UX & Research Framework · 9 min read
AARRR measures growth. OKRs measure business outcomes. But neither answers the question: is the product actually easy, pleasant, and effective to use? Poor UX is the silent killer of retention — users who find your product confusing or frustrating rarely tell you directly. They just churn. Google's HEART framework, developed by Kerry Rodden's UX research team, gives product and design teams a structured way to measure UX quality with real metrics — not just NPS or CSAT scores. It works at the scale of a feature, a flow, or an entire product.
Happiness captures the attitudinal dimension — not what users do, but how they feel. NPS, CSAT, app store ratings, and in-product satisfaction surveys all measure Happiness. The challenge: Happiness is a lagging indicator. Users who are about to churn often score Happiness fine because they haven't yet articulated their dissatisfaction. Pair Happiness metrics with behavioural metrics (Engagement, Retention) to get the full picture.
Engagement measures how deeply users interact with the product — frequency, depth, and breadth of use. DAU/MAU ratio is the classic engagement metric, but HEART pushes you to measure engagement at the feature level too. An app might have strong overall DAU but terrible engagement on a new feature you just shipped. Engagement tells you whether the product is becoming a habit or staying a novelty.
Adoption measures how many users start using a product or feature for the first time. For a new product, this is activation rate. For an existing product shipping a new feature, this is feature adoption rate (% of eligible users who try it in the first 30 days). Adoption answers the design question: "Is the entry point to this feature discoverable and compelling enough to get users to try it?"
Retention in the HEART model is the same retention you track in your AARRR funnel — Day 1, Day 7, Day 30 cohort retention — but applied at the UX level. A design change might improve Adoption (more users try the feature) while hurting Retention (they try it once and never return). Tracking both simultaneously tells you whether a UX change creates genuine value or just novelty.
Task Success is the most directly actionable HEART dimension. It measures whether users are able to complete specific tasks efficiently and accurately. Completion rate, error rate, and time-on-task are classic usability metrics that belong here. Task Success is particularly important for complex flows — KYC, loan applications, mutual fund investments, BNPL checkout — where a single confusing step can abort the entire transaction.
HEART doesn't give you a fixed set of metrics — it gives you a process for deciding which metrics to track for your specific product and context. That process is Goals → Signals → Metrics (GSM): for each HEART dimension, define a goal (what success looks like), identify signals (observable behaviours that indicate that goal is being met), then translate signals into measurable metrics.
What does success look like for this dimension?
What user behaviour would we observe if the goal is being met?
How do we measure the signal at scale?
Here's how a payments app would complete a HEART worksheet for their onboarding flow — from app download to first successful transaction.
| Dimension | Goal | Signal | Metric | Current | Target |
|---|---|---|---|---|---|
| Happiness | Users feel the signup was quick and painless | High CSAT score at end of onboarding · Positive app store reviews mentioning "easy signup" | Post-onboarding CSAT survey (shown after first transaction) | 73% | 82%+ |
| Engagement | Activated users return and transact regularly in week 1 | User opens app more than once in first 7 days · Makes at least 2 transactions in first week | % of activated users with ≥2 sessions in D7 · Transactions per activated user in D7 | 38% | 55%+ |
| Adoption | New users complete KYC and link their bank account | User reaches "KYC complete" state · Bank account linked before leaving app | KYC completion rate (% of new signups who complete full KYC) | 51% | 65%+ |
| Retention | Activated users are still transacting 30 days later | User opens app and makes at least one transaction in days 25–35 | D30 transacting retention rate | 28% | 38%+ |
| Task Success | Users complete KYC without dropping off or restarting steps | No step in KYC flow retried more than once · KYC completed in single session | Single-session KYC completion rate · Step-level retry rate per KYC screen | 62% | 78%+ |
AARRR is a funnel framework — it tells you at which stage of the user journey you're losing people. HEART is a quality framework — it tells you whether the design of specific flows is working well for users. Use AARRR to diagnose where the problem is in your funnel. Use HEART to diagnose why the UX is causing that problem and what to fix. In practice, teams run AARRR monthly at the product level and run HEART worksheets when designing or evaluating specific features or flows.
HEART is especially valuable when your team is debating whether a UX change made things better or worse after shipping — it forces you to measure all five dimensions, not just the one that looked good in the A/B test.
We run HEART measurement workshops for product and design teams — building the GSM worksheet, identifying the right instrumentation, and setting targets. Book a free session.
Book Free Session