Commentary

Media Metrics: The Reader Is the Message

  • by July 22, 2008
Back when TV was young, the visionary communications theorist Marshall McLuhan noted: "When faced with a totally new situation, we tend always to attach ourselves to the objects, to the flavor of the most recent past. We look at the present through a rearview mirror. We march backwards into the future."

For an advertiser in today's fragmented media landscape, reach and frequency are such objects; they have a nostalgic value, being the traditional metrics of audience exposure to the advertising. For the past five decades, these metrics and their derivatives, the ratings points, served as media's gold standard. They were the common denominators for evaluating campaign performance across different media channels and schedules. They directed billions of dollars in advertising spending.

As marketing plans started to include online placements, advertisers sought a similar measurement. Once comScore and Nielsen came up with digital R/F estimating tools to show what audiences new media placements bring to the table, the amount of online spending increased exponentially. With a familiar object now appearing in their rearview mirrors, advertisers could start their march into the future.

But now, years later, many of us are still facing backward, holding on to reach and frequency, totems of the television age. Advertisers using digital cannot rely on reach and frequency alone to project the value of digital advertising relative to other communication channels. By doing so, they restrict their ability to capture the whole picture of exposure to online advertising - and by and large, they can only understand a fraction of what happens after an ad appears online.

Estimating online media reach turned out to be quite different from doing so for television or radio campaigns. Unlike with TV programs, simply knowing the site's audience numbers at the time when the ads are running is not enough to project the reach of online advertising. After all, Web sites can only serve ads when users request new content; thus, the amount of ads served is directly proportional to the pages being requested and viewed by individual users. In short, few visitors ever see all the ads served by a given site.

Eventually, the industry accepted site page views as a proxy for ad impressions. For example, the number of times the CNN.com front page was viewed last month is used to estimate how many banner ads can be placed on that page in the future. To project the number of people exposed to those ads, digital reach and frequency equations account for how often people visit the site within a certain time frame and how many pages they request per visit.

The page-view based models, however, have a sizable blind spot: Using publishers' page views as a proxy for ad impressions captures the exposure to search placements and to streaming video. But Google or Yahoo Search page views would not tell you how many searchers will see your particular paid search listing. Neither does broadband video fit into the models based on page views: It comes in streams, not pages, and the number of ads served within a stream depends on the number and duration of streams.

Presently, the commonly used R/F tools, based on comScore or Nielsen metered panel data, can at best project only a portion of the reach of a typical digital media buy. Ostensibly, new variables, such as time spent on site or video stream duration, can be added into equations to extract the "eyeballs" count from users' interactions and perusal of the new advertising channels. But it becomes harder and harder to bypass the fundamental difference in advertising delivery between analog and digital media: Unlike in TV or radio, an online campaign's ability to reach specific audiences is often independent of the content near which the ad appears.

While reach and frequency numbers are the beginning and the end of traditional media planning, online they are just the starting point. Using sites' registration databases and cookies, digital marketers can target their ads to serve just the right number of impressions exclusively to their desired segments. The individual sites' overall audience compositions and historical audience numbers are thus much less significant to an online marketer than they might be to a traditional marketer.

What's more, the ratings-points approach is not just antiquated but can be detrimental to an online campaign's efficacy. Focusing just on reach and frequency limits a digital campaign's ability to serve ads based on specific audience characteristics other than demographics, such as previous purchases and online behavior. In the end, these narrower criteria may be more effective in predicting future purchase behavior than demographics.

It is also shortsighted to adhere to sites' historical stats alone when planning social media campaigns. New technology has empowered users to manipulate media, and as a result, advertising content gets picked up and passed along, altered and redistributed far away from its original placement. In social media, the relationship between the original media placement's audience and the combined audience for the advertising is very loose, as the ads can have long and productive lifecycles beyond the initial media placements and campaign flight dates.

As consumers' relationship with media becomes more active, the eyeball count will become less important. After all, with each user acting as an editor, slicing and dicing though available content, modifying and disseminating it, he or she sees or creates a somewhat different picture, anyway. McLuhan would probably have described this social media phenomenon by altering his most enduring remark: "The reader is the message."

Advertisers are slowly waking up to the reality of media without a gold standard of reach and frequency. Although a plethora of new and telling metrics is available, no single metric would ever serve as a universal currency for media value across media channels and industries.

The advertisers have no choice but to assess the relative roles of all available measurements and to create their own exclusive currencies informed by the unique economies of their specific products and brands. For example, direct retailers should concentrate on media-attributable visitor traffic, conversions by product type and revenues. CPG advertisers should monitor search behavior, attitudinal response to ads, consumer interactions with their brands' online entities and ads, pass-along behavior, word-of-mouth, and online sentiment. At the same time, they should match media exposure to purchase behavior (as measured by offline panels or databases) to assess the impact of advertising on sales. For most advertisers, evaluating media investments in the digital world requires working with a number of measurement and tracking vendors and developing proprietary systems to electronically monitor consumer media and purchase behavior.

Of course, no matter what media you're dealing with, it is important to know how many potential customers the campaign has reached. But now it is more important to get a holistic picture of what happens in a media system as a result of advertising. This means recognizing individuals' spontaneous responses to advertising, tracking users' manipulation of ads, and understanding how people influence each other. We're already well into our march into the future. It's time to start facing forward.

Yaakov Kimelfeld, Ph.D., is vice president of digital research and analytics director at MediaVest USA.

Next story loading loading..