What is a heatmap?
A heatmap is a visualisation that overlays aggregate visitor behaviour onto a screenshot of a page, using colour to show intensity. The three common types are click maps, scroll maps and attention or movement maps. Each summarises thousands of sessions into a single image you can look at rather than read.
Compared
The three types of heatmap They are often bundled under one label and they answer quite different questions. Only one of them is reliable enough to make decisions from on its own.
| Type | Shows | Good for | Weakness |
|---|---|---|---|
| Click map | Where people clicked or tapped | Finding non-clickable things people try to click | Says nothing about what happened after the click |
| Scroll map | How far down the page people got | Deciding what is actually above the fold | Depth is not attention: people scroll past things |
| Attention or move map | Where the cursor lingered, or estimated attention | Directional read on what draws the eye | Cursor position correlates weakly with gaze, and not at all on mobile |
Mouse position is a poor proxy for eye position, and the correlation studies behind attention maps are weaker than the marketing suggests. Scroll maps are the most trustworthy of the three.
What heatmaps are genuinely good for
They are a fast way to spot a specific class of problem that aggregate numbers describe badly.
- People clicking things that are not links: a styled heading, an image, a stat card that looks interactive. This is the single most useful heatmap finding.
- Content below a scroll cliff: if 70% of visitors never reach your primary call to action, that is a layout decision made visible.
- Navigation nobody uses: a menu item with almost no clicks across thousands of sessions is a candidate for removal.
- Mobile versus desktop layouts diverging: the same page can have completely different click patterns on each.
What they are bad at is anything causal. A heatmap shows where attention went, never whether the page achieved anything. A beautifully engaged page that converts at 0.4% is still failing.
Live demo
See it on real data. The live demo is a fully populated dashboard on demo traffic, with no signup and no card.
Open the live demoWhere heatmaps mislead
Dynamic and personalised pages
Heatmaps overlay one screenshot. If your page varies by user, by test variant, or by viewport, the overlay averages layouts that never coexisted.
Sample size
A heatmap over 200 sessions looks just as convincing as one over 20,000. There is no confidence interval on a colour gradient, which makes them easy to over-read.
Segments blended together
New visitors and returning customers behave differently on the same page. A blended heatmap shows a pattern that describes neither group.
Screenshot drift
Ship a redesign and the historical heatmap is overlaid on a page that no longer exists. Most tools handle this badly.
mrkr does not offer heatmaps
This is worth stating plainly rather than hedging. mrkr has no click map, no scroll map and no attention map. If a heatmap is a requirement for your team, Hotjar and Microsoft Clarity both do them well, and Clarity is free.
What mrkr offers instead is the underlying behaviour in a form you can query rather than squint at:
- Engagement time per page, which is the closest honest proxy for attention.
- Session replay, which shows one real visit in full rather than an aggregate overlay.
- Funnels and exit rates, which say whether the page achieved anything, which no heatmap can.
These measurements answer different questions from a visual heatmap. Replay helps explain an individual visit; funnels and conversion events show whether visitors completed a meaningful action.
FAQ
Does mrkr have heatmaps?
No. mrkr does not offer click, scroll or attention heatmaps. It reports engagement time per page as numbers, and it has session replay for watching individual visits. For visual heatmaps specifically, Hotjar and Microsoft Clarity are the established options.
Are heatmaps accurate?
Click and scroll maps are accurate about what they measure. Attention and mouse-movement maps are weaker, because cursor position is only loosely related to where someone is looking, and on touch devices there is no cursor at all. Scroll maps are the most trustworthy of the three.
How many sessions do I need for a useful heatmap?
A few thousand sessions per page and per device type before the pattern is stable. Below that you are looking at the behaviour of a handful of people rendered in a way that implies statistical weight it does not have.
Start with one line of HTML.Your first visitors show up within a minute.