The story of advertising on the Web is, in a sense, far more important than that of content. Long before schemes to milk the Web for dollars came as regular as the second hand on a Timex, the content people believed; they believed there was a growing, literate audience on the Web that was willing to click. The challenge of Web publishing wasn't easy then, and still isn't, but it was straightforward. Explaining the simple mechanics of the Web to advertisers, however, has proved somewhat less elementary. And from the day the first dollar changed hands in exchange for a link, the case for hypertext has been argued time and again, employing increasingly arcane and superfluous claims.
The latest salvo from the proponents of the obvious comes in the form of studies proving that "ad banners build brands." Berkeley Systems, America Online, the Internet Advertising Bureau, and HotWired each conducted research essentially aiming to demonstrate that people remember online ads. Some of the reports even added claims of increased purchase interest. The methodology was not dissimilar to what you might devise if forced to create one at gunpoint: Show people ads, quiz them later, compare to control group. The results are clear and positive, though by no means radical: Somewhere around three to five out of ten people remember seeing a given advertisement. Why wouldn't they, when the ad was only about 3-5 inches from their faces?
But there'll always be good business in ritual contrarianism toward all things Web. In last week's Adweek feature on the studies, Forrester media analyst Bill Doyle was fingered to provide requisite technoskepticism, contending that "the Web is not a good place for branding." Revealingly, Doyle's criticism had little to do with branding per se; it merely reiterated many of the most resilient misconceptions about the Web. "There isn't enough reach among the demographic groups major advertisers want," Doyle is quoted as saying. "The lack of bandwidth and slow modem speeds at home greatly limit the animation capabilities of most online ads. Most Web sites have poor or negligible measurement systems in place to track users. Production headaches abound."
The demographics assertion would surely raise eyebrows at Intel and Toyota, and the idea that branding efforts require animation should come as a surprise to Conde Nast's ad-sales team. But the measurement comment is most discouraging; you could build a Web site's usage-tracking system with Wrigley's and curly straws and still offer more accurate measurement than radio, television, and print combined! How many years of pinpoint-precise impression, clickthrough, flowthrough, and attrition stats will it take for this cosmically obvious proposition to sink in?
Brand studies are best understood as icing on an already fattening cake constructed to appease link-buyers upset to learn that, on average, only 2 percent of users are actually clicking on their teasers. In an ideal world, rather than convincing advertisers that webvertising clicks even when users don't, the message would be far simpler. There'll always be more links than clicks. Advertisers everywhere should rejoice that - for the first time - whether their fishhooks get seen, swallowed, or rejected, they'll feel the tug. The ones who are successfully dealing with this knowledge aren't wasting time damning the medium; they're building better links, better ads, or better brands.
This article appeared originally in HotWired.
