HEART maps happiness, engagement, adoption, retention, and task success onto goals and signals

What it is

HEART takes Happiness, Engagement, Adoption, Retention, and Task success and hooks each category to the three layers of Goals–Signals–Metrics: write a goal for that category, write a signal that would be visible if the goal were met, and only then name a metric. The contribution is not another list of five nouns. It is forcing every category through the mapping “goal → signal → metric,” so teams cannot jump from a category name to a ready-made number. A happiness goal might be “authors feel the tool is helping them finish the draft”; the signal is showing the finished piece to a colleague unprompted; only then comes an attitude item or a count of recommending behavior.

Why it happens

Category names are too easily treated as metric names. Engagement is filled with session length; Happiness is filled with a satisfaction item. The mapping layer breaks that short circuit: the engagement goal on this product might be “several complete authoring sessions each week,” the signal is a session that ends in export or publish, and duration is only one candidate metric, which may lose. HEART and GSM are bound together because the five categories supply horizontal coverage and GSM supplies vertical derivation; without the mapping, horizontal coverage decays into five unrelated common fields. The mapping also exposes blanks: if a category cannot produce an observable signal, that category is not yet an operable measurement problem on this product, rather than a missing industry-standard number.

Studying it

Demand three lines per HEART category: a goal sentence, a signal sentence, a metric definition; missing any line means the mapping is unfinished. Ask colleagues outside the plan to read only goal and signal and judge whether the proposed metric captures that signal; record mismatches. Compare a “fill each category with a common metric” plan against a “map, then choose” plan by later citation in experiments: columns never cited in decisions are often the ones filled from the warehouse without mapping. When one category has two different metrics, adjudicate at the signal layer, not by which number looks better.

Where it stops holding

HEART is a thinking framework for large-scale product measurement, not a psychometric scale: the five categories have no standard items and no norms. Adoption on internal tools is often an entitlement being switched on, which is not the same mapping as consumer feature uptake. Do not fold the five initials into a composite index; that undoes the split the mapping just made. The framework can complete the mapping for only some categories; empty ones should be marked unmapped, not filled with placeholder numbers.

Applying it

• Fill a five-by-three table of HEART × GSM; do not fill the metric column first and then write goals and signals backwards.
• Every cell must be readable as “what we need to see is …, which is why we count ….”
• Leave categories with no signal blank, and write why; do not top them up with industry-common fields.
• Review only categories whose mapping is complete; unmapped categories do not enter the ship gate.

Related

• Same group: Q6.07.2 Frameworks name dimensions differently, but all require goals before metrics · Q6.07.3 Applying every framework dimension without product stage dilutes measurement focus · Q6.07.4 A framework supplies a thinking structure; specific metrics still need product-level customization
• Adjacent: Q6.01 Classification frameworks · Q6.02 Goals–Signals–Metrics
• Search terms: HEART · Goals-Signals-Metrics · GSM mapping