|
|
Why Web Design Matters to Internet Advertising
|
|
 |
 |
|
As interest in the Internet as an advertising medium has grown, significant energy has been expended on uncovering the impacts that ad format, ad creative, and ad targeting have on key branding and direct response metrics such as Brand Awareness and CTR. As a result, it is now widely understood that optimizing these three campaign elements is integral to optimizing the performance of any Internet-based branding or direct response campaign.
Yet, while the Internet Advertising industry has recognized ad format, creative, and targeting as being critical factors impacting advertising effectiveness, it has largely failed to recognize the depth of impact that Web site design has on advertising effectiveness.
How many ads should be displayed per Web page? Should skyscraper ads be positioned to the right side of the page (as they most commonly are) or to the left? Should square ad units be "embedded" in editorial content or not? How many distinct page elements (colors, navigational elements, etc.) should be used in the design of an ad-supported web site? Should 468 x 60 banners be displayed above or below a web site's main navigation bar?
These and related factors are rarely given consideration by either advertisers or ad-supported content publishers. Yet, there is clear, growing evidence that such factors substantially drive the bottom-line branding and direct response performance of Internet Advertising campaigns.
Just as ad format, ad targeting, and ad creative before it, what we see is that web site design is mistakenly approached by both advertisers and publishers as a matter of "art" rather than "science".
As Leslie Laredo, founder of Internet Advertising consulting firm Laredo Group told Media Life magazine in a recent interview, "Very few designers really think about or understand the ad proposition before they design the sites. Designers are coming up with designs for sites that might work on one level but don't work in terms of revenue production. The art of site design is not [yet] a science of site design."
A "scientific" approach to the design of ad-supported Web sites, and the use of site design as a media buying criterion is, in all probability, still at least two to three years from being mainstream. Admittedly, the reason for this is not simply shortsightedness: Research and data on the subject is still largely lacking, and will take time to produce.
Nonetheless, important research does exist, and smart advertisers and publishers must seek to comprehend this research sooner rather than later.
In short, we believe, advertisers now can and must attempt to use an assessment of a site's design as a factor in assigning value to that site as a piece of advertising real estate. At the same time, ad-supported content publishers must strive to determine means of content presentation that satisfy their content presentation objectives while maximizing advertiser ROI.
The facts: How Web site design impacts advertising effectiveness
According to research recently released by MSN, between 7% and 12% of the total time Internet users spend on Web pages is spent viewing ads. Yet, strikingly, according to MSN's research, only two of the four factors that determine how much time users devote to looking at ads on a given page - ad format and ad creative quality - are considered by the typical advertiser in formulating their campaign strategy, and in conducting media buys. The other two determining factors, rarely considered, are Web site design factors: page layout and the placement of ads on the page.
MSN's research in this instance was powered by usability company, Eyetracking, Inc., which uses patented technology to track the movement of users' eyes as they move across the page.
In a recent interview with avant|marketer, Eyetracking Inc. Senior Usability Manager, Sylvia Knust explained that a strong correlation between the amount of time that a user spends viewing ads and the user's ability to recall the various characteristics of those ads (which is directly related to their branding impact) has been consistently observed across multiple research studies conducted by Eyetracking Inc.
In 2001, both Terra Lycos, and C|Net separately retained Eyetracking, Inc. to assist in comprehensive redesigns of their respective Web sites. The main objective of the redesigns was to improve the effectiveness of the advertising being served via these sites. In specific, Eyetracking Inc. technology was deployed to enable each publisher to ascertain precisely how to re-configure the layout of its site so as to maximize the amount of time users spend viewing the ads placed on them (thereby optimizing the branding impact of those ads).
The Eyetracking, Inc.-backed Lycos site re-design, yielded powerful results. By re-configuring the design of the site such that the top placement 468 x 60 standard banner appears below, as opposed to above the main navigation bar of the site (which runs horizontally across the top of the page), Lycos increased the number of users that viewed their 468 x 60 top placement banner by 200% (thereby, effectively tripling the reach of this placement across their site), while simultaneously increasing the amount of time users spend viewing each of these ads, by 600%.
Brian Payea, a Terra Lycos spokesperson close to the Lycos site redesign, tells avant|marketer that the redesign has created significant, tangible ad performance improvements, pushing the CTR of the top placement 468 x 60 banner to levels that match the performance of the much larger skyscraper units. These results defy conventional wisdom on the relationship between ad unit size and ad effectiveness. The implication is clear: Web site design is a critical determining factor in the effectiveness of Internet Advertising.
Payea also tells avant|marketer that the Lycos/Eyetracking, Inc. research evidences an important and little-discussed fact that sheds further light on the closeness of the relationship between Web site design and ad effectiveness: Even within the course of a single usage session on any given site, ad effectiveness is subject to substantial decreases. Why? Early on in their usage session on a site, users come to recognize the location of advertising on the screen, and so quickly "train" themselves to tune out anything placed in those screen locations. Thus, ad tune-out is in many cases largely a function of Web site design.
To preempt such ad tune-out, sites in the redesigned Lycos network (as well as those in the re-designed C|Net network), utilize something Lycos refers to as "Dynamic Ad Placement", wherein the position of ads, as well as the particular ad formats that are being served (skyscrapers vs. square units vs. standard banners, etc.) change dynamically, as the user proceeds through the usage session. In this way, Dynamic Ad Placement prevents the user from forming reliable expectations as to the position of ads on the screen, thus increasing ad effectiveness.
What we have outlined thus far is data that points to ad placement as a very real driver of ad effectiveness. In addition to ad placement, there is also good reason to believe that a second Web site design factor, ad clutter (i.e. the number of different ads and advertisers per page), also helps to determine the effectiveness of Internet Advertising.
While there is currently no data that indicates specifically the optimal number of ads that should be served per page, it is well known from the traditional media (specifically, Television and Print) that ad clutter adversely affects advertising effectiveness. For instance, a recent study conducted jointly by Nielsen Media Research and the Cabletelevision Advertising Bureau demonstrates that viewer ad recall decreases by 45% when the number of Television ads shown during a commercial break is moved from three or fewer ads, to seven or more. Internet advertisers and Web site designers, we believe, would be wise to assume similar facts hold for the interactive arena, too.
Even in cases in which the number of ads being served per page is relatively low, as in the case of Lycos (which limits the number of ads per page to two), there is the separate question of whether advertising effectiveness (on a per ad basis) is diminished when ads from multiple advertisers are displayed on the same page.
Results from early testing of New York Times Digital's now well-known Surround Sessions concept, in which a single advertiser's message is presented throughout the usage session, lend support to the idea that Internet ad campaign performance and ROI can be significantly enhanced when limitations are placed on the number of different advertisers whose ads are displayed per page. Specifically, initial results from Surround Sessions show that presenting the user with advertising from a single advertiser can help Internet campaigns achieve four times the lift in the areas of Message Association and Purchase Intent, and close to three times the lift for Brand Awareness and Brand Favorability, as compared with sites that run ads from multiple advertisers per page. Surround Sessions have been much discussed for being a media buying innovation. Clearly, Surround Sessions also represent a web site design innovation, the likes of which will be far more prevalent in the Internet Advertising landscape of the near future.
Finally, recent research conducted by Dynamic Logic indicates that a fourth design factor - site clutter (the number of distinct Web site design elements - colors, text areas, navigational elements, etc. - per page) - also plays a critical role in determining the effectiveness of Internet Advertising campaigns.
Although Dynamic Logic has so far been unable to formulate specific design benchmarks that would enable site designers and advertisers to estimate the likely impacts of a given site's design on the performance of advertising served from that site, Dynamic Logic has succeeded in demonstrating that there is a real connection between user-perceived site clutter and ad effectiveness. Specifically, Dynamic Logic's research shows that sites that are perceived by users to contain less site clutter generate materially higher Brand Favorability and Purchase Intent for their advertisers.
The strategic view
Based on an analysis of the available research, our present position is that there are principally four Web site design factors that impact overall advertising performance:
1. Ad Placement (location of advertising on the page)
2. Ad Clutter (number of ads per page)
3. Ad Clutter (number of advertisers per page)
4. Site Clutter (number of Web site design elements per page)
Advertisers and publishers, we believe, should use an understanding of these factors to direct their advertising and site design objectives. Our general strategy recommendations for advertisers and publishers are as follows:
Use Dynamic Ad Placement: Advertisers should seek to advertise, wherever possible, on web properties that support the dynamic placement of ads. Publishers are advised to design sites that leverage Dynamic Ad Placement technologies.
Reduce ads per page: Leading Internet Advertising measurement and intelligence company, AdRelevance tells avant|marketer that, at the present time, ad-supported Web sites average between 2.5 and three ads per page. Therefore, we advise advertisers to buy media on sites that display between one and two ads per page, as - all else being equal - this is likely to contribute to superior advertising ROI. Publishers should seek to tailor sites to fall below this average. The exception to this guideline is the case in which all ads being served on a given page are from a single advertiser (see below).
Reduce advertisers per page: Advertisers should seek to buy ads on sites that enable the advertiser to be the exclusive advertiser on pages on which the advertiser's ads appear; publishers should seek to re-arrange how they serve and sell advertising media in order to make this possible.
Reduce site clutter: Although no comprehensive guidelines as to how to assess site clutter to optimize ad performance have yet been formulated by any research organization, generally, we believe, it is prudent for advertisers to seek out sites that have a minimum variety of page elements, as understood in terms of colors, navigation schemes, fonts, and related page elements. Our sources at Lycos also suggest that it is prudent for advertisers to seek out sites that use consistent design elements across all site pages, as this has been observed to increase advertising effectiveness. Publishers are advised to seek to minimize design complexities, as such, on their sites, and - wherever possible - should seek to employ usability firms such as Eyetracking, Inc. to determine the precise impacts of site clutter at their sites on advertising effectiveness.
Ajay Segal is Managing Editor of avant|marketer. He may be reached at
. Click on "Send Input" (below) to send us your publishable input on this article.
|
|
|
|
