measurably rolled up into the North Star. The abstract metric became a set of concrete, ownable levers.
The deliverable is a decomposition from North Star (root) through drivers to owned input metrics (leaves), each with a verified line of sight to the top.
| Level | Contains | Owned by |
|---|---|---|
| Root | The North Star | Whole company |
| Drivers | The few composing factors | Leadership / function |
| Input metrics | Actionable levers | Specific teams |
| Test | Each leaf moves the root | — |
The metric tree is what turns a North Star from inspiration into instruction. It gives every team a line of sight from the lever they pull to the outcome the company cares about — the structural form of alignment.
The decomposition does two things at once. It makes the North Star actionable, by breaking an un-pullable headline metric into levers teams can actually move; and it makes local work coherent, by ensuring those levers compose mathematically into the outcome rather than pulling in different directions. The verification step — checking that each leaf genuinely rolls up to the root — is also a powerful vanity-metric filter: an input metric with no real path to the North Star is activity masquerading as progress. A well-built tree is the practical bridge between Module 5's strategic metrics (North Star, hierarchy) and the daily prioritisation of what to actually work on — because it shows, concretely, which inputs are worth a team's effort.
A sprawling tree guides no one. Decompose into a few key drivers, not twenty.
An input metric that doesn't roll up is a vanity metric. Verify each path to the root.
A tree no team owns is a diagram. Assign input metrics to specific teams.
The tree makes the North Star actionable; it doesn't replace it. Keep the root clear.
The tree's root is the North Star (Tool 04); the tree is what makes it actionable.
The tree is the explicit line of sight the hierarchy (Tool 02) calls for.
Knowing which input metrics move the North Star informs what's worth building (Module 1).
Teams' input-metric targets become their OKRs (Module 2), tying daily work to the outcome.
Take a North Star metric (e.g. weekly active users completing the core action). Break it into two or three drivers that mathematically or causally compose it.
For one driver, name an input metric a specific team could own and directly move. Trace its line of sight back to the root.
If you can show how one team moving one leaf metric rolls all the way up to the North Star, you've built the line of sight that turns an abstract outcome into concrete, ownable work.