• by Paul Gough , August 6, 2002

You’ve got a client who has a product perfect for 18-to-34-year-olds. You know you need to include the Internet in your media plan because this demographic spends a lot of time online, but it’s a big Web and you’ve got a small budget. You’ve got to find the best sites and figure out how many people you want to reach and how often.

Welcome to the often-confusing, ever-changing world of online reach and frequency. This isn’t your parents’ R&F. Until the Internet, reach and frequency were simple concepts. Offline, reach is how many different people are exposed to a particular ad at least once during a four-week period. Frequency is how many times someone has been exposed to the ad during the four weeks.

The Web’s more complicated. An advertiser can buy 10 million ad impressions on a site, but that doesn’t mean 10 million people are going to see the ad.

An Advertising Research Foundation committee recently defined online reach as the number of different people exposed at least once to an Internet advertising message during a four-week period, beginning with week one. The committee defined frequency as the average number of times a person has been exposed to an advertising message over the same period. The definition sounds similar, but tracking online R&F is still a problem for most advertisers, even though it seems that the Web is tailor-made for every metric imaginable.

"It’s been hard getting reach/frequency [data] for new media in [a format] media planners are used to using day in, day out for traditional media," says Gian Fulgoni, chairman of comScore Networks. "They’ve had to say, ‘We don’t know,’" says David L. Smith, president of Mediasmith Inc. in San Francisco and chairman of the ARF committee on Online Reach and Fl_miency. Advertisers need empirical data on effectiveness, he says, and that’s something that just wasn’t available.

So how can you pitch and sell your advertiser on interactive media when you don’t have the reliable data on reach and frequency that’s available for every other medium? And how do you adjust the plans if you have nothing but intuition and anecdotal evidence to go on? "It’s very clear to me that the online industry wants and needs the capability to do media planning using reach and frequency," says Doug Knopper, VP/GM at DoubleClick.

Well, things are changing. There’s a growing awareness that the old ways don’t work anymore, particularly with advertisers who are more interested in branding than in counting clicks or total sales. Those advertisers are looking for a gain in awareness and an overall increase in sales. And they want to know how many individuals in their target audience were reached, and how many times. That’s a different strategy than the one that would be used by a company that can measure ROI on sales tied to the ads. You can’t use the amount of traffic on a particular website to determine reach or frequency, because the number of impressions is meaningless without context. Companies have been figuring out new ways to track information about a specific campaign (usually by counting cookies) along with demographics and how Web users behave. Atlas DMT, comScore, NetRatings (p.15), DoubleClick, and others are developing programs, or pieces of them, that help plan an Internet campaign, forecast effectiveness, carry out the campaign, and then measure its effectiveness in reach and frequency. The companies vary in approach but not in goals.

The standard questions about reach and frequency — how many people will see the message and how many times each — weren’t answerable with any reliability. Studies on websites’ claims found they weren’t accurate, because traffic measurement isn’t linked to reach and frequency.

Moreover, ad server logs can’t tell you whether site visitors are male or female, or how old they are. And they can’t dig further to give the demographic data that brand advertisers crave: How well did the campaign do reaching working parents, males 18-34, or any other target audience?

Demographic information is available from third-party providers who survey consumers’ activity on the Web. The combined data helps you determine whom you’re reaching and, as the technology develops, it allows finer and finer levels of detail, or "granularity."

"It allows us to slice and dice" the data, says Young-Bean Song, director of analytics at the Atlas Institute in Seattle. "That level of detail and forecasting of reach in their target area wasn’t available before."

That makes it gold for brand advertisers. For instance, comScore, has a database of 1.5 million Web users who have given the company permission to monitor their behavior online. The company tracks not only what sites they visit but also whether they buy anything online, what they buy, and what they subscribe to.

Panel data can add an important factor to demographics: How much users are spending online. All of this is a huge leap from what’s been available before, and it helps interactive media catch up to what is standard in other media.

But all of this isn’t going to do much good for planners if it can’t be compared to reach and frequency data for traditional media. Some think all this work will be in vain without that comparison, which will isolate online media and make it little more than a novelty.

DoubleClick’s Knopper believes the first step needs to be a standardized metric, or at least some kind of consensus so that the figures can be considered accurate in the offline world. Having eight companies trying to figure out their own way won’t do that, he says. "The results of reach and frequency have to be comparable to the offline world," Knopper says. "That’s a very important and critical component."

He believes data from all media in the plan — TV, radio, print, newspapers, online — should be intricately linked to compare results.

"We want to be able to compare the results of your [online] campaign and look at its place in the overall media plan," Knopper says. "You want to be able to plan and assess all your media, not just one medium."

It goes beyond reach and frequency."Reach and frequency is just one tool to plan media," he says. "It’s not the most important tool, it’s not the least important tool." Which tool is best for planning depends on your goals, he says.

"This is a work in progress," Smith says. He and other analysts are in the process of evaluating products, but there’s no long-term data beyond the theory. "Right now, vendors are lagging behind the demand," Smith says.

Smith counseled planners and buyers to experiment with the products and become knowledgeable about the products and strategies that will meet their needs. Some planners might find that they need to adjust their buys, increasing or decreasing Gross Ratings Points.

"Then they know whether they’re overbuying or under-buying," Smith says.

It’s key, because some brands want frequency and others care more about reach. With these new tools, they know how they’re covering their websites.

"You’re really trying to figure out the optimal and most effective media buy," Knopper says.

comScore’s Fulgoni says it’s high time online reach and frequency was developed to its full potential.

"I think the reason it hasn’t happened [until now] is that it was too easy to produce other measures" like click-throughs, he says.

"Early on, [the online industry] was spoiled. They went down the wrong path. They were seduced into thinking that direct response and the click was the metric that would drive the whole buy. The bubble burst, the pure online money dried up and went away. They were left with no boat. Now we’re scrambling to catch up."

But if everything can be worked out technologically, the developments could make the online advertising future bright.

"We’re entering the next phase. I call it the end of the beginning. But it’s going to be very different," Fulgoni says. "As those tools become available, I think we’ll see a rejuvenation of online advertising."