New ads: 'Rich concepts, not rich media'

As the industry works to hammer out new types of Internet ads, a basic tension is emerging: Advertisers want larger, more complex spots to grab viewers’ attention. Web site publishers don’t.

A study commissioned by 10 major advertisers concluded that bigger ads were more effective than simpler banner ads, with consumers getting the main message one-third of the time they saw full-screen ads but only 16%of the time they saw banners.

Yet the bigger ads require bandwidth. And Jupiter Communications Inc. in New York estimates that more than three out of four home surfers will access the World Wide Web through dial-up connections -- a maximum of 56Kbit/s -- through 2002.

America Online won’t run large, technically complex ads that take too long to download or “look like junk” for people with slow connections, says Bob Pittman, president and chief operating officer of America Online. “We don’t let our ‘art monsters’ ... make stuff people can’t see,” he said. About 20% of AOL subscribers are still running connections slower than 28.8Kbit/s.

“When bandwidth gets here, it’s not going to be a problem anymore,” said Mike Donahue, executive vice president at the American Association of Advertising Agencies in New York. “But bandwidth isn’t going to be here for a while.”

At an Internet advertising summit last month hosted by Procter & Gamble, major manufacturers, site producers and advertising specialists created an organisation aimed at boosting the Web as an advertising medium. One of Fast Forward’s four main tasks is devising new advertising models that everyone in the industry will accept along with banners and buttons.

The committee has pledged to come up with proposed online ad models by November.

It is too early to predict what the models might look like. In general, “we’re going for rich concepts, not rich media,” says Donahue, a member of Fast Forward’s steering committee. That means looking for innovative ideas but not necessarily cutting-edge technology that many consumers might not be able to access. “Some of the best ideas can be done in 12K,” Donahue said.

But in terms of what the new ad types will accomplish, several speakers at the Procter & Gamble summit cautioned against expecting Web ads to translate into immediate click-through sales.

“You start your brand building in the window-shopping phase,” Pittman says.

Still, measuring an ad’s effectiveness was very much on the minds of summit attendees. When asked to rank a half-dozen issues surrounding online advertising models, they put return on investment first.

A growing number of experts say banner ads are less effective among consumers and will become less desirable among buyers, lending some urgency to the search for new ad models acceptable to Web publishers and advertisers.

At the Software Development ’98 conference in Washington earlier this month, for example, author and former Sun Microsystems Inc. engineer Jakob Nielson says, “users are completely ignoring banner ads. Click rates are falling through the floor.”

According to Evan Neufield at Jupiter Communications, statistics on banner ads can be read as either good or bad news. Although 20% of consumers have no interest at all in clicking on banners, another 28% click at least once in every 10 Web sessions (and 51% click rarely).

“We would say the banner is not a home run but no strikeout either,” he said.