Take the 2-minute tour ×
User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It's 100% free, no registration required.

Heatmaps are one of the best tools we have to mesure the impact of a design, not just in UX terms but also in marketing/conversion terms.

I've found that most of the free or less-expensive solutions are based on tracking the position of the cursor instead of tracking the eyes of the viewer.

I can understand the obvious technology restraint tracking eyeballs implies, but the question is: Are cursor-generated heatmaps reliable? Or rather they make an impressive picture but don't faithufully represent the areas of real interest?

Side note: Here is a list of 10 free heatmap applications (Spanish site but English links). I'd appreciate mentions to any other cursor and eye-tracking software to generate heatmaps.

share|improve this question
5  
Somewhat related: I also just learned that seeing something is not at all the same as noticing it (meaning that eye-tracking data can be misleading). Source: amazon.com/Things-Designer-People-Voices-Matter/dp/0321767535 –  Phil Dec 21 '11 at 13:38
5  
They're 100% reliable. They show where the cursor has been. They do not indicate where the eyes have been. Clicks generally imply that a user looked at a location, but a cursor heatmap is not a substitute for an eye tracking heatmap. Interest is a third thing that can't easily be tracked by either heatmap. –  Ben Brocka Dec 21 '11 at 15:55
    
redant.com.au/blog/… .. short version: don't use on Ajax/interactive sites, or sites that rely on cookies or session variables –  Erics Feb 10 '12 at 4:50
add comment

5 Answers 5

up vote 9 down vote accepted

I think that they are reliable as what they are - cursor mapping heatmaps. They indicate where the user is positioning the cursor. In the same way, eye-tracking heatmaps indicate where the users eyes rest. Neither of them actually indicate what the user is doing when their eyes or the cursor rest there.

So I think that they are useful for some sorts of pages ( in particular, highly interactive ones ), to give a certain type of feedback as to what users do. They are not a replacement for eye-trackers, and neither is a replacement for talking to users. Each provides a different type of input, that requires interpretation.

share|improve this answer
add comment

According to this study which is referenced in this blog post by Clicktale (who happens to sell a mouse tracking tool) there is a an 84%-88% correlation between eye tracking and mouse tracking.

A more recent study from Microsoft and the University of Washington suggests the correlation depends a lot of what the user is doing. When the mouse cursor isn't moving, the average distance is 233px, but when they are doing an action, it is on average 77px.

share|improve this answer
    
But why was this answer down voted?? I'm the only one here who actually presented numbers instead of opinions. –  Tony Bolero Dec 22 '11 at 16:26
    
+1 it actually points to a study of sorts on the correlation, which was my question's underlying motive. Wording is important, and I guess I didn't do a good job explicitly asking about correlations. But of course voting is subjective, some may find your answer out of topic and some may see how it contributes to the subject. –  Naoise Golden Dec 22 '11 at 17:02
1  
Re-reading the question, a more fitting answer would be asserting that eye-tracking is more reliable than mouse-tracking, and that by pointing out the correlation between both, we can extract the reliability of the mouse-tracking techniques. –  Naoise Golden Dec 22 '11 at 17:06
2  
This study seems flawed. Look at the heatmap. It is focused on the text. Sure, a lot of people, but not all, will follow lines of text with their mouse to keep their place. But in a site that is not based on reading large sum of text, this has to be way off. Of course I could be biased myself, but I NEVER follow the path my eyes take on a site with my mouse. I only move my mouse when attempting a to take an action. –  Matt Rockwell Dec 22 '11 at 17:13
add comment

I think heatmaps based on cursor movements are misleading and I would guess the companies are are doing this to suggest that this is some proxy for eyetracking, it is not. As many people have mentioned the cursor is moved in many different ways, I have seen people use it as a reading aid and others that throw it to the side so they can see what they are reading.

In my opinion mouse movements and clicks should be displayed in the balls and lines display which also avoids the potential confusion between clicking and pausing your cursor. This is complexity is obfuscated when displayed as a heat map.

share|improve this answer
add comment

Random cursor movements are generally an indication that the user doesn't actually understand what to do on a page.

share|improve this answer
1  
Personally I find myself often positioning my cursor just to the right of the 'block' that I'm reading - especially noticeable somewhere like here at stack exchange. It may appear random, but isn't. –  Roger Attrill Dec 21 '11 at 15:12
    
You need to look at the 'whole picture' though - when you tie up what the user is saying + the expression on their face + the random movements around the screen - I'd deduce that the user doesn't know what to do. You can capture all of this on something like Morae. –  PhillipW Dec 21 '11 at 16:42
    
yeah - definitely a good idea to sync feedback from multiple channels if you can! –  Roger Attrill Dec 21 '11 at 17:26
1  
I'd generally agree with your statement, but all of the mouse-tracking heatmaps I've looked at are made from data collected from dozens or more users, not just a single user. And they also don't show the path of movement as all mouse positions are flattened into a single image. I think this behavior is more useful when analyzing playbacks of individual user tests rather than heatmaps. –  Lèse majesté Dec 22 '11 at 3:11
    
Yes that's a good point Lese. –  PhillipW Dec 22 '11 at 16:50
add comment

This is a question I've pondered myself. One major problem with heatmaps that I can think of is that I often move my cursor out of the way of what I want to look at, whether they be images, video, or text. So aside from when I'm actually moving the mouse to click on a link or button, I usually place my cursor in an empty area like the page margins to not block any content.

There's also little evidence that people will move their cursor towards non-clickable page elements which they're interested in or are focused on.

So it's most-likely only going to reveal what your most-often-clicked elements are, and maybe also highlight the typical path that the cursor travels to go between these elements. So if you have a CTA and a login button that are very popular with visitors, the heatmap might show a region between the two that is highly trafficked because it's the most direct route between two frequently clicked items. But that doesn't mean users are particularly drawn to the content in that region.

Maybe cursor-movement-generated heatmaps take these issues into account some how (I believe some marketing materials for mouse-tracking heatmap services do show side-by-side comparisons between eye-tracking-generated heatmaps and their own to show that their heatmaps are a reliable substitute). But if they don't, the results could be very misleading for people who try to interpret the heatmaps the same way as eye-tracking heatmaps.

share|improve this answer
1  
+1, I also move mine into an empty spot of the page –  Izkata Dec 21 '11 at 16:06
1  
I do the same. Having read this question twice now (and thus paid attention to my mouse), I seem to move my cursor to the left of the text around the middle of the browser while I scroll with the mouse wheel. In this type of scenario, a heat map would seem to be utterly useless. –  Karen Dec 21 '11 at 19:43
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.