The project
Can't see or just not looking?
The Internet allows us to gather information about world affairs, search for a home, car, or holiday, entertain ourselves or just browse. We can do all this on congested websites, without the slightest prospect of being able to view or read through all the information appearing on such sites at the same time. Our eyes are not a camera taking a snapshot of the object in view at any time, and nor are our brains a film or digital memory, from which all previously registered information can be called upon whenever we like.
Since we don't possess this capacity, we are forced to separate those things that matter to us from the unwanted, in other words we make selections from the deluge of information and continually direct our attention towards input that we consider important.
This is increasingly true of today's websites, where the user has to extract the precise information that is important to him/her from the rapidly changing chaos of colorful space, films and text. A question that is of interest to an increasing number of people arises here: what is that attracts the Internet user's attention on a page, what and how does he/she look at the page, and in what way does this influence decisions in the present and future?
Advertising experts with furrowed brows
Advertising specialists, who live in a world of impact and effectiveness, face the abovementioned dilemma, not only with regard to the Internet, but also for almost all channels of communication. Whereas several industry research studies measure encounters with individual channels, pages, websites, etc., or often with the given advert itself, the databases containing the results of such studies, which generally operate as an industry standard, do not always provide an exact answer to the question of what has been seen - and what has not been seen - by the person the message is intended to reach.
One of the largest media agencies, OmnicomMediaGroup, and the C3 Foundation launched a research project for improved understanding of the above problem. The study, known as Eyetracking, involved experts in vision research, which is considered a relatively new branch of science, and communications specialists from Omnicom agencies. A new instrument was employed during the course of the study - the eye camera.
Who or what are the C3 Foundation?
This nonprofit organization formerly associated with George Soros has already worked on a large number of vision research projects, and has several employees who teach related sciences (neurobiology, computer applications for visual information, etc.) at higher education institutions. Nor are new media instruments unfamiliar to C3, since they have examined their effect on society in various projects. On more than one occasion they have also been involved in popularization - just think of the development and communalization of the free mailing system Freemail, now the most popular domestic electronic mailing system, which forms part of the [origo] group.
Who or what are OmnicomMediaGroup?
Media agencies, also known as communication agencies, are responsible for conducting business, selecting the appropriate environment and placing adverts or even inserting a break in our favorite journals, shows or sites. They are generally companies with multinational interests on the advertising market, which is currently undergoing a strong process of concentration. One of the largest agencies is OmnicomMediaGroup, at least in respect of the number of employees, and according to data from TNS Media Intelligence, which measure list price expenditure, they also handled the largest turnover in 2006. Innovation within the industry and the future of the advertising profession are recurring ideas at Omnicom, and several professional initiatives or forums may be linked with these notions. Examples include the national Media Student Contest, which has been held annually since 2001 for undergraduate students at higher education institutions. Possibly as a result of such initiatives OmnicomMediaGroup is the agency with the greatest number of awards, both domestically and internationally.
How did it start?
In the first round of research, initiated a few months ago, the Internet was placed under the microscope. The two companies first intended to analyze the extent to which the selected methodology and technology were suited to studying perception. Given that they wanted to examine the subjects under real conditions, the monitoring of Internet usage seemed most advisable because of the technical possibilities. They made arrangements for members of a sample group of 120 persons, representing domestic Web users of working age (15-49) based on gender, age group and educational qualifications, to sit down at a computer, and asked them to look at some virtual Internet sites. The eye camera recorded the way they tracked the unprecedented Hungarian-language content, and the data obtained in this way were crosschecked with an exit interview, in which recall was examined.
What did they learn? - We look at whatever catches our attention
Attention and eye movement go hand in hand, i.e. they scan the visual environment. Regardless of whether we look at something intentionally, or are automatically attracted, our attention always turns to this first (not by much, by less than one tenth of a second) and prepares our brains to efficiently process information following eye movement. And if we are awake, this process continues in the same way without interruption - we direct our eyes to another part of the surroundings three or four times a second. The pattern of eye movements formed in this way precisely demonstrates what it was that attracted and escaped the attention of the consumer, who is both a product and media consumer at the same time. Measuring patterns of eye movement is therefore an objective way and, based on our current knowledge, one of the most promising methods for examining user habits with regard to obtaining information carried in complex visual space.
Big Brother is watching you - the eye camera
The eye camera barely seems to differ from a slightly larger web camera. It is capable of tracking eye movement by following infrared light reflected from the retina. Lamps releasing infrared light form accessories to the camera. The solution is completely harmless to the human eye, since it works with very low-strength infrared light. With the assistance of a computer and the camera operating in the infrared domain (i.e. invisible to humans) and placed before the subject under observation, it is capable of tracking the eye movements of persons involved in the study without any kind of intervention or disruptive factors.
During the course of such studies, observers view the websites we have selected at their own pace and discretion. Through evaluating and processing data acquired in this way, we are given a graphic illustration of what the individual looked at on any page. However, in this case, if the eye movement pattern of several observers is recorded for the same website, the specific browsing habits of various groups (such as men and women) may be described and compared through averaging and evaluating the received data.
Top or bottom? Right or left?
One of the most important objectives of the studies was to examine, by tracking eye movement, whether considerable deviations really can be detected in the extent to which browsers pay attention to adverts appearing in banners placed on various parts of websites. It was revealed that we watch for almost twice as long, and view the same adverts and creatives when they appear in the banner on the right or at the top of a website than when they are placed in banners on the left or at the bottom of the page. These results are of particular interest and importance because it is widely accepted that browsers scan pages in a pattern resembling the letter F. However, it might be deduced from this that adverts placed on the left-hand side and at the top of pages are most effective. In contrast, the results of the studies show that banners on the upper and right-side are the most viewed, which coincides with experience and is also reflected in advertising costs: adverts in banners on the right and at the top of websites are more expensive than in banners on the left and at the bottom.
One of the most important objectives of the studies was to examine, by tracking eye movement, whether considerable deviations really can be detected in the extent to which browsers pay attention to adverts appearing in banners placed on various parts of websites. It was revealed that we watch for almost twice as long, and view the same adverts and creatives when they appear in the banner on the right or at the top of a website than when they are placed in banners on the left or at the bottom of the page. These results are of particular interest and importance because it is widely accepted that browsers scan pages in a pattern resembling the letter F. However, it might be deduced from this that adverts placed on the left-hand side and at the top of pages are most effective. In contrast, the results of the studies show that banners on the upper and right-side are the most viewed, which coincides with experience and is also reflected in advertising costs: adverts in banners on the right and at the top of websites are more expensive than in banners on the left and at the bottom.
Differences between genders
The studies also revealed that in the case of individual creatives, a significant difference can be detected between male and female browsers in respect of the length of time they spend watching, and the extent to which they are subsequently able to recall and identify creatives. The clear lesson from these results is one of the hobby horses long referred to by media agencies, which suggests that it is worth approaching different target groups with adverts and creatives that best conform to their individual characteristics and user habits.
Another interesting lesson revealed by the studies was that, by crosschecking the results of questionnaires describing user habits with the results of eye-movement patterns, no correlation was found between the extent to which the observer considers himself/herself an "advert avoider" and the rate at which he/she watches creatives. This means that we cannot exclude adverts from our lives, they even have an impact if we don't want them to and when it seems we have managed to ignore them.
Where next? - Supplement not a replacement
C3 and Omnicom have made a very positive assessment of the initial experiences, and definitely intend to continue research in future, this time using a larger sample. They would like to acquire further experience on the perception of adverts appearing on the Internet, but have not excluded the possibility of extending the study to other types of media, and for example consider it important to study use of the press and watching TV. OmnicomMediaGroup is able to exploit the results obtained in this way to improve the service to its clients, since such additional information may involve more precise targeting, and the lessons of Eyetracking may also bring new elements into media owner relations.
With regard to using the results, they have stressed the fact that they are not looking to find alternatives to existing industry solutions, but to work alongside them and approach the issue from a new perspective.