HEART Framework: Five Metrics for Product Health

WHY IT EXISTS: Product teams naturally gravitate toward easy-to-count numbers like page views, downloads, or raw signups because they move up and to the right. The problem is that these vanity metrics often hide terrible user experiences. You can have record traffic and a product that nobody wants to use twice. The HEART framework was created to give teams a structured way to choose metrics that actually reflect user value and product quality across the entire journey, not just surface-level activity.

THE MENTAL MODEL: Think of HEART as the five vital signs on a hospital monitor. A doctor does not treat a patient based on heart rate alone; they look at blood pressure, oxygen, and temperature together to spot the real problem. Similarly, HEART gives you five lenses to diagnose product health. A spike in new users means nothing if those users hate the experience and never return. The framework forces you to look at the whole patient.

HOW IT WORKS: HEART stands for five dimensions you measure. Happiness captures user sentiment through surveys or net promoter scores. Engagement tracks how deeply people use the product, such as sessions per week or actions per visit. Adoption measures new users reaching a key milestone, like completing a first project or enabling a feature. Retention is the rate at which existing users come back over a defined period. Task Success covers usability, like completion rates, error rates, or time on task. In practice, teams pair HEART with the Goals-Signals-Metrics method. You start with a goal, identify signals that indicate success or failure, and only then choose the specific metric to track.

WHEN TO USE IT: Use HEART when you need to define success for a feature launch, a redesign, or an overall product health dashboard. It shines when cross-functional teams argue about what matters because it forces explicit tradeoffs. If the business cares about growth but the user team cares about quality, HEART gives both sides a shared language.

WHEN NOT TO USE IT: Do not use HEART for pure infrastructure projects with no user interface, or during very early validation when you have ten users and qualitative feedback is richer than any metric. It is also the wrong tool when you need surgical focus on a single conversion bottleneck; adding four extra dimensions just dilutes attention.

ONE CANONICAL EXAMPLE: Imagine a team launching a new collaboration feature in a design tool. They might track Happiness through a post-use satisfaction survey, Engagement through comments left per week, Adoption through the percentage of teams that invite a colleague within seven days, Retention through the share of users who return to the feature within thirty days, and Task Success through the percentage of invited colleagues who accept and join without support tickets. If Adoption is high but Retention is low, the team knows the onboarding is catchy but the feature lacks staying power.