When is a click not a click?
It's an urgent question pitting Web sites against their advertisers. Fortune's Devin Leonard explains why answering it won't be easy.
By Devin Leonard, Fortune senior writer

(FORTUNE Magazine) -- The other day, Yahoo said it had developed another cool technology for Internet advertising: a technique for collecting "traces" of paths its users take, without the need to record any personal data on them.

That way, without invading anyone's privacy, it hopes to sell targeted ads that are a step beyond the kind where you type a phrase like "fishing rod" into Google or Yahoo and get a bunch of blue links for sites like compactfishinggear.com. Now if Yahoo knows you are in the market for fishing gear, it can send you a jazzy display ad from L.L. Bean.

The old dial-up service plays catch-up by going free, but Fortune's Stephanie Mehta argues the plan's unlikely to result in big gains. (Read the column.)
The company took a beating last month, but CEO Terry Semel tells Fortune's Adam Lashinsky that he likes its chances against Google, YouTube, and MySpace. (more)

It's the kind of innovation we've come to expect from the likes of Yahoo (Charts) and Google (Charts). Promising to deliver ads that consumers want and to bill advertisers only when their messages are received, they are often praised for transforming advertising from a black art into a transparent, measurable business.

No wonder advertisers increased their spending on the Internet by 30% last year, to $12.6 billion, according to the Interactive Advertising Bureau. They are rebelling against old media, which seem drab and unmeasurable compared with the web.

Marketers, increasingly unhappy with Nielsen's television program ratings, want television commercial ratings. They have their doubts about magazines and newspapers. Can anybody blame them after the scandals at Gruner & Jahr and Newsday that forced the publishers to restate their circulation numbers?

But there is an irony here: Although the Internet may be the most measurable of advertising media, advertisers and Web sites are actually having huge battles because they can't agree on what they should be counting. If Google, Yahoo, AOL, and Microsoft's (Charts) MSN want to see continued double-digit spending increases, the two sides are going to have to resolve some fundamental issues.

Consider, for instance, the click. The simple click of the mouse is the currency of "search advertising" - the aforementioned compactfishinggear.com example - which accounted for $5.1 billion of last year's Internet ad-spending binge. You would think it would be easy for advertisers and sellers of web ads to agree on a common definition of something so fundamental.

But there are valid clicks, and there are invalid ones. Advertisers argue that they should be billed only for clicks that bring customers to their Web sites to shop around. They don't want to pay for clicks made by accident or by, say, malicious competitors trying to make them exhaust their marketing budget.

They also worry about unscrupulous Web site operators who click on the ads on their own sites - sometimes employing computer "bots" to do the job - so they can put more money in their pockets. But sellers of web advertising say it's not easy to distinguish the good clicks from the bad.

As Shuman Ghosemajumder, Google's business product manager for trust and safety, wrote in a recent blog: "Our servers can accurately count clicks on ads, but we cannot know what the intent of the clicking user was when they made that click."

Nevertheless, the IAB, which represents search engines and web publishers, recently announced it was forming an industrywide working group including Google, Microsoft, and Yahoo to "provide a detailed definition of a 'click' and the standard against which clicks are measured and counted, including the identification of invalid clicks and/or fraudulent clicks."

Here's why the IAB's members are so interested in click definition: The search-advertising business has had some scandals of its own lately. Advertisers complain that Google and Yahoo haven't done enough to protect them from invalid clicks.

In some cases, they say, it's competitors who are driving up their costs, but there are all kinds of ways that unscrupulous users can use clicks. Let's say you pay Google $50 for an ad that will stay up for ten clicks. If someone who doesn't wish you well knows about it, it's an easy matter for him to make your ad vanish. All in all, some consultants say, as much as 30% of all clicks on the web are fraudulent.

Google says that number is wildly inflated. But it recently agreed to a $90 million settlement in a class-action lawsuit brought by advertisers claiming that they were charged by the company for invalid clicks. Yahoo has agreed to pay $4.9 million to settle similar charges. (AOL, a division of Time Warner (Charts), which owns Fortune's publisher, was also named as a defendant in the Google case. It hasn't settled.)

Yahoo and Google say they are putting new technology in place to sniff out phony clicks. Yahoo is hiring what it describes as a "full-time traffic-quality advocate" to act as a liaison to advertisers. But none of this means much if the two sides don't have a common definition of a click.

Even when the Web's ad sellers and buyers have agreed on measurement terms, there are still big squabbles. Consider what is happening with display ads. In 2004 the IAB and major advertisers hammered out a definition of an "impression"--a term for when someone sees a display ad. They agreed that websites should bill advertisers only when a display ad has been received by a user's browser. That way, advertisers could be fairly certain that customers had gotten their messages.

But both sides come up with different numbers when they count those ads too. Advertisers complain that they are often charged for ads that didn't fully load. How do they know? They look at numbers from third-party ad servers, such as DoubleClick, which they hire to deliver their ads to Internet sites.

Joel Lunenfeld, vice president of media services at Moxie Interactive, an Internet ad agency, says he has seen discrepancies from 20% to 100% in the numbers from the two sides. Web site people blame their customers' ad servers for the confusion.

Recently a group of big advertisers, including PepsiCo, Hewlett-Packard, and Kimberly-Clark, demanded audited numbers from Web sites by next year. They also want common measurement standards by 2007. "We think it's important, because the new standards will eliminate questionable reporting," says Pepsi spokeswoman Nicole Bradley. "Now we will have confirmation that our ads have been fully delivered."

Sounds like a job for Nielsen. The company measures the unique users of Web sites - those who visit at least once a month - using a panel of 30,000 Internet surfers with software installed on their computers that records their activities. But that's an old-media sample in a new-media world.

IAB CEO Greg Stuart says his group's members are better equipped to count their own ads. "On the Internet, you can literally have a single graphic ad served a billion times," he says. "Ad servers can measure that. It's the virtue of the medium."

Perhaps the best Pepsi can expect is for an independent auditor like Ernst & Young to review web publishers' technology and the processes used to count ads. Web sites hope that will be enough to persuade advertisers to accept their numbers and quit using those from the third-party servers. But will big spenders like Pepsi be so acquiescent? It doesn't seem likely.

Eventually the IAB hopes to come up with a definition for unique users. It should. Magazine publishers - like Fortune - that are extending their business onto the Web tout their unique-user counts to lure web marketers just as they use circulation to attract magazine advertisers.

But funny things happen with those numbers too. Nielsen recently restated Entrepreneur magazine's Web site numbers for April from 7.6 million to two million after concluding that "urls within the entrepreneur.com domain had not been user requested" - meaning that it was counting things like pop-up ads as additional visits to its site.

Strange things go on in traditional media as well. But if you are a magazine publisher and you want to boost circulation by distributing free copies to tanning salons, you have to disclose that to the Audit Bureau of Circulation. There's no such oversight on the Web, but there will be soon. Yahoo, Google, and the rest will have to get used to stricter measurement policing just like their old-media brethren. There's only so much anyone can do to transform the advertising business.  Top of page

YOUR E-MAIL ALERTS
Follow the news that matters to you. Create your own alert to be notified on topics you're interested in.

Or, visit Popular Alerts for suggestions.
Manage alerts | What is this?
Market indexes are shown in real time, except for the DJIA, which is delayed by two minutes. All times are ET. Disclaimer LIBOR Warning: Neither BBA Enterprises Limited, nor the BBA LIBOR Contributor Banks, nor Reuters, can be held liable for any irregularity or inaccuracy of BBA LIBOR. Disclaimer. Morningstar: © 2014 Morningstar, Inc. All Rights Reserved. Disclaimer The Dow Jones IndexesSM are proprietary to and distributed by Dow Jones & Company, Inc. and have been licensed for use. All content of the Dow Jones IndexesSM © 2014 is proprietary to Dow Jones & Company, Inc. Chicago Mercantile Association. The market data is the property of Chicago Mercantile Exchange Inc. and its licensors. All rights reserved. FactSet Research Systems Inc. 2014. All rights reserved. Most stock quote data provided by BATS.
Market indexes are shown in real time, except for the DJIA, which is delayed by two minutes. All times are ET. Disclaimer LIBOR Warning: Neither BBA Enterprises Limited, nor the BBA LIBOR Contributor Banks, nor Reuters, can be held liable for any irregularity or inaccuracy of BBA LIBOR. Disclaimer. Morningstar: © 2014 Morningstar, Inc. All Rights Reserved. Disclaimer The Dow Jones IndexesSM are proprietary to and distributed by Dow Jones & Company, Inc. and have been licensed for use. All content of the Dow Jones IndexesSM © 2014 is proprietary to Dow Jones & Company, Inc. Chicago Mercantile Association. The market data is the property of Chicago Mercantile Exchange Inc. and its licensors. All rights reserved. FactSet Research Systems Inc. 2014. All rights reserved. Most stock quote data provided by BATS.