If you belong to a commercial enterprise with a Web presence, how do you view your company's home page? Frankly, your views do not matter. Like Richard Serra's award-winning sculpture, which had to be removed from the New York City Federal Building in the face of public pressure, what really counts is what your audiences think about your home page. And we have reasons to believe that your audiences might be viewing your home page primarily as an advertisement designed to entice them to experience the site.
At first glance, the proposition that a Web home page is an advertisement may seem somewhat farfetched. But thinking of a home page in advertising terms not only makes sense, it has far-reaching and profound implications for the design and evaluation of Web sites. In this article, we will flesh out the details of the home-page-as-an-ad proposition and describe the implications of this communications view.
Back to Top
Home Page as an Ad: The Conceptual Bases
Can a home page qualify as an ad? We approach this question from three angles: conceptual, physical, and functional. We consider whether a home page meets the conceptual definition of advertising, whether it physically resembles an ad, and whether it performs the same functions as an ad.
By definition, an advertisement is a "paid form of nonpersonal presentation of ideas, goods or services, by an identified sponsor, with predominant use made of the media of mass communication" . And indeed, a home page meets these conceptual criteria: it is paid for by a sponsoring institution/company; it is designed with large audiences in mind and hence is impersonal; it is an entry point to a Web site that promotes one or more ideas, goods, or services; it has an identified sponsor; and the Web, with its hundreds of millions of users, is a mass medium.
Do home pages resemble ads? Given that the Web is a relatively new mediumquite unlike traditional mediahaving a physical resemblance to traditional ads is not a fundamental requirement for home pages to be considered ads. However, at least at first impression, many home pages are beginning to look like traditional ads. For example, at the time of the writing of this article, the home page of the Dos Equis beer company at www.dosx.com/ and the company's print advertisement were virtually identical. Another recent trend is the use of an entry page with enticing graphics and minimal hyperlinksnot unlike a colorful print adwhich links to a more traditional home page.
Do home pages perform the same functions as advertisements? Advertisements are fundamentally communications messages and, like all communications, they serve two basic functions: to inform and to persuade. To inform is to make consumers aware of the product/service and to persuade is to create positive attitudes and behavior vis-a-vis the product/service. For example, a print ad of the Dannon yogurt company informs consumers about the health benefits of Dannon plain nonfat yogurt and hopes to persuade them to buy it. Likewise, the Dannon home page at www.dannon.com provides links to detailed information on its yogurt offerings and attempts to persuade existing and potential customers to buy Dannon products.
In sum, home pages meet the conceptual definition of advertising, they resemble ads in physical appearance, and they perform the same basic functionsto inform and to persuadeas other communications messages. To be sure, a home page also differs from a traditional ad in important ways. For example, it is available to the consumer on demand 24 hours a day, and a surfer may visit a Web site from any place at any time.
Furthermore, unlike traditional ads, a Web site offers a user the ability to interact with the site and to navigate to various pages. Using internal and external hyperlinks, a company can make tremendous amounts of information available using a combination of media, text, graphics, and video. Moreover, the content of this multimedia communication can be continually updated.
That said, the differences between home pages and traditional ads merely reflect the unique advantages of the Web medium; they do not undermine the premise that a home page is fundamentally a communications message.
Back to Top
Advertising and Home Pages: Learning from Theories of Communication
If home pages are like ads, what lessons can we learn about home pages from advertisers? In recent years, advertisers of commercial products and services have spent enormous amounts of money to persuade consumers to view their brand and company more favorably than those of their competitors and to convert consumers into ideal customers. An ideal customer not only holds a positive attitude toward the advertiser and its brands but also buys the company's products repeatedly and spreads positive word-of-mouth advertising about the company. In short, an ideal customer is an evangelist for the brand and the company. It is not difficult to extend this idea to the Web context. From a sponsor's point of view, an ideal Web visitor will hold positive beliefs and attitudes toward a site and sponsor, will frequently visit the site, buy products from the site if available, and will spread positive word-of-mouth advertising about the site. Designing effective messages (ads or Web sites) is a key ingredient in creating an ideal customer.
The information function of a message is only the first step in the two-step inform-persuade hierarchy. The ultimate goal of any ad is to persuade the target audience to view the advertiser and its products favorably and to buy the advertised brand. In short, an ad informs only to persuade, that is, to create favorable attitudes and (purchase) behaviors vis-a-vis the advertised brand. Information is the means, persuasion is the end.
Of course, "inform" has a broader meaning in this context and it also includes the role of producing awareness, impressions, knowledge, and beliefs about the advertised brand or product and the sponsor of the ad. Thus an ad or a home page is serving its information function when it creates positive beliefs about the advertised brand ("This seems like a good brand of yogurt"), its sponsor ("Dannon cares about health"), or the ad itself ("This is a very creative ad"). This idea that a home page, a Web site, or an advertisement are all communications messages designed to inform and persuade is depicted in the basic communications model shown in Figure 1. The figure also provides examples of beliefs, attitudes, and behaviors in advertising and in a Web context.
How can this communication model be applied to the Web? When a visitor arrives at a Web site, the first thing he or she typically encounters is the home page. If it is to be effective, the page must provide the right information content and it should have a look and feel that persuades a viewer to spend time exploring the site as well as engaging in other favorable behaviors such as bookmarking the page and revisiting the site. If for any reason the home page fails in its inform/persuade role, the visitor is likely to move on to other sitesan easy enough action, especially when other attractive Web sites are literally mouse-clicks away.
Back to Top
The Information Function of a Communications Message
Whether a surfer explores or skips a Web site largely depends on his or her beliefs and attitudes toward the home page and the site. Even a brief exposure to a home page is capable of creating beliefs, which in turn influence attitude and behavior toward the site. Most beliefs are formed on the basis of information provided in the message. For example, a car ad might refer to objective laboratory tests and claim that a particular model has a fuel consumption rate of 60 miles per gallon. This information will tend to create a belief that the advertised model of the car is highly fuel efficient.
Not all beliefs are formed on the basis of explicit information contained in the message, however. Certain beliefs are "inferential" beliefsthat is, they are inferred from the message even though there is no specific information in the message about those beliefs. For instance, an unusually long ad (say 10 pages) for a car may create a belief that the car's price will be high. The fact that this belief may be false is beside the point. What is critical is that the consumer "believes" the brand to be high-priced; the source of that belief is less relevant.
Home pages perform the same basic functionsto inform and to persuadeas other communications messages.
Inferred beliefs play a crucial role in advertising. One widely held and well-investigated inferred belief about commercial messages has to do with the nature of their appeal. When exposed to a message, audiences typically make an inference about the degree to which a message is rational or emotional in character [3, 8, 9]. Designing a message that will be perceived as emotional, rational, or mixed in its appeal or tone is a primary strategic consideration in advertising.
All advertisements can be placed along a continuum from predominantly informational or rational ads at one end to predominantly emotional ads at the other. A predominantly rational ad (such as an ad for a yogurt brand) may contain information about a brand's attributes and benefits, its availability, price, warranty, and comparisons with competing products. An emotional ad, in contrast, may have an affective theme that attempts to appeal to one's feelings. Such ads typically are less wordy and contain more visual elements. Regardless of their nature, all ads "inform" because they create awareness, impressions, knowledge, and beliefs about the products, the ad, and the company.
If home pages are like ads, then it is reasonable to predict that like ads, home pages too are reliably classified by Web visitors along emotional and rational dimensions; that is, Web visitors form beliefs about whether a home page is primarily emotional or rational in its tone. Designers of Web sites should therefore be aware of the type of appealrational, emotional, or mixedthat would be most effective with their target audiences. However, the Web is awash in rational home pages, suggesting that Web designers have either assumed that rational appeals will be most effective or that they have ignored other kinds of appeals.
Back to Top
The Persuasion Function of a Communications Message
An important advertising finding is that the persuasiveness of a rational or emotional ad depends on how involved or motivated the audience is in processing the ad. Rational ads are typically more persuasive when audience involvement is high, that is, when the audience pays attention to the informational content of the ad and is willing to spend time processing it. When audience involvement is low, however, the emotional aspects of the ad hold sway. The logic behind this assertion comes from a communications model called the elaboration likelihood model .
According to this model, exposure to an ad is likely to generate two types of thoughtscentral and peripheral. Central thoughts relate to the main points of the ad and are based on a careful, in-depth consideration of the merits of the advertised product. Peripheral thoughts, on the other hand, are secondary to the focus of the advertisement; require little mental effort; and are generally based on such characteristics as the emotional tone of the message, the use of graphics and colors, the setting of an ad, and the spokesperson for the ad, if any.
A motivated audience, when faced with a communications message, tends to engage in central processing and the persuasiveness of the message largely depends upon careful weighing or elaboration (hence the name elaboration likelihood) of the information presented in the message. Elaborating or focusing on the informational aspects of the message generates thoughts about the attributes of the advertised product. In such a case, the informational aspects of the message play the major role in persuading the audience and peripheral cues such as celebrity endorsers are likely to be ignored.
In contrast, an uninterested audience has little motivation to scrutinize the informational aspects of the message. Such an audience is more likely to rely on peripheral processing, a low-involvement activity characterized by shallow, perfunctory consideration of the message at a sensory levelpresumably at the expense of the more relevant or central aspects of the message. Such ad processing typically generates thoughts and feelings that are not directly related to the central message itself but that nevertheless influence the persuasion process by association. An audience engages in peripheral processing when it is unwilling to engage in deeper, effortful processing because it may lack interest in the advertised product or does not have enough time to process the message, or if the message is too technical or otherwise complex for the audience to comprehend.
Consider a popular ad for a fabric softener whose sponsor is fully aware of the low involvement nature of the product. Realizing that most people are unlikely to engage in the cognitively taxing exercise of scrutinizing the finer qualities of the advertised product, the advertiser cleverly includes a picture of a lovable, cuddly bear. Given the low-involvement situation and the resulting peripheral processing, the audience is more likely to focus on the cuddly bear's picture than the informational attributes of the fabric softener. This processing is probably involuntary, effortless, and takes virtually no time. The audience's attitude toward the fabric softener is thus shaped not by the rational information contained in the ad but on the basis of the favorable emotional reactions to the bear's picture. Notice that the ad designer's accent here is on "how did the message make you feel?" rather than on a rational consideration of the ad's merits.
How do the preceding ideas apply to the Web? Visitors to a Web site can be classified into two broad categories, low-involvement hedonistic surfers and high-involvement utilitarian searchers henceforth, surfers and searchers, respectively. Surfers are fun-seekers and explorers who desire entertainment and stimulation; they are likely to land at a Web site, linger for a brief period and then take off for another more attractive site in their path. In contrast, searchers are goal-oriented, looking for specific information, and are likely to spend more time at their preferred sites. Applying the elaboration likelihood model, it is easy to see that surfers are more likely to engage in shallow, sensory-level, peripheral processing of the executional aspects of the message (graphics, images, headers, and footers) but searchers will pay more attention to the contents of a message, processing it at a deeper, semantic level and elaborating on the merits of the message's arguments. In the case of the surfer, the emotional component of the message will tend to exert greater persuasive influence than the informational or rational aspects of the message, but to a searcher, the opposite should hold true, that is, the informational component of the message is likely to play a more dominant role.
Back to Top
Empirical Test of Home-Page-as-an-Ad Proposition
So far, our thesis has been that home pages and advertisements are fundamentally communications messages that behave alike. Both create beliefs and attitudes in order to engage audiences in favorable behaviors. Equally important, these attitudes and beliefs are predictable, giving us a measure of control in influencing audiences' behavior. Based on the preceding discussion, if home pages truly behave like ads, then it is reasonable to expect the following:
- Web surfers can reliably classify home pages along emotional and rational dimensions.
- When surfers (rather than searchers) process a Web home page, the emotional dimension of the home page will exert a greater influence on persuasion judgments than will the rational dimension of the page.
We tested these two propositions in a lab study. A total of 19 student volunteers browsed 10 home pages of existing commercial Web sites as they would normally surf and then provided their reactions to these pages. Five of the pages were highly informational and seemed apparently rational. The other five pages, which combined information and eye-catching pictures/graphics, appeared to be employing a mixed (rational plus emotional) appeal. After browsing a page, the students provided written feedback, which consisted of the following:
- The extent to which the students thought that the page was rational-emotional. This was measured by asking students to think of each home page as a person and to rate it on seven characteristics: tender, factual, heart warming, sensitive, gentle, rational, emotional, and logical on a seven-point semantic differential scale.
- Two persuasion measures were also used. One measured the attitude toward the home page ("What is your overall impression of this home page?"), the other measured the attitude toward the sponsor of the home page ("What is your reaction to the sponsor of this home page?"). Each was measured on a seven-point scale with the endpoints labeled as "Favorable" and "Unfavorable," respectively.
- Finally, students indicated the likelihood that they would further explore the Web site on a seven-point scale ranging from "very likely" to "very unlikely."
Back to Top
Our findings indicate that people were able to classify home pages on rational and emotional dimensions according to our first proposition. For example, all five pages judged a priori as informational scored the highest on the rationality dimension (an average score of 5.0 or more on a seven-point scale) and lowest (less than 3.0) on the emotional dimension. The remaining five mixed-appeal home pages had moderate scores on the two dimensions. Further statistical analyses revealed that subjects in our study were able to accomplish the classification task with a high degree of reliability. The reliability index (Cronbach's alpha) value was 0.95 for the emotional items and 0.91 for the rational items. These values are very robust considering the fact that 1.0 is the highest possible value for this reliability index.
The emotional aspects of a home page had a significantly greater impact on the persuasion responses than did the rational aspects.
To test our second proposition, we conducted three separate regression analyses on the three persuasion measures: 1) attitude toward the home page, 2) attitude toward the sponsor of the home page, and 3) propensity to further explore the underlying Web site by clicking on an icon/button on the home page. The independent variables in all three cases were the emotional and rational scales. Our expectation was that the emotional aspects of a home page would be more important in determining the degree of persuasion than the rational aspects. That is exactly what the results showed.
The emotional, but not the rational, dimension had a statistically significant effect on the persuasion measures in all three regressions. Both of our predictions were thus supported by the datahome pages were reliably classified as emotional or rational by the surfers and, like other forms of communication under the low-involvement processing conditions, the emotional aspects of a home page had a significantly greater impact on the persuasion responses than did the rational aspects.
Back to Top
Implications of the Web Home-Page-as-an-Ad Concept
Viewing a home page as an ad whose main goal is to entice potential audiences to explore the site has significant implications for the design and evaluation of home pages.
Diagnostic pretesting of home pages before they go online. The Web is increasingly awash in online and banner advertising. However, one must realize that the most important online advertisement of all is the home page itself. An organization's home page is the gateway to organizational information on the Internet, and hence has a tremendous potential to entice customers to the site or to drive them away. The expenses involved in creating and maintaining a large commercial Web site can easily run into millions of dollars. Despite the expense, if a home page fails to entice a surfer to explore the underlying Web site, the other aspects of the site become irrelevantat least on that occasion.
For these reasons, it is absolutely essential to know what features of a home page entice viewers to further exploration and which ones are distracting, neutral, or offensive. This suggests that home pages and Web sites, like advertisements, should be routinely subjected to diagnostic pretesting. In diagnostic pretesting, a group of potential consumers is first exposed to the ad being tested and then asked to respond to a number of questions concerning the ad. This consumer input is then used to further refine the ad before it is placed in circulation.
Diagnostic pretesting is not quite the same as standard usability testing of software, which is not quite the same as usability testing of Web sites. In conventional usability testing, real users test a system on various criteria such as ease of learning, retention of learning over time, speed of task completion, error rates, and subjective user satisfaction . Usability testing of Web sites as currently practiced is focused more on specific attributes of a page or site, such as download time, ease of navigation, style, backgrounds, or the use of graphics. Such usability tests are designed to improve the "usability" of a Web page or site under the assumption that people exposed to a Web page will explore it. This assumption may well be true for a utilitarian searcher but is not so for a casual surfer, whom the page must invite, attract, or pull by its look and feel to engage in further explorationat which point usability becomes critical. Traditional copy testing (or pretesting) methods can be used to assess the attractiveness or pull potential of a home page. Copy testing methods focus on such issues as: How quickly does the home page capture and hold the viewer's attention? Do audiences find a home page, on the whole, attractive or unattractive? Which parts of the home page need modification to enhance their appeal?
Advertisers have perfected the art and the science of copy testing. They can evaluate an ad based on its potential effectiveness and simultaneously diagnose why the evaluation turned out the way it did . Because a Web page behaves like an ad based the entire accumulated repertoire of advertising copy testing tools can be used in diagnostic copy testing of home pages to make them more appealing. Thus, even before a home page goes online, one can pretest to see whether a supposedly informational home page is indeed perceived as informational; if it is received favorably; and to ensure that it will not create boredom, or negative feelings in potential surfers.
Implications for post-testing home pages. An equally important issue is measuring the overall effectiveness of a home page once it is online. At present, the most common measure of a home page's effectiveness appears to be the click-through rate or hit rate, which is one of a class of behavior-based measures of overall site effectiveness. (Other behavior-based measures of site effectiveness include the time spent at the site, navigational patterns through the site, and repeat visits to a site.) However, the hit rate, while useful as a gross measure of the popularity of a page, is limited in several ways. Typically, the hit rate is inflated by spurious hits recorded by a server for reasons that have nothing to do with the effectiveness of a page. For example, large numbers of hits for a page may originate from users who have set it up as their default page for opening.
Moreover, the hit rate is a gross measure; it only tells us that a surfer has been at a site. It does not say anything about how a surfer reacted to the site, which aspects of the page he or she liked the most or the least, and what made the surfer explore the page in the first place. Focusing merely on behavioral measures has several negative consequences, including underestimating page/site effectiveness and ignoring much-needed diagnostic information that could reveal which aspects of a home page led to what behavioral response(s)information critical in designing and maintaining an effective home page and Web site.
Relying exclusively on behavior-based measures of communication effectiveness could be misleading in other ways too. Consider a surfer who, after viewing a home page, decides to skip the rest of the site. Does this behavior imply the home page was ineffective from the sponsor's point of view? Not necessarily, especially if one takes the view that informing and persuading are the primary functions of a home page. First, if nothing else, the surfer may have developed some awareness (information) about the sponsor and the sponsor's offerings. (In a related study currently in progress, preliminary results indicate that after 11 seconds of viewing a home page, subjects show up to 25% recall and 30% recognition of information on home pages.) Second, it is possible that the home page left a positive impression on the surfer. Third, a positive attitude about the home page could transfer to a positive impression about the company and its products as well. (In the present study, attitude toward the home page and attitude toward the sponsor were significantly correlated.)
So how does one measure the effectiveness of a home page or a site? From an advertising perspective, the answer is clearmeasure the effectiveness in communications terms. The essential question to ask is this: Did the message create desired beliefs and persuasive intent toward the message, the object/idea being propagated by it, and toward the sponsor of the message? A message that generates more favorable attitudes and beliefs in the target audience is deemed more effective than other messages.
For most communications messages, an organization should set information goals to inform the consumer of benefits or attributes of the product and persuasion goals to develop liking for the organization's ad and brand. Consider a car manufacturer who is introducing a new, highly fuel-efficient model. The manufacturer may set information goals that within three months from the onset of the ad campaign, 70% of the target audience should be aware of the brand and 50% should know the car is highly fuel efficient. Likewise, persuasion goals that 65% of the target audience will have a favorable impression of the ad and 45% will have a favorable impression of the advertised brand might be set. (Note that progress on these goals is measured within a specific time frame using prespecified scales or measures.)
Because a home page and its underlying Web site are fundamentally communications messages, it should be possible to develop goals and to measure the effectiveness of the site in communications terms. Sample communications goals for a Web site include:
- Information/Awareness/Belief Goals. Within a month of the site's launch, 20% of the target Web audience should have visited the site and should be aware of the sponsor of the site and its major products and services. The home page should be perceived to have emotional appeal, although other pages at the site could be perceived as being rational. At least 50% of consumers who visit the home page (while casually browsing the Internet or by using directed searches) should be encouraged by the emotional appeal of the home page to further explore and navigate to another page in the site.
- Persuasion/Attitudinal Goals. All visitors to the site, regardless of whether they explore the rest of the site, should have positive feelings, positive evaluations, and favorable impressions of the home page, the site, and its sponsor. (These goals will, of course, have to be quantified.)
We picked these commonsense goals as examples. In an actual situation, a site manager will have to carefully select specific communications goals for a site in light of overall company goals and measure its effectiveness using reliable scales.
Back to Top
Conclusions and Future Research Directions
We have conceptually and empirically shown how home pages are similar to ads and we have provided broad guidelines on how this knowledge can be applied by managers and technicians.
Specifically, we have drawn from advertising research to present some ideas about measuring communications effectiveness and copy testing. Happily, there is much more accumulated knowledge in advertising that can be tapped to enhance our understanding of Web-based communications. For instance, knowing how a surfer develops positive attitudes toward home pages can be very valuable in designing effective pages and advertising research suggests some answers to this question.
To the extent home pages behave like ads, prior research suggests that exposure to a home page, however brief, generates feelings (happiness, for example) and evaluations ("this page is dull"), which contribute to the formation of an attitude toward the home page. (In a recent study, we were able to verify this idea: feelings and evaluation responses accounted for between 47% and 67% of variation in the attitude toward home pages. Moreover, we discovered a strong correlation between these attitudes and people's intentions to further explore Web sites.)
The home-page-as-an-ad proposition leads to many other interesting questions. For example, do favorable reactions to a home page transfer to the products described on the site? In other words, do favorable reactions to a home page lead to greater purchases of advertised products?
An important set of questions revolves around Web design issues. For example, how effective is a home page (read "ad") for surfers if it has too much information and far too many hyperlinks? What is the implication of having an entry page that is not a home page in the traditional sense but is really designed to draw audiences to the actual home page, which is linked to this entry page? At what stage in Web design should inputs be sought from marketers?
In seeking answers to these questions, we need not limit ourselves to a communications viewpoint or even a computer-based perspective. Consider the following statement: "When people view a [home page], they are making a judgment, however intuitive and unconscious the process may be. This judgment concerns the sorts of experiences they would have, the ease of locomoting, of moving, of exploringin a word, of functioningin the environment they are viewing."
This quote, which seems to offer an appropriate description of Web browsing, originally described how people evaluate landscapes in the context of architectural aesthetics, not home pages; we simply changed Kaplan's quote  to replace landscape with home page. This example shows the promise of but one established discipline, architectural aesthetics, in complementing and enriching our existing computer-based perspectives of the Web environment. Numerous other disciplines (for example, anthropology and cognitive psychology) can likewise contribute to our understanding of the Web.
Back to Top
1. Engle, J.F., Washaw, M.R., and Kinnear, T.C. Promotional Strategy. Irwin, Burr Ridge, Ill., 1979.
2. Hofacker, C.F. Internet Marketing. Digital Springs, TX, 1998.
3. How advertising works: A planning model. J. Advertising Research 26, 1 (Feb./Mar. 1986), 5766.
4. Kaplan, S. Perception and landscape: Conception and misconceptions. In J.L. Narar, Ed., Environmental Aesthetics: Theory Research and Applications. Cambridge University Press, 1988.
5. Leckenby, J.D. and Plummer, J.T. Advertising stimulus measurement and assessment research: A review of advertising testing methods. Current Issues and Research in Advertising, Vol. 2, 1983.
6. Nielsen, J. Multimedia and Hypertext: The Internet and Beyond. AP Professional, Boston, 1995.
7. Petty, R. and Cacioppo, R. Attitudes and Persuasion: Classic and Contemporary Approaches. Brown, Dubuque, IA, 1981.
8. Vaughn, R. How advertising works: A planning model. J. Advertising Research 20, 5 (1980), 2733.
9. Zeitlin, D.M. and Westwood, R.A. Measuring emotional response. J. Advertising Research 26, 5 (Oct./Nov. 1986), 3444.
Back to Top
Back to Top
Figure 1. A basic communications model.
Back to top
©1999 ACM 0002-0782/99/0800 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 1999 ACM, Inc.
No entries found