Web Informant #115: 22 June 1998:
The death of banner ads

Other than web site owners who collect revenues from advertising on their sites, I don't know very many people that like banner ads. And the latest reports are an encouraging sign that their death might be imminent. It couldn't happen fast enough for me.

• Banners waste bandwidth: many of us turn off images on our dial-up accounts just so we can get the "meat" of the pages we access without wasting time. We are starting to see programs that can selectively filter out banners, such as WRQ's @Guard and others. This will only reduce the overall audience for banners.
• Banners aren't memorable: my own informal polls show little or no recognition of them. Magazine publishing and TV broadcasting would die a quick death if this were the case. There are several theories as to why but I think the best explanation has to do with the changing surfing patterns of web users. Early web users tend to click around and visit many new sites out of curiosity. Banners might be memorable for these users and under those conditions. But as my own informal polls show, as users gain experience, they tend to surf less and get less patient with trying to find information. Who has the time? This makes it hard for banners to stick out.
• Banners often take you places you don't want to go. If I am looking for something, clicking on the banner takes me out of the site and away from the content I want. Think of having to change channels on your TV to VIEW a commercial: that would be a disaster for broadcasters. Several vendors such as Narrative and First Virtual Holdings have tried to correct this problem by placing the entire purchase information directly inside the banner, so you never leave the page you are viewing. These are somewhat intriguing, but these banners take even more bandwidth to load.
• Revenue from ad banners depends on users clicking on the banner, and the ad server recording this click in its logs. Yet recent studies (such as the one mentioned in Interactive Week last week) have shown clickthrough rates plummeting, in some cases half of what they were a year ago. And a combination of using the BACK button in a browser and technologies such as caching and proxy servers confound these logs. There is a bright spot, though: the purchase-enabled banners mentioned above tend to increase clickthrough rates.
• Banners have few sizes and standards. Think of 6.3 second TV or radio commercials or odd sizes for print publications. Where would these industries be without standard sizes? Yet on the web we have everything from rectangles to squares to full-page ads. No two sites are alike it seems. This makes it harder for ad creators obviously, since they have to create different banners for different sites.
• Banners are only useful on pages where users actually visit. It seems elementary, but many sites have little-seen pages, or have traffic patterns that are hard to predict. This often results in many sites having pages that don't have banners on them, what our industry amusingly calls unsold inventory. So placing banners means a solid understanding of traffic patterns by a media buyer. That is wishful thinking, as many media buyers are young with little technical background. The airlines "yield management" programs (that charge continuously-changing discounts for little-used seats) would be worthless under the current web conditions.

Taken together, these trends make for tough times ahead for banners. And I predict that in another year we'll see fewer and fewer of them on our pages. Web advertising is a billion-dollar business, according to the Internet Advertising Bureau. Yet it is a slippery slope: should we add together a dollar for an ad that Microsoft bought on Netscape's site to the dollar for an ad that Netscape placed on Microsoft's? Did any real money actually change hands?

So if banners die off, where will all the real money currently invested in them go? How about towards sponsoring better sites? Maybe it is wishful thinking on my part, but all of us would benefit: we'd save on bandwidth, improve the content, and deliver goodwill and customers to the sponsors.

David Strom
david@strom.com
+1 (516) 944-3407
back issues
entire contents copyright 1998 by David Strom, Inc.
Web Informant is ® registered trademark with the U.S. Patent and Trademark Office