• by Carl Langrock , May 29, 2007

Part of the allure of interactive advertising is that it's measurable. In fact, marketer demand for accountability is driving an increasing portion of ad dollars to interactive. But the stepped-up use of interactive also has affected offline ads. Now, offline often serves to ignite the relationship. In fact, it's difficult to find an offline ad that doesn't have a URL pushing consumers to branded Web sites.

As online transactions establish new peaks, marketers must tie online activity back to the offline budget. Failure to make this connection devalues offline advertising and drives more budget into interactive. For many, making this connection remains a challenge.

In general, agencies use three effective techniques for relating online activity back to the offline spend. I call the techniques proximity, deduction and micro-focus. Agencies use proximity when spikes of Web site activity closely follow broadcast spots. Deduction is used when offline activity is so constant that the spike-to-spot relationship is blurred. Micro-focus adds precision when media drives consumers to URLs dedicated by channel.

TV commercials pushing people to a Web site can generate breathtakingly fast results. When Gotham launched Go Daddy, the agency only got credit from the client for checkouts within 10 minutes of the spot airing. If instead of a Web site, the ad urged people to call an 800 number, the agency never would have expected the majority of callers to complete checkout so quickly. But Go Daddy's target customers are often on a computer while watching TV - apparently even during the Super Bowl.

For agencies, sharing the advertiser's data is the key. But data sets vary by product, which means that agencies need to develop a common denominator that most clients can produce - home page views, sales, unique visitors, whatever. For example, one agency shares the advertiser's results data by time period. This data feeds their response analytics system which, in turn, ties the activity back to media. The process overlays response patterns in ways traditionally seen with 800 number activity.

Advanced agencies develop success metrics for each media outlet. The combination of results and media data yields cost per acquisition by outlet and by time period. To optimize the buy, agencies identify under-performing outlets and shift weight accordingly.

Advertising volume can obscure the relationship between a specific spot and online response: If you're constantly on air, how do you tell how much traffic each spot generates? Enter deduction. To paraphrase Sherlock Holmes: If the answer is hidden within the data, eliminate all that is not the answer; whatever remains must be the answer.

For Earthlink/People PC, Omnicom's Direct Partners tracks online applications vs. media activity. Tracking across periods produces clear evidence that sales correlate with TV weight. The agency developed formulas for ascribing a percentage of online activity back to the broadcast schedule. This reduces TV's relative cost per lead. Without this adjustment, TV's portion of the spending mix might be reduced; sales would suffer. Direct Partners uses its media response analytics system to combine media with sales and customer lifecycle data.

Tied back to media, the lifecycle data shows relative value of acquisition by channel. Direct Partners' process has reduced customer acquisition cost while maintaining volume. Long-term quality and stability in the results are evidence that the tracking methodology works.

Marketers also employ microsites - or separate, smaller versions of Web sites - to judge the success of campaigns. Microsites are usually tied to a special offer that consumers can't find anywhere else. They are most useful in measuring push if the microsite is designed around the "special offer."

Some agencies assign a URL prefix to connect microsite activity for their clients. An example of a prefix is 123.url.com, where the "123" is tied to a specific cable network or broadcast station, and to a specific destination on the advertiser's Web site. For Nike's Shox IV sneakers, for instance, an ad on ESPN might send people to 123.nike.com, while an ad on The Golf Channel might be coded 132.nike.com. With this method, advertisers need only one domain and can create an infinite number of prefixes, each taking 30 seconds to set up. This strategy means marketers don't have to worry about multiple sites, multiple domains, updating and "page not found" errors.

But, while microsites can make the TV or radio push to URL easy to track, search can add a wrinkle. Instead of dialing an 800 number or typing a URL, many people use a search engine to find the advertiser's site. These search customers are more challenging to tie back to a specific offline channel. If you can't track where orders come from, your offline might appear to under-perform. For example, if there is no clear relationship to a TV spot, and if it appears that the lead came to the Web site from a search, it would seem logical to give credit to the search program, increase its budget and cancel some TV spots. However, TV may have driven many of those search orders. The solution for some agencies is to combine deduction with the microsite technique. Microsites track the relative strength of each outlet. Using deduction to distribute a percentage of search activity leads to the appropriate media balance.

Online auction company uBid.com wanted to use direct-response TV to drive and track customer registrations and bidders. Florida-based SendTec developed 60-second and two-minute spots, and tracked the results using coreMedia's media buying and response analytics and SendTec's proprietary iFactz microsite tracking system. SendTec watched search activity to determine the impact of the TV campaign, which ran from November 22, 2006 through December 17, took a week's hiatus, and then resumed after Christmas. The change in bidder volume (see graph) correlates closely with spot volume. Prior to the campaign, bidder activity was much lower. Post-campaign, the level of weekly bidders remained 56 percent higher than pre-campaign levels.

The campaign also had a positive effect on search engine marketing, with the volume of searches on branded terms increasing 45 percent, clicks to uBid.com rising 62 percent, bidders on branded terms up 58 percent and on nonbranded terms increasing by 18 percent.

The flow of dollars into interactive shouldn't disrupt effective media planning. Planning has always been about mastering the mix: using each medium in a complementary way with balanced spending. Meeting this challenge is the core of quality planning. By combining interactive results data with response-to-media analysis, savvy agencies can optimize the schedule, produce concrete ROI analysis in-flight and deliver client results.

Carl Langrock is president of Fairfield, N.J.-based COREMedia Systems. (clangrock@coremedia-systems.com)