What is eye tracking?

Eye tracking is a process in which people’s line of sight and eye movements are recorded when they are in front of displays and products, or when they are in different types of settings. Classic uses include analysing reading habits on the web, eval­u­at­ing in­struc­tion manual use and observing how users navigate online marketing.

User in­ter­faces and packaging can also be re­searched to see how com­mu­nic­a­tion reaches users. Marketing strategists can gain important knowledge from eye tracking when people enter a store or walk through a su­per­mar­ket.

By analysing the data, analysts can find out which areas within a person’s field of view have attracted the most attention. The chro­no­lo­gic­al sequence in which in­form­a­tion is viewed is also important and can be re­searched. Eye tracking is a useful tool, which in the classic marketing mix can help both optimise the placement of products as well as a company’s com­mu­nic­a­tion method.

To carry out research into eye tracking, spe­cial­ised hardware and software are needed. The hardware includes cameras, which can record the face or eye movements of a person. Using image re­cog­ni­tion from specially designed software, the position of the pupils is detected, and their movement is then followed.

Pro­fes­sion­al hardware works in the infrared (IR) range, making it easier to detect irises and pupils. To do this, the area being recorded is il­lu­min­ated with an infrared light source. Depending on which way someone is looking, the re­flec­tions from the IR light source will be on a par­tic­u­lar place in the eye. The re­flec­tions from the IR source are then captured by the camera. The software can then draw con­clu­sions relative to the line of sight from the re­flec­tions in the pupils.

Eye tracking is used to optimise the user ex­per­i­ence on a visual level. User ex­per­i­ence design (UX design) includes the analysis, man­age­ment and op­tim­isa­tion of the user ex­per­i­ence. Some tra­di­tion­al uses for eye tracking include:

In public spaces:

• Signs in a large building
• Des­tin­a­tion boards

The design of control elements:

• Man­age­ment of control elements in a ticket machine
• Layout of control elements in a cockpit
• Layout of the user interface in an app

In ad­vert­ising and marketing:

• Design for new packaging
• Design and set up of a su­per­mar­ket shelf
• The placement of photos, texts and images in an online advert

In web design

• The layout design of a website’s nav­ig­a­tion
• The placement of con­ver­sion elements in online shops

After re­devel­op­ing their website, the operator sees a doubling of their bounce rate. Something seems to be putting off visitors, keeping them from com­plet­ing their purchases. The operator decides to use eye tracking to monitor 15 test par­ti­cipants as they set out to complete a test purchase. After com­plet­ing the user test, the operator realises that the button which is meant to take them to the checkout page can only be found after searching the screen for a con­sid­er­able amount of time. For many visitors, this search was enough for them to not to complete their purchases. Small ad­apt­a­tions to the layout and the design lead to a big decrease in the bounce rate and to more purchases being carried out.

For eye tracking, you’ll need both hardware and software. Some companies spe­cial­ise in one or the other, while others offer a complete solution.

The following hardware can be used in order to conduct research with eye tracking:

• Eye tracking glasses are worn by testers and track all head movements. A testing method supported by glasses makes sense since during the testing people will move or have to turn their heads, e.g., when driving.
• An external camera opposite the tester can record their eye movements. This method can be used when mon­it­or­ing displays or other in­ter­faces.
• Webcams can be used for eye tracking with the correct software. The per­form­ance is however nowhere near as good as with hardware that has been specially developed for eye tracking. For example, webcams built into the screen are not able to work in the IR spectrum.

In terms of software, there are some freeware programs offered by different de­velopers. Among these include:

• Xlabs from xlabsgaze
• Opengazer from Samsung and the Gatsby Char­it­able Found­a­tion
• Pygaze from Oxford, Aix-Marseille and Utrecht uni­ver­sit­ies
• Ogama from the Freie Uni­versität Berlin
• OpenEyes from Thirty Sixth Span Internet Tech­no­lo­gies

The main dis­ad­vant­age here is the lack of support, making these tech­no­lo­gies more for hobbyists who have pro­gram­ming knowledge.

In contrast, there are some com­mer­cial solutions that offer customer support as well as the ability to make customer-specific ad­apt­a­tions.

• Tobii Pro from Tobii Pro
• iMotions Core from iMotions
• VERIEYE from neur­o­tech­no­logy
• D-LAB from Ergoneers

By using eye tracking, you can determine where a user looks first. You can also generate a heatmap analysis, which shows the areas in a user’s field of vision that are looked at the most. This is done by using different colours and varying the intensity within the in­di­vidu­al colours. Anything displayed in red shows an area which is viewed often, while anything in blue is hardly seen at all. This means you can find out which areas in someone’s field of vision are barely looked at.

The phe­nomen­on of not looking at online ads is known as `banner blindness´. It describes a situation when certain elements or areas of a website are knowingly or un­know­ingly not seen by users. This ‘blindness’ towards ads can result from the habit of our brains trying to separate important and un­im­port­ant in­form­a­tion. By using eye tracking, you can improve the placement of ads on websites, therefore, gen­er­at­ing more attention for them.

By analysing eye tracking data, UX designers can improve user ex­per­i­ence. This means dis­play­ing complex systems as easily as possible so that users can do what they want to as easily and quickly as possible. This helps to make apps and terminals easier to use, the dash­boards in cars clearer, and improve the read­ab­il­ity of in­struc­tion manuals and websites.