In this story: Web Revenue Slowdown Slap in the Face Do Banners Work? Standards Problems

Those were the days. “The Internet is a transforming technology,” gushed a New York Times article on the impact of the Web on the advertising industry written only a little over two years ago. “Between it and the frenetic pace of the digital revolution, advertising will never be the same.”

Perhaps not. Today, talk of transforming technology and digital revolution rings hollow when conversations about Web advertising focus on cash burn rates and “poor visibility”—the embattled Web CEO’s new euphemism of choice for telling investors that nobody has a clue where the business is going.

The entire Web advertising industry is on the ropes, and with it, the many Web content providers who were counting on advertising profits. Almost all are desperately cutting costs and scrounging for revenue. Cases in point:

• On April 12, DoubleClick, the industry’s largest and most successful Web advertising network, announced that its first-quarter 2001 network revenue fell 23 percent over the year-ago quarter to $114.9 million. For the full year, the firm expects revenue in its Media division to be down 45 to 55 percent. The problem: “softness in online advertising.”

• On March 21, DoubleClick competitor 24/7 Media reported a net loss of $779 million on revenue of $185 million. “No one could have anticipated the enormous challenges the Internet advertising industry has faced in the past year,” CEO David J. Moore griped. “The poor visibility within the industry has caused company after company to continually lower guidance, redefining the term ‘rock bottom.’ ”

• On April 16, New York Times Digital, the New York Times’ online operation, reported a 9.8 percent decline in revenue to $14.1 million and a 23 percent increase in operating losses to $7.7 million. The problem: weak online advertising.

• On April 11, Web blue chip content provider Yahoo announced that its first-quarter revenue fell to $180 million from $231 million a year ago, and its net income fell to $8 million from $61 million. The once-aloof firm, long one of the few profitable Web content sites, reportedly is eagerly wooing advertisers and cutting ad rates while chopping its staff and hunting for opportunities (sometimes in unsavory places) to boost subscription revenue.

According to the Industry Standard’s Layoff Tracker, almost 13,000 people in Web businesses dependent on advertising have been laid off in the past year (see chart “Pink Slips for Web Ad Workers”). Three thousand and fifty-two were sent packing at Internet marketing and advertising firms, including hundreds each at leading companies such as 24/7 Media, Avenue A, Beyond Interactive, DoubleClick, and Engage. 3,921  employees at portal sites such as AltaVista, Ask Jeeves, Britannica.com, and NBCi got pink slips, and 5,694 more were fired at a long list of media and entertainment sites, including Bolt, CNBC.com, CNET, CNN Interactive, Primedia, Red Herring, TheStreet.com, Walt Disney Internet Group, and even, as dutifully noted by the Industry Standard, two layoffs of about 100 people total at the Industry Standard.

Web Revenue Slowdown

Oddly, the carnage on the street is not immediately obvious in industry statistics. For example, recent advertising reports from the Interactive Advertising Bureau (or IAB, which was called the Internet Advertising Bureau until last month and widely regarded as the preeminent supplier of Web advertising revenue statistics) shows a slowdown but not a collapse in Internet advertising revenue.

In the third quarter of 2000, Internet advertising revenue fell slightly, to $1.986 billion, down $138 million or 6.5 percent from $2.124 billion in the second quarter of 2000 (See chart “Web Ad Revenue Hits The Brakes”). This was the first sequential quarterly fall in Web ad revenue since the IAB, an association of Web publishers, began its tracking efforts. The IAB emphasized that, year-on-year, Internet ad revenue still rose 63 percent.

In the fourth quarter of 2000, Web ad revenue growth resumed, according to the IAB, albeit at a snail’s pace by Web standards. Total ad revenue was up 9 percent over the third quarter figure, to $2.2 billion, and 22 percent on the year-ago quarter. Because the fourth quarter has generally been by far the industry’s strongest quarter, the results for subsequent quarters will likely be lower, perhaps even negative.

A Hard Number to Get

Another key measure of the industry’s health–average ad rates–is also deceptively strong, at least if the two industry firms that track these numbers are to be believed. AdRelevance, a data tracking subsidiary of Jupiter Media Metrix, asserts that the average cost of a full banner ad in the fourth quarter of 2000 fell to $25 per thousand impressions (CPM), a relatively modest decline of $5 from the $30 mean banner CPM in the second half of 1999 (see chart “Average Banner Ad CPM”). Engage’s AdKnowledge subsidiary reports an even more modest decline in banner ad CPMs, from an average of $33.75 in December 1999 to $33.64 in September 2000. (AdKnowledge has yet to publish a fourth-quarter report.)

Both AdRelevance and AdKnowledge base their CPM rate figures on “rate card” figures provided by publishers, but publishers and advertisers alike know that few if any advertisers, whether on the Web or in any other media, actually pay rate card rates for their advertising. “I dare you to find a single person in the industry that goes by those numbers,” says Tig Tillinghast, an industry writer and columnist who spent many years as a media buyer and manager at the interactive divisions of agencies such as Leo Burnett, J. Walter Thompson, and Anderson and Lembke. Such rates are just a starting point for negotiations. The actual rate paid varies with the extent of commitment and the amount of clout an advertiser has with a publisher.

In print advertising, a 30 percent discount off the rate card is typical. The key question, of course, is how big that discount is for Web advertising. Unfortunately, publishers never disclose this figure because it would weaken their bargaining position with advertising agencies, and the agencies don’t reveal it because their contracts with publishers forbid them to do so. “It’s a hard number to get,” explains Marc Ryan, director of media research at AdRelevance. “We would collect it if people would give it to us.”

While nobody publishes actual, or “effective,” ad rates at present, various observers offer estimates. For example, in a December 2000 report, Forrester asserted that $8 is the effective banner ad CPM, roughly a 70 percent discount off the average rate card CPM. A February report on Web advertising from Morgan Stanley Dean Witter is more pessimistic, pegging the effective CPM at $3.50, a hefty 88 percent discount.

The general consensus seems to be that the average effective banner ad CPM today is somewhere in the neighborhood of $10. “The average CPM is now $9 to $11 after negotiation,” says Tig Tillinghast. “But the effective CPM is lower because of overdelivery.”

“The big differential between rate cards and actual rates probably shows that people are getting a little more desperate for advertising,” Marc Ryan notes, adding that CPM figures don’t take into account the huge amount of barter and pay-for-performance Web advertising.

Meanwhile, it appears that effective CPMs are still headed down. Peggy O’Neill, an advertising analyst at NetRatings, says that preliminary data from a new research program that tracks effective Web advertising rates indicates that the effective banner ad CPM has dropped 3.5 to 6 percent between the first and second quarters of this year.

Silver Lining?

Given that ad rate CPMs appear to have plunged around 70 percent, it is remarkable that total Web advertising spending, as measured by the IAB, has continued to grow. One explanation of this anomaly—the “silver-lining-in-every-cloud” explanation—is that tremendous growth in total paid ad impressions has all but compensated for the catastrophic fall in the CPM rate.

There is, however, a more ominous possibility. Why would virtually every enterprise associated with Web advertising be in severe financial distress if total Web advertising revenue is still growing, albeit slowly? Perhaps the answer is that Web advertising revenue has in fact fallen much further than the IAB’s numbers suggest. Until the IAB releases its first-quarter figures, this question will have to remain unanswered.

Many prognosticators are quite gloomy. A recent report from Merrill Lynch, for example, forecasts a 25 percent decline in Web advertising revenue in 2001.

Supply and Demand

Although observers disagree on the severity of the collapse in Web ad rates, there is widespread agreement on the root cause of the plunge: a simple imbalance in supply and demand. On the one hand, supply, in this case the portfolio of Web sites eager to take advertising, has risen continuously for the past six years (see chart “Ad Space Supply Is Up”). Indeed, it is likely that not only the number of Web sites taking advertising but also the depth of ad-supported sites has grown, boosting the inventory of Web ad space even more.

Ironically, because of the way that advertising is sold, the huge expansion in the percentage of the population on the Web and the amount of time that people spend on the Web has actually worked against Web publishers. More people spending more time on the Web means more “page views” and therefore more ad space inventory (see chart “Page View Surplus”). “Particularly over the last five years, the number of people coming online, and the time they spend online, has skyrocketed,” explains Jim Nail, a senior analyst at Forrester Research. “This has created a huge growth of page views and therefore ad impressions.”

On the other hand, demand for that exploding space has not kept pace. In particular, the dot-com implosion that started in the spring of last year wiped out many of the most enthusiastic—or reckless—Web advertisers and erased big chunks of the marketing and advertising budgets of most of the rest. In today’s chilly financial climate, surviving dot-coms have done just what traditional firms do when things get tight: chop discretionary expenses, of which advertising is very high on the list.

Slap in the Face

Inevitably, as dot-coms disappear from the Web, traditional brick-and-mortar advertisers grow in importance. The AdRelevance division of Jupiter Media Metrix, a Web market researcher, says that dot-com firms represented 69 percent of all Web advertisers in January 2000 but fell to only 52 percent of the total a year later. Nielsen/NetRatings, yet another Web researcher, asserts that as of February, dot-coms represented 41 percent of new Web advertisers but only 35 percent of new Web ad impressions (see charts “Dot-coms Disappear”).

Nevertheless, traditional advertisers have not rushed to fill the big vacuum in the Web ad market created by vanishing dot-coms. Indeed, this vacuum was inevitable because traditional advertisers were hardly going to boost their budgets just to compensate for the advertising cash that the dot-coms poured down the drain. “People lost sight of the fact that people who do advertising have budgets, and within a range, it’s a finite budget,” explains Ken Marlin, managing director of Veronis, Suhler, and Associates, a New York investment and merchant bank that specializes in media deals. “All these new channels don’t change the budget. They allocate a set amount to advertising.”

Instead, traditional advertisers are watching with glee as Web ad rates collapse. “Advertisers are getting back at the dot-com arrogance,” says Nick Nyhan, president of Dynamic Logic, a New York firm that markets tools for measuring Web ad effectiveness. “I can see why they might feel that way. ‘Hey, you’re not worth more than GE and GM combined. So we’re going to pay you what we think we should have been paying you all along.’ It’s a slap in the face.”

Bryan Russiano, director of e-marketing for New York-based interactive ad agency Novo, agrees. “The collapse of dot-com business models, that took away probably half the revenue these Web sites were getting. Now they must rely more on traditional advertisers—and most of these advertisers have savvy agencies. No longer can a Web site get all kinds of silly money from a dot-com that is just trying to build traffic. In the last six months, we have been able to leverage that for the benefit of our clients. It’s more a buyer’s marketplace than a seller’s marketplace.”

Sinking Click-Throughs

While all agree that traditional advertisers are now warming to Web advertising, their embrace has hardly been passionate. Indeed, there is widespread agreement that the Web has deep problems that must be overcome before mainstream advertisers will accept it as a mainstream advertising vehicle. First and foremost, many advertisers are skeptical that Web ads work.

Ironically, the biggest factor in this uncertainty is a Web ad metric that once was billed as absolute proof of Web advertising’s effectiveness: the click-through rate. When the first Web advertising space was marketed, salespeople emphasized that, unlike broadcast or print advertising, the response to a Web ad could actually be measured. “We have the first truly accountable advertising medium,” boasted Rick Boyce, an ad salesperson for HotWired, the Web site that launched the idea of Web advertising in 1994. “We can literally count each consumer that responds to the ad banner with a click-through to the advertiser’s Web site.”

True enough and, at first, many Web surfers did click through, often as many as 4 percent of all who viewed an ad, or more. But the novelty soon faded, and the number of  click-throughs sank. By 1997, the average click-through rate dropped below 1 percent, according to Nielsen/ NetRatings, and today, it hovers at around 0.4 percent—one click-through per 250 people who view an ad (see chart “Click-Through Crash”). And this is for home users; at work, the click-through rate is just above a meager 0.2 percent, or one per 500 viewers.

Instead of proving the effectiveness of Web advertising, the click-through rate wound up proving its ineffectiveness, to the dismay of the industry. “When they initially sold online ads, the fastest way to get money was to sell it as different than offline advertising,” explains Nick Nyhan of Dynamic Logic. “It was crack cocaine at the beginning of online advertising: a way to get people to take money from TV and spend it on a little banner ad. It was the one differentiating factor: ‘It’s better because it’s got this click-through rate.’ ”

Come a Cropper

Both Web sites and ad agencies were complicit in peddling click-through rates as advertising’s manna from heaven. “People like me … promised to clients that interactive stuff would be so great,” says Tig Tillinghast, who was a media buyer on the interactive side of various agencies at the dawn of the Web advertising era. “We didn’t quite know how, but we knew it would somehow be more valuable and better. We got people to start sending money. This money initially came out of media research budgets, not marketing budgets: literally, it was experimentation. But now we’ve come a cropper: are we actually offering that value? Someone with fiduciary responsibility has to say: no.”

But at the beginning of the Web advertising era, fiduciary responsibility was a scarce commodity. As long as the money was flowing, thanks to the infinite generosity of the dot-coms and their venture capital backers, falling click-through rates did not matter much.

Of course, in Biblical fashion, what went around has now come around, inducing considerable schadenfreude among those on the sidelines.

After the well ran dry and Web sites had to plead on bent knee with traditional advertisers for support, the weakness of the click-through rate as an advertising metric became glaringly apparent. “Now the industry is saying, ‘Geez, this is awful. The thing we used to hang our hat on can’t let us hang anything,’ ” says Nick Nyhan. “That is where the industry is right now.”

Do Banners Work?

With click-throughs in the basement and Web advertising in a slump, a great debate has been triggered: does Web advertising actually work? Or, more precisely, how well does Web advertising work, and for what purposes? These, of course, are key questions because the answers ultimately will determine the value of—and price of—Web advertising and therefore the fate of advertising-supported business models on the Web.

Some think Web advertising doesn’t work, at least in its current incarnation. “The first generation of Web advertising was all about accountability and proof of performance, and in fact a lot of it proved to be not that effective, with exceptions,” says Richard Notarianni, director of media planning at DDB Worldwide. “The environment is somewhat rough, the creative is limited and weak.”

Notarianni scoffs at banners. “The current model of Web advertising was, ‘Let’s make the most innocuous unit possible, the banner, that does not offend the audience,’ which was the UNIX coder. "It's like this peace, love, and freedom hippie B.S. comes out. Oh man, this is not about commercials, this is about ideas, man!" These guys dominated, and they were very aggressive about this, so they made a unit that was the least advertising-oriented possible. And we all act surprised that it doesn’t work well! If you had designed a TV ad that was not designed to attract attention or glamour it would not have worked either.”

Notarianni argues that the banner is intrinsically doomed to fail because it does not mesh well with the mindset of the Web surfer. “I’m proactively using the medium, and this banner is encouraging me to do something else. [It’s as if] I’m busy reading this magazine and now you want me to answer the phone. That doesn’t make any sense.”

Notarianni does not argue that the Web is by its nature a poor advertising medium—quite the contrary. It is just that the banner ad does a very poor job of exploiting the Web’s capabilities. “The medium’s potential is mind-numbing, but the attempts at delivery to date have been kludgy or sort of apologetic,” he explains. “We created an advertising environment in which it was a shame to be an advertiser. We still wonder why it doesn’t work.”

Notarianni argues that Web advertisers and publishers must develop Web advertising that is better aligned with the active nature of Web use, that somehow is inserted into the context of that activity rather than distracts from it.

Doug Jaeger, interactive creative director at TBWA\Chiat\Day in New York, agrees that banners are limited as an advertising tool, although not so categorically as Notarianni. “You should use banner advertising like outdoor advertising or an envelope to deliver a simple message,” he says. “It’s about being at the right place at the right time. Banner ads aren’t for all advertisers. A banner can deliver news or information … [They’re] good for mass awareness of simple ideas.” But, of course, most advertising must deliver a lot more than simple ideas.

Paying for Performance

One ironic consequence of the lack of faith in banner effectiveness is the emergence of new “pay-for-performance” Web advertising payment models that exploit the medium’s unique measurability—not to mention its current abject poverty—to reduce or even eliminate any risk on the advertiser’s part. Instead of paying for ad impressions, many advertisers now demand to pay by the click (“cost per click”) or even by the “conversion” of a customer who has clicked into a sale (“cost per action”).

Such deals are anathema to Web sites, which fear that pay-for-performance advertising will drive ad rates even lower and leave them vulnerable to ill-executed advertising campaigns over which they have no control. “There’s an enormous gulf between marketers and media companies,” says Jim Nail of Forrester. “The marketers say, ‘Hey, I don’t want to pay on anything but a performance basis.’ I can understand: it’s a great tool to manage the budget. But the media companies say, ‘Wait a minute, why do I have to take the risk that your ad is any good, your product is any good, that you know what you’re doing? That’s not my job; my job is to bring you a good audience.’ ”

Nonetheless, pay-for-performance advertising has become increasingly common. Forrester Research estimates that pure performance-based and hybrid ads, which combine a pay-for-performance structure with some sort of fixed CPM, represented 33 percent of all Web ads in 1999. In 2000, that figure grew to 56 percent, and Forrester estimates that such ads will make up 62 percent of the total this year (see chart “Advertisers Flee the CPM”).

Despite its apparent inevitability, ad agencies report continued resistance to the pay-for-performance concept. Tig Tillinghast reports that he sometimes turns a site’s sales pitch against it in an effort to cut a pay-for-performance deal. “Sites say, ‘We have the perfect media for you. It’s uniquely well suited,’ and so on. I come back to them, ‘If you have such confidence that I’m underappreciating your targeting, then you should be very comfortable with a cost-per-action deal.’ You see clouds of dust as they back out of the room.”

Branding versus Direct Response

A recurring theme in the debate over Web advertising effectiveness is the question of branding versus direct-response advertising. Many observers agree that advertisers, agencies, and publishers alike are confused about the difference between these two very different classes of advertising, and this in turn has led to a tremendous waste of Web advertising dollars.

The problem, some say, is that the Web is intrinsically a direct-response medium, but has been used, or abused, as a branding medium. “The first model that Web agencies acted on was a brand ethic. Finally, someone figured out to use the disciplines of the direct-response industry, which is suited for one-to-one marketing,” says Richard Notarianni of DDB. “At the end of the day, advertisers coming to the Web haven’t had a clear sense of their objectives, haven’t understood the direct connection. If the advertisers had a better idea of the specific benefit they are seeking, they would do better. You see the same problem with television and magazine advertising: people run a brand ad and count responses.”

Guy Forestier-Walker, a vice president at Beyond Interactive, an interactive agency in New York, agrees that Web advertisers have been slow to understand the direct-response character of Web advertising. “Cost per acquisition, cost per sale, the lifetime value of customers: these are traditional direct-marketing concepts. It took awhile for the industry to realize that the Internet works just the same way as traditional direct-response marketing. American Express did it for us 40 years ago. All we have to do is apply the model.”

Odd Discrepancy

One interesting measure of the Web advertising industry’s confusion over branding versus direct marketing is a discrepancy between what advertisers think their ads are intended to do and what independent market research firms think the ads do. According to ad networks such as DoubleClick and 24/7 Media, 80 to 90 percent of all Web advertising is direct-response advertising.

But, according to market researcher AdRelevance, 61 percent of the Web advertisements it tracked in January of this year were branding-oriented (see chart “Branding vs. Direct”). Marc Ryan of AdRelevance explains that others in the industry who track the split between branding and direct response advertising base their figures on what the advertisers themselves say, but AdRelevance actually looks at the ad to determine its focus. Even if an advertiser regards an ad as a direct-marketing ad, Ryan says that it is counted as a brand ad if it “works better as a branding ad.”

In short, if an advertiser can’t tell what its Web ad is for, is it any wonder that Web advertising is not as effective as it could be?

Lazy Marketers

Jim Nail of Forrester thinks that the advertising industry’s delayed comprehension of the direct-response nature of the Web is partly a reflection of the people who have been involved in Web advertising. “There are not really that many direct-mail people in the industry,” he says. “It has shocked me that the direct-marketing side of the world hasn’t gotten onto it more quickly. It puzzles me. It’s an opportunity.”

Even many of those who have used the Web as a direct-response medium have not done it right, in Nail’s opinion. He believes that instead of blaming banners, they should blame themselves for the ineffectiveness of their advertising. “[Banners] are not the reason that online advertising is going through its Death Valley period,” he argues. “Marketers haven’t learned how to use them.”

The key, Nail says, is attention to detail. “To make online ads work, it takes a lot more care and feeding. I’m an old direct-mail guy, it’s very much like direct mail. You try something, you test it, you don’t go right out the door with 20 million impressions. If it doesn’t work, you look for other sites, other offers, you give people a reason to click through. You tweak the variables to get the results. Marketers are lazy.”

Even when an advertiser and its agency fully understand what their direct-response Web ad is for, they can get in trouble from a brand standpoint. Tig Tillinghast says that as interactive media director in a traditional agency, he was once called on to develop a direct-response Web campaign for a high-profile luxury goods company that would drive surfers to the sites of the client’s retailers. At first, the banners had a “high-touch” look befitting the client’s lofty image, but over time, Tillinghast says, the ads came to look more and more like “Kmart specials” as they were continually modified to improve click-through. Finally, the ads were killed when everyone realized that they were “doing horrible irreparable harm to the brand.”

Using the Wrong Data

While some say the Web cannot be used effectively for branding, many others disagree. They believe that Web brand advertising may have failed to an extent because advertisers have used the medium improperly, not because the medium is intrinsically flawed.

One common cardinal sin has been to use click-through rates, which measure direct response, as a measure of a brand ad’s effectiveness, even though the goal of traditional brand advertising is to subtly build a favorable attitude that at some point in the future pays off with a purchase, rather than to evoke an immediate response.

“Advertisers have long thought the only way to get value out of online advertising is by looking for click-through, especially if you’re a brand like Kelloggs or Nike or P&G,” says Marc Ryan of AdRelevance. “But you don’t expect someone to buy shampoo through an ad. There has been confusion that click-through is the magic number to measure effectiveness, especially in packaged goods. But what they really want to know is what will bring the product to the top of mind.”

“I think people early on threw money at online advertising,” says Sherry Szydlik, senior vice president of marketing at PrimaryKnowledge, a New York-based firm that provides interactive ad tracking services. “Because they had data—click-through rates—to make optimization decisions, they did. But they shouldn’t have, because it’s not the right data.” Often, she says, effective Web ads have low click-through rates while ineffective Web ads generate high click-throughs. “There can be value hidden in a low click-through rate,” Szydlik claims.

Doug Jaeger of TBWA/Chiat/Day puts it another way. “When you look at a McDonald’s billboard on the highway, do you count the cars that go by and don’t turn? It’s a mistake to look at [Web ads] as solely direct response.”

Nick Nyhan of Dynamic Logic argues—not disinterestedly, given that his firm has developed technology for measuring Web ad brand effectiveness—that it is absurd to assert that the Web is useless for branding. “If less than 1 percent of ad impressions are clicked on, then it looks like 99 percent of the ad campaign is wasted,” he says. “But common sense tells you that an ad that isn’t clicked on has some value. Maybe I buy the Coke later, not now.”

CPMs Shoot the Moon

While advertisers and their agencies can be blamed for their confusion about how to best use the Web, much of the blame for the failure of the Web to blossom as an ideal advertising medium, particularly for brand advertising, can be pinned on the Web sites themselves. To start, the Web advertising price model was absurd, especially in light of the inability of its proponents to deliver on their bold effectiveness-tracking claims.

This absurdity was apparent from the very beginning. “The first CPM rate ever published was by the HotWired folks and Rick Boyce,” explains Tig Tillinghast. “I met him in Chicago when I was at Leo Burnett, and he told me that the site’s CPM was $150. I asked where that came from. Rick was very honest and shrugged his shoulders. That’s where things started.”

Tillinghast says that the two kinds of content providers that initially emerged on the Web, lifestyle sites and technology sites, both followed the same strategy: “shoot the moon with the prices.” And, says Tillinghast, “In the absence of any precedents, why not? It was very rational.”

Peggy O’Neill of NetRatings suspects that these initial astronomical ad rates were driven not by any particular consideration of value or demand, but rather by a more pressing issue: the desire to inflate the perceived value of the site to the financial community, no doubt to accelerate an IPO and its accompanying riches. “I think ad rates were initially set for Wall Street,” she says.

Whatever the cause, the Web’s high ad rates, while scarcely an impediment for cash-rich pound-foolish dot-coms, were a serious problem for traditional advertisers, who were intimately familiar with and concerned about the cost of other media. “If you go back to rate card rates, AdKnowledge says the average CPM is in the $34 range,” says Jim Nail of Forrester. “But a 30-second Super Bowl spot has a $29 CPM. When you say to advertisers, ‘You must pay more for a banner ad than for a Super Bowl commercial,’ it just doesn’t work.”

Inadequate Research

Even advertisers and ad agencies that believe in the Web’s potential as a branding medium complain that, for a long period, Web publishers did little to prove the effectiveness of online branding.

“Every company feels it has to adapt to the Internet, but at the end of the day, they need to see value for the dollars they spend,” says Nick Nyhan of Dynamic Logic. “Big traditional advertisers got pitched by online agencies every day. ‘Spend more money online,’ they were told. The advertisers said, ‘Why? Who is going to click on an ad and buy toothpaste, or a Cadillac?’ ” Now, Nyhan says, “The online ad community has woken up to the reality, we have to show value beyond the click.”

Doug Jaeger of TBWA\Chiat\Day says that publishers are “moving toward the model of being consultants, to help you use their Web site and meet your objectives in terms of lead generation or brand awareness.” He agrees, however, that Web sites have not done as much as they could. “We haven’t gotten a lot of help. [The sites] have some general pointers, they can show what has been effective on their site and what hasn’t. But in the end, you have to trust your own conclusions. It’s still a wide-open frontier.” He says that there are exceptions, for example CNET. “They ask, ‘What’s your objective? We want to help you meet that objective,’ ” Jaeger says. “They’re trying to be helpful.”

Predictably, banner-skeptic Richard Notarianni is scornful of the Web sites’ efforts to demonstrate the effectiveness of online advertising. “They have this feeble research to show that banners are effective advertising tools,” he snorts.

Even when sites have provided research to advertisers, it has seldom been in a form they are comfortable with. “They [traditional advertisers] were not given things in the way they were used to getting them,” says Nick Nyhan. “They were being asked to change the way they make decisions. You can’t expect a big old company to change itself.”

Marc Ryan of AdRelevance agrees that the sites need to do more to help advertisers. “Web sites need to offer unique solutions for their advertisers: value-added services such as designing interesting sponsorships or helping clients show results,” he explains. “Many sites forget to do this, especially when they default to click-through rates. They need other metrics to prove that this ad or sponsorship has helped to increase a brand’s top-of-mind awareness. They should just be creative: the new IAB standards open up lots of opportunity to experiment, to make progress in getting advertisers better bang for their buck.”

Ken Marlin of Veronis, Suhler, and Associates notes that the handful of sites that have learned to work with advertisers are generally more successful. “The winners are still able to get good prices, the Yahoo Finances, the CNETs, the Expedias,” he says. “They are the market leaders in their niche. They have their own sales forces, people who are able to talk intelligently and professionally to the person buying the advertising about the characteristics of the people looking at the site, who they are, how long they stay, their demographics.”

Meanwhile, Marlin says, smaller sites that cannot afford a dedicated sales force have a chicken-and-egg problem: they contract sales out to third parties that sell all kinds of ads, but the third-party salespeople don’t understand why to sell on one site versus another, so they sell on price, which only intensifies the financial problems of the sites they represent.

Standards Problems

Advertisers and ad agencies also take publishers to task for not managing a vast and expanding array of Web advertising standards issues better. To start, there is controversy over the physical size of Web ads. Although the IAB established a set of eight standard banner and button ad sizes in 1996, even today many sites use non-standard banner sizes, which means that advertisers who use those sites must redesign their banners to advertise on these sites (see chart “Web Ad Confusion”).

Worse, there are dozens of rich media technologies in use today, most of which are supported by only a handful of sites (see chart “More Web Ad Confusion”). Any advertiser that strays beyond basic GIF, JPEG, and HTML formats may find that its ad has very limited distribution.

“The problem now is that you have to produce so many versions of your creative that it’s more expensive to produce your creative than to make your media buy,” says Doug Jaeger of TBWA\Chiat\Day. “An Enliven ad can cost $30,000.”

While they wish there were fewer non-standard ad sizes and formats, many advertisers are also frustrated by the absence of more standardized advertising formats. Five years—an eon in Internet time—passed before the IAB issued seven new standard ad sizes this February. These ads, most of which are larger vertical “skyscraper” and horizontal rectangular ads (see example), were intended to address the limitations of the banner by providing advertisers with a larger, more arresting canvas to work with.

Still, many in the advertising world feel that the new ads were slow in coming, especially in light of the obvious problems with banner ads. “I admit we made a mistake not to get bigger earlier,” says Richy Glassberg, chairman and CEO of Phase2Media, vice chairman of the IAB, and chairman of the committee that developed the new ad formats. The IAB does not intend to be years late with standards for rich media: the organization has already formed a committee to address rich media issues such as ad tracking and reporting, file sizes, and quality assurance.

A Dollar Short, Too

Others argue that the IAB is not only a day late, but also a dollar short. “We need not just more creativity, but also some technical innovation,” says Richard Notarianni of DDB Worldwide. “We need to reinvent approaches and strategies here, not just do a better ad. It’s an inherently flawed medium that we’re trying to make less flawed, instead of throwing it out and starting over.”

Notarianni views the new IAB ad formats as mere imitations of passive magazine advertising. “It’s amazing that we’ve run out of ideas already in a medium that is five years old and are going back to what we know: print ads on the Web. I think we can do better than that.”

The Web advertising standards problem extends beyond the format and technology of the ads themselves. One issue is business terms and conditions. While standard “insertion orders” have long been used to handle the ordering of a print or broadcast ad, only in February did the IAB propose a similar set of Web advertising terms and conditions.

Ad tracking and measurement is another problem. Because of the many complex ways in which an ad may make its way from the advertiser’s or agency’s server to a third-party server to an ISP proxy cache to the Web surfer’s browser, there are multiple points along the journey where that ad can be counted, each yielding a different final count of “ad impressions” for which the advertiser is ultimately charged.

“Some sites count ad requests,” says Leslie Laredo, a longtime Web ad sales person who now runs the Laredo Group (Newton, MA), a Web advertising training and consulting firm. “Others count ad deliveries.” Still others count ads actually rendered. Meanwhile, a multitude of additional factors distort ad tracking: Should ad requests generated by Web spiders and bots be counted? What about pages rendered from a computer’s local cache? What about refresh pages? The IAB is working on ad tracking, too.

Over and above these advertising-specific issues are problems intrinsic to the Web and to computers that are beyond the control of either Web publishers or advertisers. “The biggest challenge is the disconnect between what you can do in a TV ad and an Internet ad,” explains Doug Jaeger of TBWA\Chiat\Day. “You have to worry about bandwidth and processor speed [with Web advertising], but all TVs are the same. TV is a very easy platform to come up with ideas for. But when you get to online experience with a mouse pointer and things you click on to deliver info from a server in the ether, it is difficult to conceptualize the consumer experience.”

Systemic Bias

Just as advertisers and their agencies blame Web publishers for many of the problems plaguing Web advertising at the moment, Web publishers have some gripes to air about advertisers and agencies. In particular, many feel that traditional agencies and their clients have been very slow to understand, let alone take advantage, of the Web.

To some extent, this is no surprise. Until very recently, traditional advertisers and agencies had very little to do with the Web, instead leaving the Web advertising business to the dot-coms and their dot-com advertising agencies, firms like Avenue A, Engage, Modem Media, Organic, and Razorfish. Even when traditional clients did launch a new advertising campaign with Web components, a traditional agency would typically handle advertising in traditional media while an interactive agency would take care of Web advertising.

For example, Nancy Johnston, VP of professional services for Fullmoon Interactive, an interactive agency in Los Angeles, says that BBDO is Dell’s traditional ad agency, but her firm handles Dell’s interactive ad business. “We try really, really hard to integrate, but it’s tough,” she says.

Only in the past year or two have traditional agencies started to take the Web seriously: most have now formed interactive divisions, and some have snapped up struggling interactive agencies. But even if a traditional agency has acquired or created an interactive division, that does not mean the agency as a whole is up to speed on the Web. “There is a struggle within agencies, too,” says Tig Tillinghast. “There is often a ‘systemic bias’ within groups.”

As a result, the interactive group within an agency can be almost as isolated as it would be were it an independent firm. It may wind up operating all but independently, creating Web advertising for a client that has little or no relationship to the campaign the agency has created for other media.

The Glamour Factor

Financial considerations can be an issue, too: the fees that advertising agencies make on a banner or two pale compared to a thirty-second TV spot. “Agencies are in business to make money,” says Fullmoon Interactive’s Nancy Johnston. “They’re going to recommend things that make money.”

Tig Tillinghast agrees, “One of the things that clouds our judgment is the profitability of the media. So media selection goes to the highest-margin media, which is usually not online.”

Another deterrent to Web advertising from an agency’s point of view is the hassle of Web media buying. “It takes two and a half times as much labor to buy online [than to buy space for the equivalent dollar value of print or broadcast ads],” Tig Tillinghast says, thanks to the multiplicity of sites and the absence of standard practices. Done properly, Web advertising also requires more labor after the buy because Web ad campaigns can and should be constantly tweaked.

Then, there is a glamour factor: top “creatives” would much rather work on a complex, interesting television or print campaign than design a little 468 × 60 banner. “They’d rather be shooting a TV commercial in Monte Carlo,” Scott Schiller, a senior VP at Walt Disney Interactive Group, recently told the Industry Standard.

Another barrier to the full embrace of Web advertising within traditional agencies is political, Tig Tillinghast believes. “There’s an interesting dynamic,” he says. “Traditional ad guys are twenty years older, so the online people can’t play as an equal in office politics. This puts online at a disadvantage.”

Old Guys Can’t Live Forever

Still, the agencies are changing or at least they say they are. “There are barriers between interactive and traditional advertisers, but the smart agencies are bringing those together,” says Richard Notarianni of DDB Worldwide. “The old guys can’t live forever. The young people who are bringing this in, even the not-so-young people who are bringing this in, are coming to the table with ideas that you can’t resist for very long because they’re smart.”

“Even at the top, they’re accepting this,” says Notarianni. “The line people accept it, the top accepts it, but the recalcitrant people in the middle who want to preserve their comfort level are resisting. But that was last year; most people now are using the Web and understanding it. As they integrate it into their own lives, they won’t have as much problem with it as a marketing vehicle.”

Media directors at other agencies also assert that they are committed to the Web. “All media directors should know both traditional and online,” says David Song, interactive associate media director at Arnold Worldwide in Boston. He jokes, “Online media is media, too.”

Clearly, the agencies have a long way to go, but progress is being made. “Fifty years of television buying has to be broken down,” says Richy Glassberg of Phase2Media and the IAB. “It’s going to take awhile.” He notes that the IAB has been running training seminars on Web advertising for people in traditional agencies for three years.

Ad Dollars Follow Eyeballs

Between the collapse of the dot-coms, the start of what may be the worst recession the advertising industry has seen in 20 years, and the various and numerous missteps of both Web publishers and Web advertisers, it might seem that the Web as an advertising vehicle is doomed.

Web publishers are flat on their backs financially, their business models shattered by the collapse in ad rates and their bankrolls dwindling as they desperately launch subscription schemes that have no chance of compensating for the ad revenue they so badly need (see below table). The ad agencies that have the power to divert a huge stream of advertising dollars to the Internet are indifferent to its plight because of a combination of self-interest, the absence of convincing proof that Web advertising is effective, and internal paralysis.

Yet few in the publishing and advertising industries believe that the Internet will not emerge as a powerful advertising medium in the very near future. While resistance to the Web is entrenched, mighty forces are gathering to drive Web advertising forward.

An old cliché—advertising dollars follow eyeballs—is perhaps the simplest way of explaining these forces. With the Web now occupying a large and growing percentage of the average person’s time—often at the expense of other advertising vehicles—advertisers have no choice but to turn to the Web to peddle their wares. That means advertisers must and will find solutions to the numerous but far from insuperable technical, marketing, financial, and corporate obstacles that have hindered the full development of the Web as an advertising medium.

With that strange temporary hallucination known as the Internet boom now receding into history, signs abound that Web publishers and advertisers alike have finally achieved the sobriety required to fully grasp what must be done to make the Web work for advertising. The publishers now know that it is no longer good enough to talk about why the Web is great for advertisers; they must give advertisers what they need to make Web ads work, be it creative new ad formats, more realistic rate structures, or disciplined research that demonstrates advertising effectiveness. The ad agencies, on the other hand, are realizing that even after the great crash, the Web is for real, and they have no choice but to learn how to use it, even if that means a radical change in their business practices and corporate organization.

In addition, Web publishers need to learn how to compete effectively with rival media. The industry is young and naïve, softened by its easy dependence on dot-com revenue, but is doing battle with other media that have honed their competitive weapons in decades of bitter competition with each other. While newspapers, magazines, and broadcasters all have heavily funded industry trade associations that crank out reams of research that members can use as tools to fine-tune their products or persuade advertisers of their medium’s superiority, Web publishers have only the little IAB, which until February, when it hired its first employee, CEO Robin Webster, was entirely a volunteer organization.

No More Wild, Wild West

David Song’s advertising career is a crisp encapsulation of the experience of the Web advertising industry. Before moving to Arnold a few months ago, Song worked for an interactive agency in Boston that, like all interactive agencies, handled a lot of dot-com business. “It was like the wild, wild west,” says Song. “We could do whatever we wanted and get away with it.”

For example, Song says that he would buy space on a hundred sites. “You would never do that in traditional advertising, buy a hundred TV stations.” Or the agency “would pay $100 CPMs for stuff that didn’t work, and now we pay $20,” he admits. “We spent a lot of our clients’ dollars recklessly. Why did we get away with it? Because the Internet was new, and the clients thought we knew.”

After the Web collapse last spring, however, Song says the world changed. “We had to be a lot more careful with client dollars. We really had to understand what the Internet should be used for. We redefined Internet advertising: it’s not a separate rogue medium, it’s just part of an overall communications plan.”

“Looking back,” Song reflects, “it makes me feel like ‘Wow, I was a kid, and I just grew up.’ When you’re a senior in high school, you think you’re the greatest, and now it’s, ‘Well, whatever.’ We made terrible mistakes,” he confesses cheerfully. “We were young and reckless.”

In a bland remark of the sort that never was heard a year or two ago, Song says of the Web, “It’s like TV, radio, and print: it contributes to sales of a client’s products.” Bland though it may be, Song’s remark is a perfect reflection of a new attitude in Web advertising and indeed in the Internet industry as a whole: the party’s over and it’s time to get back to work.

Next month in Content Intelligence:

Web Ad Resurrection: A Chastened Industry Moves to Claim Its Rightful Market Share. Shell-shocked Web publishers are learning the hard way what it takes to win market share in the brutally competitive media market. The key lesson: listen to your customers. That means a lot of things, including new ad formats, a new focus on effectiveness research, and new sales strategies. But will these be enough to turn leaden business models into gold?