Introduction
Analytics can tell you a thousand people visited the page; a heatmap shows you what they did once they arrived, rendered as weather: warm zones of concentrated clicking, cool stretches nobody touched, the exact scroll depth where attention fell off a cliff. Heatmaps are the most immediately legible visualisation in behavioural research, which is both their power and their trap, because legible is not the same as unambiguous. This article covers the main heatmap types, what each can honestly claim, and how to read them without fooling yourself.
What are Heatmaps?
A heatmap is an aggregated visualisation of user behaviour on a page or screen, using colour intensity to show where activity concentrated. In product research the term covers several distinct instruments. Click (or tap) maps aggregate where users clicked, including on things that aren't clickable. Scroll maps show how far down the page visitors travelled, colour-banding the drop-off. Move maps aggregate mouse position. And attention or gaze maps, built from eye-tracking hardware, show where people actually looked, a different and more expensive claim than where their cursor drifted. Tools like Hotjar generate the first three passively from live traffic at scale, which is why heatmaps are often a team's first contact with behavioural data.
What Each Map Can Claim
Click maps are the most trustworthy of the passive family: a click is a deliberate act, and clusters of clicks on non-interactive elements (the underlined heading, the static product image) are among the clearest design feedback available, users voting for affordances that don't exist. Scroll maps are similarly honest about one thing: content below the fade line was never seen, whatever its quality. Move maps require the most caution. The folk assumption that cursor position tracks gaze has only weak and inconsistent support in the research literature; many people park the mouse while reading elsewhere or run it along text as a pacer. Treat move maps as suggestive texture, click and scroll maps as evidence, and reserve claims about attention for actual gaze data.
Reading Heatmaps Well
1. Segment before interpreting.
An aggregate map blends new visitors with returning ones, mobile with desktop, buyers with bouncers, and the blend can show a pattern no actual segment exhibits. Split by device and audience first; mobile and desktop maps of the same page are routinely different documents.
2. Check the sample behind the colours.
Heatmaps render twelve visitors as confidently as twelve thousand. Find the n before believing the weather, and be doubly cautious comparing maps across pages with very different traffic.
3. Mind the page-variant problem.
Dynamic content, personalisation, and responsive layouts mean "the page" may be many pages; a map averaged across variants can place clicks on elements some visitors never saw. Map variants separately where the tool allows.
4. Hunt the classics.
The recurring finds: rage-click clusters (rapid repeated clicking that signals frustration or breakage), dead clicks on non-interactive elements, primary calls-to-action sitting below the effective fold, and navigation items nobody touches. Each is a hypothesis, located.
5. Follow the map with a why.
A heatmap shows where behaviour concentrated, never what anyone wanted. The cold CTA might be unnoticed, mistrusted, or irrelevant, three different fixes. Session recordings narrow it; a quick task-based study with follow-up questions (a Ballpark test against the live page or a revised prototype) settles it. The map finds the crime scene; other methods interview the witnesses.
The Benefits
Heatmaps collect themselves from real traffic at full scale, no recruitment required, and compress thousands of sessions into one glance. They excel at locating problems (where interest dies, where phantom affordances live, where the fold actually falls) and at communicating them: no research artefact converts stakeholders faster than their own homepage rendered mostly cold. As a triage layer that directs deeper research, they are excellent value.
The Limitations
Heatmaps describe aggregate surface behaviour on a single page with no context of intent, task, or outcome; they cannot distinguish fascination from confusion. Move maps overpromise on attention, aggregates hide segments, low-traffic maps masquerade as evidence, and dynamic pages muddy what was actually seen. They also tempt teams into page-level optimisation while journey-level problems (why visitors arrived with the wrong expectations at all) go unexamined. They are a locating instrument, not an explaining one.
The Takeaway
Heatmaps are behavioural research's establishing shot: fast, cheap, and vivid about where the action is and isn't. Read them segmented, sample-checked, and variant-aware; trust clicks and scrolls more than cursor drift; and treat every cold zone and rage-cluster as a located question whose answer lives in methods that can ask why.
Further reading
For the tooling and the reading patterns behind the maps:
Articles:
1. Heatmaps: The Complete Guide - Hotjar
A thorough practitioner's guide to click, scroll, and move maps from the tool most teams meet first, including setup and interpretation guidance.
2. F-Shaped Pattern of Reading on the Web - Nielsen Norman Group
The gaze-based research that explains many of the patterns click and scroll maps surface, and the formatting lessons that follow.