Ever watched a TV commercial and not known what it was advertising? Sometimes we can see the same advertisement day after day and even become familiar with the advertisement’s narrative content. Yet when asked what the advertisement is trying to sell, we are at a loss. The question is why is the commercial failing so badly?One way to answer this question is to run a marketing research study and simply ask respondents why they didn’t or couldn’t engage with the branding message in the advertisement. This might provide an answer. However, research has shown that visual attention is complex and involves both conscious and unconscious impulses. Because visual attention often depends upon unconscious impulses, respondents may not really understand their own visual behaviour. This can lead respondents to give rationalizations for their patterns of visual attention that are, in fact, quite wrong. This is a serious problem as, in marketing research, a wrong answer is often much worse than no answer at all.You may well have heard of eye-tracking for marketing research. When used in a marketing research study, eye-tracking can give important insights into viewers’ engagement with marketing material through visual behaviour analysis. At a very basic level, visual behaviour analysis allows the marketing researcher to see through the eyes of the customer and to determine the customer’s focus of attention at any given point in time. The hope is that by conducting visual behaviour analysis, we can spot potential problems with the marketing material before the campaign is launched.What can visual behavior analysis tell us that we don’t already know? Marketing professionals rely upon marketing research to garner insights into customer opinions and behaviour. This data is often interpreted with the aid of empathic skills, intuition and experience. However, eye-tracking gives a more direct access to the viewer’s thought processes through visual behaviour analysis. This is important as eye-tracking is not merely about viewers’ eye-gaze patterns: visual behaviour analysis helps us understand what the viewer is thinking. When we watch a viewer’s eye-gaze pattern over an advertisement, we gain an understanding of the viewer’s thought processes. What they are looking at and why? Are they paying attention to the key branding visuals? What is the link between attention to branding visuals and the ability of the viewer to recall branding information at a later date? Do the viewers read textual information? If so, how much of the text do they read?These are just some of the generic insights offered by visual behaviour analysis. However, when we combine visual behaviour data with contextual information relating to the advertisement, the respondents’ demographic data and the respondents’ self-reported data, it is possible to build up a rich picture of the viewers’ overall engagement with the advertisement in terms of both behaviour and underlying opinions. This data helps us to better understand the viewer. It helps us determine what marketing messages work for viewers and what marketing messages leave them cold. As part of a multi-modal marketing research study, eye-tracking allows us to determine if the viewers ‘get’ our marketing message. If the viewer does ‘get it’, eye-tracking studies will tell us why and if the viewer doesn’t ‘get it’, the visual behaviour analysis will give us the data we need to determine why the advertisement has failed.Eye-tracking involves three important steps. These are:-The study – for the results of the eye-tracking study to be valid, the study itself must be performed using a rigorous research methodology. What this means is that the study should be performed in a scientific manner. This is often a point of confusion as some people claim that eye-tracking is not a science but rather qualitative and subjective. This is both true and false. It is true that eye-tracking data can be analysed in a qualitative way. The analysts can draw subjective inferences from the eye-tracking data. However, the validity of these inferences depends upon the validity of the data upon which they are founded. In order for the data to be valid, it must be collected in a scientific fashion. Failure to do so will not only lead to validity problems with the data but will seriously undermine the validity of any inferences drawn from the data.The Analysis – at its most basic level, eye-tracking data reduces to a series of ‘point of regard’ co-ordinates. For screen based test media, this can be a data file containing time-stamped screen co-ordinates of the tracking subject’s eye-gaze. This data needs to be analysed to gain useful insights from the study. What can be done? Well there are many useful eye-tracking metrics. For instance, it is possible to track every glance test subjects make on the product as and when it appears on the screen. To do this, the product visuals are tracked within the advertisement and intersected with the test subjects’ point of regard co-ordinates. This will allow the analyst to quantify the test subjects’ focus of attention on the product and monitor their level of attention over time. Basically, if a metric involves viewer’s focus on attention to media visuals, it can be used.The interpretation – provided the eye-tracking data has been collected in a valid way and processed so as to produce useful information, the eye-tracking analyst will provide you with a rigorous set of data and metrics relating to the viewer’s engagement with the advertisement and highlight potential problem areas. The eye-tracking data will be complemented with test subjects’ self-reported data. Respondents will be questioned about problem areas within the media and their overall level of recall of branding information will be assessed. Where retention of key marketing messages is wanting, the analyst will review the respondents’ eye-tracking data to try to discover what went wrong.Consider the benefits of running eye-tacking studies against prospective marketing campaigns before they are launched. The visual behaviour analysis could identify problems with a campaign which could be corrected before the campaign begins. This has the potential to make campaigns more effective and allow you to avoid the situation where viewers are watching your advertisement with little idea of what you are trying to sell.