Reading the Quiet UX Signals in Your Analytics

Most visitors never leave feedback. Their behaviour still tells a story, if you read the numbers as clues to investigate rather than verdicts.

Signals are questions, not answers

Analytics show what people did, never why. A page with a short visit may have failed, or it may have answered the question in ten seconds. Treat every number as a prompt to look closer, ideally by combining it with what people say in feedback or testing.

Bounce rate

Bounce rate is the percentage of visitors who enter a site and leave rather than continuing to view other pages on the same site. A high bounce rate on a contact page with a phone number may be perfectly healthy. A high rate on a page meant to lead somewhere else is worth a look: is the next step visible on a phone screen, without scrolling?

Heat maps

A heat map represents values as colours. Applied to a web page, it shows where people click, move or scroll. Look for:

• Clicks on things that are not links, such as images or bold headings. People expect them to do something.
• A sharp drop in scrolling just above important content, which suggests a "false bottom" where the page looks finished.
• Ignored navigation items, which may be badly named rather than unwanted.

Session replay

Session replay reconstructs a visitor's interaction with a site, letting you watch the clicks and scrolling as they happened. Repeated rapid clicks on one spot, frantic back-and-forth between two pages, or a form abandoned at the same field all show friction clearly. Replay is powerful, so use it with care: mask form fields, avoid recording sensitive pages, and tell visitors in your privacy notice.

Search logs

What people type into your site search is feedback in their own words. Searches that return nothing, and searches for things that are already in the main menu, both point to labels that do not match the way visitors think.

Form analytics

Forms are where many journeys end, for good or ill. If your tools can show which field people leave a form on, look for the one that loses the most. Typical culprits are fields that ask for information people do not have to hand, unclear password rules, and error messages that appear only after submission and clear the rest of the form.

Compare before and after

Signals become most useful around a change. Note the date you changed something, and look at the same measure for comparable periods before and after. Seasonal swings and campaigns can move numbers on their own, so be slow to credit or blame a single change.

Pair every interesting number with one session replay or one user quote before you act on it.

By Iris Calloway · last checked 2026-11-09