North Star Metric: Find the One Number Your Whole Team Can Build Toward
Duolingo’s daily active users grew from 26.9 million in Q4 2023 to 50.5 million by Q3 2025, a 36% year-over-year increase. That growth didn’t come from chasing revenue targets or download counts. It came from a company-wide commitment to one metric: DAU.
That metric is Duolingo’s north star, and it shapes everything from streak notifications to lesson difficulty calibration. The north star metric is one of the most powerful (and most misunderstood) tools in a product manager’s toolkit.
Where the concept came from
Sean Ellis coined the term “north star metric” around 2010, defining it as “the single metric that best captures the core value a product delivers to its customers.” John Cutler, then a product evangelist at Amplitude, expanded the idea into a full framework with input metrics, game types, and organizational alignment practices.
The core premise is simple. One metric unifies every team around a shared definition of progress. Engineering, design, marketing, and sales all point at the same scoreboard. When it works, it eliminates political arguments about whose priorities matter more.
What a north star metric actually measures
A good north star metric sits at the intersection of three things: the value customers receive, the engagement that predicts retention, and the revenue model that sustains the business. It is not a vanity metric like total signups or page views. It is a leading indicator of long-term health.
Amplitude’s framework categorizes businesses into three “game types” to narrow the right metric:
Attention games (media, social, content platforms) compete for time. The metric tends to measure engagement depth. Spotify tracks total time spent listening. Netflix focuses on hours watched per subscriber.
Transaction games (e-commerce, marketplaces, fintech) compete on conversion and frequency. The metric tends to measure completed actions. Airbnb tracks nights booked. Uber counts completed trips.
Productivity games (SaaS, tools, enterprise software) compete on workflow adoption. The metric tends to measure active usage or task completion. Slack tracks the number of paid teams with active daily users. Notion measures weekly active editors.
The game type matters because it prevents a common mistake: picking a metric that sounds important but doesn’t connect to how your product creates value. A marketplace tracking monthly active users instead of completed transactions is measuring foot traffic, not commerce.
The companies that got it right
The strongest north star examples share a pattern: the metric directly reflects customer value, and improving it also improves the business.
Duolingo: Daily Active Users. CEO Luis von Ahn has repeatedly pointed to DAU as the company’s north star because it captures both learning engagement and subscription conversion potential. In Q1 2025, Duolingo reported 46.6 million DAUs (a 49% year-over-year increase) and $230.7 million in revenue. Paid subscriber penetration reached 9% of monthly active users by Q3 2025, when DAUs hit 50.5 million. The connection between daily engagement and monetization is direct: users who open the app daily are dramatically more likely to subscribe.
Spotify: Total Time Spent Listening. Spotify realized early that the real measure of success is not subscriber count alone. It is how deeply users engage. With 678 million monthly active users and 268 million premium subscribers as of Q1 2025, Spotify uses listening time to drive decisions about playlist personalization, podcast investment, and audio quality features. The metric captures value that MAU alone would miss: two users who both open the app but listen for 10 minutes versus 90 minutes are having fundamentally different experiences.
Airbnb: Nights Booked. This metric captures value delivery for both sides of the marketplace. A booked night means a guest found a place and a host earned income. It is more actionable than gross merchandise volume because it reflects the core experience rather than price fluctuation.
The layer most teams skip: input metrics
Here is where teams commonly fail. They define a north star, paste it on a dashboard, and expect behavior to change. It does not.
The north star is an output metric, a lagging scoreboard. Teams cannot directly move “nights booked” or “time spent listening.” They move the inputs that feed those outputs.
Brian Balfour, co-founder of Reforge, argues that the real power of the framework lies not in the north star itself but in the constellation of input metrics underneath it. He recommends breaking the north star into three dimensions:
- Retention breadth (how many users keep coming back)
- Engagement depth (how much value active users extract)
- Monetization efficiency (how effectively engagement converts to revenue)
Each dimension has its own input metrics. For Spotify, retention inputs might include “percentage of users who create a personal playlist within 7 days.” Engagement inputs might track “average listening sessions per week.” Monetization inputs could measure “free-to-premium conversion rate by cohort.”
These input metrics change frequently. The north star should not. That distinction is the entire point of the framework.
When the north star misleads
The criticism of north star metrics has sharpened considerably since the concept’s peak evangelism around 2019. Three failure modes have emerged from real companies.
False rigor at Pinterest. Casey Winters, then Pinterest’s growth lead, documented how the company combined two user actions (repinning and clicking) into a single metric called WARC (weekly active repinner or clicker). The combined metric created a dangerous blind spot: experiments could increase WARC by trading one action for another without anyone noticing. Winters called it “false rigor,” where a single number created the illusion of clarity while masking a real tradeoff.
Growth masking product decline. Jay Stansell traced a common failure arc in a Product Coalition analysis: growth metrics climb while engagement metrics quietly fall. As product leader Tiziano Nessi documented, “Growth was going up. Engagement was going down. The North Star was shining and the product was quietly dying underneath it.” By the time the engagement decline surfaces in the north star, the damage is structural.
Supply-side blindness in marketplaces. Pinterest’s WARC also ignored content creators entirely. No team wanted to invest in increasing unique content when they could focus on recycling existing content that drove clicks. The north star incentivized demand-side optimization while the supply side atrophied.
The lesson across all three: a north star metric works only when paired with guardrail metrics that catch the tradeoffs the primary metric cannot see.
Choosing yours: a practical sequence
Across fractional COO engagements, I have watched dozens of product and engineering teams attempt to define their north star. The ones that succeed follow a consistent sequence.
Start with the value exchange. What does a customer receive when your product works well? Not what you sell, not your feature list, but what outcome the customer would describe if asked “what does this product do for you?” That answer usually contains your metric.
Match it to your game type. If you run an attention game, your metric should capture depth of engagement. If you run a transaction game, it should capture completed exchanges. For a productivity game, focus on active usage or workflow completion.
Validate that it predicts revenue. Plot your candidate metric against revenue over 12 or more months of historical data. A genuine north star will correlate strongly (not perfectly) with business outcomes. If it does not, you have found a vanity metric.
Build the input tree. Identify three to five inputs that directly influence the north star. Assign each to a team. This is where the framework becomes operational. Without the input tree, the metric stays on a dashboard and nobody changes their behavior.
Set guardrails. Pick two or three metrics that must not decline while you optimize the north star. For Duolingo, guardrails might include lesson completion rate and new user activation. For Spotify, they might track creator uploads and content diversity scores.
Revisit annually, not quarterly. The north star should be stable across multiple strategy cycles. If you change it every quarter, you never actually aligned around it. Input metrics rotate regularly. The north star does not.
Where the framework fits in your toolkit
The north star metric works best as a complement to other strategic frameworks, not a replacement. OKRs provide the quarterly execution layer that translates north star inputs into sprint-level goals. Product roadmaps organize the work sequence. Discovery practices validate that the inputs you chose actually move the needle. A regular metrics review keeps the input tree honest.
The north star sits above all of those. It answers one question that every other framework assumes you have already answered: what does winning look like for this product?
If your team cannot answer that with a single metric and a clear connection to customer value, the framework still has work to do.