Chewing through a large dataset of numbers is like eating stale cereal – boringly unpleasant. Yet, when evaluating the User Experience of websites or web apps we do not only focus on numbers, since there is other data and insights that usually needs analysis.
The sheer volume of this task is in itself a problem. To aid us, various forms of data visualisation have came forth. Line graphs, diagrams … these are the bread and butter of visualising data in a digestible form. However, they do not fill all problematic holes.
There are two problems:
- How can we correctly understand data?
- How can we be sure that our research is valid?
You can go and grab your handbook of user testing and verify all procedures, but that takes time … and money … and many, many other factors you need to account for. There are marks on our heads from scratching while figuring out how to minimize research costs and time.
Data visualisation has grown and created a form of tracking data that displays the actual behaviour of users in their natural browsing environments. Mouse tracking, a sub form of visual analytics, tracks mouse movements and clicks – data that you can use in order to verify the usefulness and ease of use of your web app or webpage with low effort.
Mouse Tracking – How Valid Is It?
I am quite sure that most of you are saying ‘This is nothing new.’
Your reaction only proves how difficult it is to incorporate mouse tracking in your research properly. Many have tried and failed.
Eye tracking is considered a rough equivalent of mouse tracking. Eye tracking studies pose three valid problems: Cost, Limited Scope and the High Observer Effect. Indeed eye tracking studies are expensive, restricted to small sample sizes of users and technology, and subject to the Hawthorne effect, which does tend to mess up a usability study.
That is why it makes more sense to rely on mouse tracking. It is cheaper and is a passive form of study, that is, users have no idea they are being tracked. In this way, their behaviour is natural and in context of the technologies they would normally use.
These two factors are crucial in web app or webpage optimization.
Is mouse tracking that effective, though? There is a paper rolling through the web citing that there is an 84% – 88% correlation between eye movement and mouse movement. That is why most researchers are eager to substitute one for the other. However, one recent paper from Google claims that eye movement is only 32% similar to mouse movement. Other sources are also nearing in the statement that it works half of the time.
Does it nullify the validity of mouse tracking? No. Eye tracking in the same way as tracking the movement of a mouse pointer is useful in determining the areas of interest, but eyes cannot click – and a mouse does. Why is it important? It is because a deliberate click on an element is a form of showing interest.
Are We Limited To Heat Maps?
The previous paragraph was a lead-in to get you, dear reader, thinking about heat maps – yet another form of data visualisation that is well-known in the UX field.
Who could resist their charm? They are simple, quick to understand and show the density of clicks in given areas. However, I have seen many evaluators taking a heat map and looking at it as a separate picture, a picture that is separate from the environment of the whole website or web app.
‘Look at this, this button gets clicks! WE NEED MORE BUTTONS!’
Here is a small, hypothetical case study. A website introduced a fresh call to action to their front page. It was meant to invite users to set up a free trial. It read ‘Go To The Tool‘. Now imagine what a heat map would show … a scalding hot button. The marketers would already be patting their backs on a job well done. Months later, the website exits after clicking the button have skyrocketed. Why? Users were interested in seeing the tool, but they were not interested in registering. The button convinced them that they would see the tool, not find a form to fill out (before seeing it). A heat map has shown interest, but it did not show that there was sheer confusion of users.
Do heat maps suck, then? No, again no. They are a nifty trick up your sleeve, especially if you need to show something to your stakeholders. Alternatively, they are very useful for pointing you in the right direction if you are wondering what needs to be studied in detail. However, heat maps on their own are not enough to work on improving the UX of your webpage or web app.
What Tool Shows All User Actions?
It’s pretty simple – why should you only track clicks? Track everything. Track swipes, clicks, movements, highlights, interactions, browsing paths. Track all the interactions that a user makes when browsing your website or web app.
You can record them and watch them later, witnessing the entire user journey of a given person. No need to rely on any guesswork – actions seen first-hand will provide you insights that you can act upon immediately. This form of tracking is called Session Recording.
The experiences will be eye opening – I had a client who had a single button, which once removed increased conversions by 40%.
There are several tools that will help you record actions:
- ClickTale – an enterprise option for recording sessions. Expansive, large, and features a suite of other advanced tools. The only con is perhaps its pricing.
- VWO – a tool known as an A/B testing software that also offers session recording for variants. You can record actions on your variants and determine with the help of a recording which version is the champion and why.
- UsabilityTools – a tool capable of reproducing the user screen and actions in one hundred percent faithfulness. Offers a free trial for two weeks and has a cheap monthly subscription.
If you need a fast, informative and easy-to-set-up data visualisation tool for your UX projects, Session Recording is the best source of qualitative data that you can use during designing and during meetings to back up your observations.
Consider it as an expanded form of mouse tracking that will satisfy the hunger for the good design ideas that every UX designer has.
(Lead image source: markusspiske – CC0 Public Domain)