AARRR vs HEART

At a glance

Two different questions about the same users

AARRR is written from the company's side of the table. Each stage is a step in a revenue story — a visitor becomes a user, a user becomes a regular, a regular brings someone else and eventually pays. A stage is weak when that story stalls, whatever the reason. The model does not ask whether people enjoy the product; it asks whether they come back and whether the business can count on it.

HEART is written from the user's side. It was designed to give UX teams something better than page views to argue with, and it comes with a small process, Goals-Signals-Metrics: state what the product should achieve for a user, find the behaviour that would show it, then pick the number. It is usually applied to one product or one feature at a time — a new editor, a checkout flow — rather than to the whole company. That is its strength: it can tell apart two versions of a screen that AARRR would score exactly the same.

Where they overlap: retention, measured two ways

Retention is the one stage the two name with the same word, and it is the source of most of the confusion. In AARRR it means customers who keep using the product and keep paying for it, read as cohort curves over months. In HEART it usually means users who come back to a specific product or feature, read over days or weeks. A team can have a feature with excellent HEART retention inside a business with poor AARRR retention: the people who use the new report love it, and most accounts still churn before they ever find it.

Adoption and Activation look alike too, and differ in the same way. HEART's Adoption counts new users of a feature; AARRR's Activation asks whether a new customer reached the moment the product becomes worth keeping. The first is a count, the second is a threshold, and a launch can move the count without moving the threshold at all. When the two frameworks disagree on the same word, the disagreement is information: it usually means the feature and the business are not being measured on the same population.

The funnel picks the stage, HEART inspects it

Used together, the two nest rather than compete. AARRR works at the level of the business and answers where growth stalls. Once it has named a stage, say Activation, HEART is a good way to look inside it: which task in the first session fails, how long it takes, whether the people who finish it are happier than those who give up. Task success and Happiness in particular have no equivalent in AARRR, and they are exactly what a team needs once it knows which stage to fix.

The reverse order wastes effort. A HEART scorecard for every feature, built before anyone knows which stage of the business is weak, produces a lot of honest measurement about screens that were never the problem. The questionnaire on this site does the first half of the job: it names the stage in three minutes, from fifteen questions about practices rather than about screens. What it cannot do is watch your users — that is the part HEART is built for.

Which to use, and when

The terms behind this

The other comparisons