Seeking ad measurement standards

By Published on .

With so many different definitions out there for how to measure the success of online advertising, business-to-business marketers are finding it hard to compare themselves with competitors.

The Internet advertising community is finally starting to set some standards. However, because complying with the industry groups is voluntary, the groups can merely endorse terminology. But companies such as ABC Interactive, Schaumburg, Ill., are enforcing measurement standards with their clients.

A beacon

Four years ago, the Coalition for Advertising Supported Information & Entertainment started to offer direction for measurement. The organization was founded in 1994 by the Association of National Advertisers and the American Association of Advertising Agencies, both in New York, to guide the development of interactive advertising and marketing.

"Basically what we wanted to try to figure out was, `What is it going to take to build this new medium?' " said Robin Webster, co-executive director of CASIE and senior VP of the ANA.

"One of the first areas we focused on was measurement," she said. "Marketers and their agencies are responsible for justifying their expenditures. We want to know what we're getting for our money."

So CASIE came up with 13 "Guiding Principles of Interactive Media Measurement." They are "a beacon, or a goal, of when we'll be happy," Ms. Webster said.

"What we want [to measure] is simple," she said. "It's the `opportunity to see.' " Simply put, the OTS means the ad has been delivered and is present if the user cares to pay attention to it. It is the yardstick used in most other forms of advertising media.

But there are glitches, of course. "You've got a lot of funky learning to go through," Ms. Webster said. "This is the first medium, to my knowledge, that doesn't give you a proof. If it was a magazine, you'd get a tear sheet."

Then there are ads that the industry wants counted that are not, due to caching. Internet services use cache memory to store data for popular Web sites. When one of these sites is requested by a user, the service does not have to go back to the Internet, and the ad server, to find it. Instead, the site is retrieved from cache. This causes undercounting within server-side measurements, which generally measure the requests the server receives for an ad.

There are also ads counted that the industry would prefer are not•those canceled by users before the download is complete.

Plus, however well documented CASIE's principles are, following them is voluntary, and CASIE does not promote specific ad-counting methodology.

More direction needed

Industry folk clearly still think Internet ad measurement is important and needs work.

Over the past three years, Ms. Webster has polled ANA members about why they did not buy, or spend more money on, online advertising the previous year. Sixty-eight percent of respondents in the 1998 and 1999 surveys ranked "no proof of return on investment" as their No. 1 reason. No. 2: "lack of reliable and accurate measurement information," according to 56% of respondents in 1998 and 58% in 1999.

So companies such as ABC Interactive have taken up the reins. ABC Interactive audits Web sites, search engines, e-mail delivery systems, chat rooms and Internet broadcasters. It is a unit of the Audit Bureau of Circulations, also of Schaumburg.

This spring, ABC Interactive adopted four ways to count online ad activity, specifically the OTS. These definitions recognize that online ad-delivery technologies can measure ad activity at different points in the ad-serving transaction: during ad insertion, ad download, ad request or ad display (see "Standard definitions," this page). Each report that ABC Interactive creates now looks at one of these counting methods.

The auditor needs four definitions, said Evelyn Hepner, VP-marketing and sales for ABC Interactive, because a variety of counting systems are in use today. "We need to be able to provide these audits for our clients to the level that they can provide the data," she said.

ABC Interactive developed the the definitions to serve its clients, but to ensure agreement on the terms, the company worked closely with CASIE and the Future of Advertising Stakeholders (FAST), a virtual group of advertisers, advertising agencies, online media companies and technology enablers dedicated to speeding up development and broad use of digital advertising.

"This industry is only going to move ahead," Ms. Hepner said, "if we have comparability."

Need to compare

Jim Spaeth, president of the Advertising Research Foundation, a New York-based industry nonprofit group, and chairman of the FAST Measurement Committee, agrees. "Clearly what the industry needs is one set of standards," he said, "and they need comparability; otherwise, it's very hard to have a rational marketplace. There are just too many different estimates floating around."

But comparability has a few different aspects to it, said Mr. Spaeth. The first, apparent in ABC Interactive's definitions, is there are two fundamentally different kinds of online ad measurement.

One is user, or browser, based; the other is server based. In both cases, the objective is to measure clicks and basic impressions.

And within those two kinds of measurement, Mr. Spaeth said, different services are attempting to measure these clicks and impressions, often with varying definitions and measurement standards of their own.

"This is all a recipe for chaos," he said.

FAST thinks, like CASIE, that online ad measurement should be compared with the traditional media standard, the OTS. So FAST has come up with a recommendation to solve the three big measurement problems--caching, varying counting methods and comparability--at once.

Mr. Spaeth describes it as "a way of sending a little message, or code, back to the server to let it know that there was an ad that was counted." The code is written into the ad itself. This method has been referred to as "tagging," but Mr. Spaeth prefers to call it a "cache-busting technique." He expects FAST to finalize the method recommendation this fall.

"The only negative," said Mr. Spaeth, "is, in some cases, it means a little bit of investment, a little bit of work, and some effort on the part of industry."

FAST hopes to create a way to measure aborted ads; to reconcile audit standards, which are stricter, with general measurement standards; and to bring "rich media"--streaming audio, video or both--measurement into the fold, all with one intention, Mr. Spaeth said: "less confusion for the advertiser."

But, Ms. Webster said, "the industry still hasn't gotten there by a long shot."

Most Popular
In this article: