Back to Glossary
Analytics
Synonyms: product analytics, web analytics, behavioral data, usage data, quantitative data
Do not index
Definition
Analytics is the practice of collecting and reading data on how people actually use a product: clicks, pageviews, conversion, where they drop off.
It shows what's happening at scale, across everyone, not just the handful of people in a research session.
Use cases
Design on opinion alone and you'll argue in circles. Analytics shows where users actually struggle.
- The bleeding funnel: Conversion looks fine overall, but the funnel shows step 3 of checkout losing 40% of users. Now you know exactly where to dig instead of guessing.
- The regression you'd have missed: A redesign "feels cleaner," but analytics shows sign-ups dropped after launch. Without the numbers, you'd have shipped a downgrade and called it progress.
How it's used in practice
- Define events before building: Decide which actions matter and tag them up front. Untracked features produce no data, and you can't add it retroactively.
- Watch funnels, not vanity metrics: Follow drop-off and completion, not total pageviews. Big numbers feel good and decide nothing.
- Pair quant with qual: Use analytics to find where users struggle, then session replay or interviews to learn why.
- Segment everything: Split new vs. returning, mobile vs. desktop, paid vs. organic. Averages blur the exact problem you're hunting.
Analytics tells you what happened, never why. A drop-off spike on step 3 is a question. Teams that act on numbers without watching a single session replay or talking to a user usually fix the wrong thing.
Pair the "what" with the "why" every time, and ignore vanity metrics that aren't tied to a real user goal.
Challenges & limitations
- Bad instrumentation, bad data: Missing or wrongly fired events produce numbers that look real and aren't. Garbage in, confident garbage out.
- Privacy gaps: Consent rules, cookie limits, and tracking blockers mean your data is partial. You're often analyzing a sample, not everyone.
Commonly used tools
- Google Analytics (GA4) — the most common web analytics platform. Best for traffic, acquisition, and basic funnels.
- Amplitude — product analytics built around user behavior and events. Best for digging into in-app funnels and retention.
- Hotjar — heatmaps and session replay. Best for seeing the "why" behind a number.
Free resources
- NN/g: Analytics & Metrics — a deep library on using quantitative data in UX work.
- NN/g: Quantitative User Research Study Guide — a curated path through analytics, metrics, and quant methods.
- NN/g: Benchmarking UX — how to track UX metrics over time, redesign to redesign.
