Most product teams track too many metrics. They have dashboards full of numbers, weekly reports covering dozens of indicators, and meetings where half the time is spent debating which metric to look at. The result is that everything looks like it matters, which means nothing really does.
The North Star Metric is the antidote. It's a single number that best captures the core value your product delivers to users, and that predicts long-term growth when it moves in the right direction. This guide explains what it is, how to choose it, and how it fits with the rest of your metrics framework.
Definition
Defining the North Star Metric
A North Star Metric (NSM) is a single metric that:
Captures core value
It measures the moment when users receive the fundamental value your product promises. Not a proxy for value, but the actual delivery of it.
Predicts growth
When this metric improves, revenue and retention tend to follow. It's a leading indicator of business health, not a lagging report of what already happened.
Aligns the team
Anyone in the organization (engineering, design, marketing, support) can understand it and know how their work connects to it. It's a shared direction, not a dashboard for analysts.
The concept was popularized by Sean Ellis and later refined by Amplitude. The key insight is that companies with clear North Star Metrics make better decisions faster, because every team has a common reference point for "does this move us in the right direction?"
A North Star Metric is not a revenue target. Revenue is a lagging outcome of value delivered; it's a consequence of getting the NSM right, not the NSM itself. When teams optimize for revenue directly, they often sacrifice user value to hit short-term numbers, which destroys retention over time.
Why It Matters
The problems a North Star Metric solves
Without a North Star, teams default to optimizing for whatever metric is most visible, usually revenue, sign-ups, or engagement time. These create perverse incentives: a metric optimized in isolation rarely improves the product for users, and often actively degrades it.
Focus
A single shared metric cuts through backlog debates. "Will this move our North Star?" becomes the first filter for every initiative. Teams that don't have this question spend enormous energy justifying why their particular project matters.
Anti-vanity
A well-chosen NSM is hard to game without actually improving the product. "Total sign-ups" can be inflated with discounts or dark patterns. "Users who completed their first meaningful task" requires genuine onboarding improvement.
Cross-team alignment
When engineering, design, and marketing all understand the same North Star, they can self-organize around it without constant top-down coordination. The NSM becomes the implicit question in every team meeting: "Does this decision move the number?"
Long-term thinking
A good NSM predicts sustainable growth, not just this quarter's revenue. Teams that optimize for it are building a product users love, which produces compounding returns through retention and referral.
Criteria
What makes a good North Star Metric
Not all metrics can serve as a North Star. A good NSM must pass several tests.
01
It measures value delivery
It counts moments when users receive the core value your product promises. If your product helps people learn, the NSM should capture learning, not just sessions or time-on-site.
02
It predicts retention and revenue
When this metric goes up, revenue and retention should follow. You can test this historically: do users who score well on this metric churn less and spend more? If yes, it's a valid predictor.
03
It's not easily gamed
The metric should be hard to improve without genuinely improving the product. "Number of logins" is gameable with notifications. "Users who completed a meaningful workflow" requires the workflow to actually work.
04
Everyone understands it
A non-technical stakeholder should be able to understand what it measures and why it matters. If it requires a 10-minute explanation, it won't function as an organizational anchor.
Real Examples
North Star Metrics from real companies
The best way to understand the concept is to see how well-known companies have applied it. Notice how each NSM is specific to the value that company provides, not a generic metric.
Spotify
Time spent listening. Spotify's value is musical joy and discovery. When users spend more time listening, they're getting value. This metric predicts subscription renewal and premium conversion, and it can't improve without the music actually being good and discoverable.
Airbnb
Nights booked. Each booking represents a guest having a travel experience and a host earning income, both sides of the marketplace getting value. This metric is directly tied to revenue (Airbnb's take rate), making it unusually clean as a NSM.
Slack
Messages sent within a team's first 30 days. Slack's famous "2,000 messages" benchmark: teams that sent 2,000+ messages were almost never churned. This became the activation target that defined their onboarding, getting teams to the threshold of habit formation.
Facebook
Daily Active Users (DAU). Facebook's value is social connection. DAU captures whether that connection is happening daily, which predicts advertising revenue, time on platform, and competitive defensibility. Their famous "7 friends in 10 days" onboarding target drove DAU.
WhatsApp
Messages sent per day. Simple, direct, and hard to fake. Each message is a moment of communication value delivered. WhatsApp's NSM is also why they built features that increase messaging frequency (status updates, reactions) rather than features that increase sessions.
How to Choose Yours
Finding your North Star
The process of choosing a North Star Metric is itself valuable: it forces you to articulate what value your product actually delivers, and to verify whether you're measuring that delivery.
01
Define the core value exchange
Complete this sentence: "Users choose our product because it helps them ___." The blank is your value promise. Your NSM should measure whether that promise is being fulfilled, not just whether users showed up.
02
Find the "aha moment"
What's the moment a new user first experiences real value? For Slack, it's sending that first message and getting a reply. For a task manager, it's completing a task and feeling the satisfaction of clearing it. The NSM often lives near this moment.
03
Test against retention data
Look at your retained users vs churned users. What do retained users do in their first 30 days that churned users don't? The action that most predicts retention is often the NSM. This is data, not intuition.
04
Verify it can't be gamed
Before committing, ask: "How would someone increase this number without improving the product?" If there's an obvious path (push notifications, dark patterns, artificial friction), refine the metric until gaming it requires actually improving user experience.
Hierarchy of Metrics
NSM vs OKRs vs KPIs: how they fit together
These three concepts are often confused because they all involve tracking numbers. They operate at different levels and serve different purposes.
North Star Metric
What: One number that defines product success.
Who: The whole organization.
When: Tracked continuously, strategic horizon.
Changes: Rarely, maybe once in a product's lifetime.
The NSM is the destination. Everything else is navigation toward it.
OKRs & KPIs
OKRs: Quarterly goals teams set to move toward the NSM.
KPIs: Operational metrics teams track to monitor health.
Who: Teams and individuals.
When: Quarterly or monthly horizon.
OKRs are the routes. KPIs are the warning lights on the dashboard.
A team might set an OKR to "improve Day-7 retention from 28% to 38%." That OKR is in service of moving the North Star (say, "weekly active users who complete their primary workflow"). The KPIs they monitor along the way (activation rate, onboarding completion, first session depth) tell them whether they're on track. The NSM tells them whether the whole system is working.
A well-designed metrics hierarchy answers three questions at a glance: Are we delivering value to users? (NSM) Are we making progress this quarter? (OKRs) Is anything broken right now? (KPIs)