Metrics
Metrics are specific, measurable indicators that teams use to evaluate if their product changes are having the intended impact.
A testable hypothesis includes a specific, measurable impact—the metric—that makes it crystal clear at the end of a test whether the hypothesis passed or failed.
Why does specificity matter for metrics?
It's easy to be sloppy with metrics. Vague impacts like "improve user experience" or "drive engagement" aren't testable without defining the specific metric you'll measure.
How will you measure improvements? You need to be specific. Are you measuring conversion rate, time on page, feature adoption, or something else?
How do you choose the right data to measure?
Teams need to be clear about what they're measuring and think through what the data needs to look like before starting experiments.
Do you need to collect the number of actions taken or the number of people who took action? Are you tracking visits, sessions, or page views? Ask yourself: What would the data need to look like for me to refute this hypothesis? What would it need to look like to support it?
Understanding what you're measuring and how you'll make decisions with that data is essential for running good experiments.
Learn more:
- Shifting from Outputs to Outcomes: Why It Matters and How to Get Started
Related terms:
- Hypothesis
- Desired Outcome
- Assumption Testing
- Experiments
Last Updated: October 25, 2025