The Great View-Through Debate

Although behavioral marketing is about niche targeting, behavioral targeting networks blanket the Internet in an effort to aggregate enough audience volume to effectively communicate with the finely defined segments or profiles advertisers require. The networks will gladly crow about their reach. They frequently cooperate with each other to extend their reach and make sure they can deliver for advertisers.

Whether the actual number is 85 or 95 percent reach, if you work with multiple large behavioral targeting networks, you can reach most of the Internet audience. Not all of the Internet will match your specific campaign criteria, but if you’ve applied sufficient budget, subject to competitive constraints, most of the people who match your criteria will be included within your ad-served population.

Hooray! You’ve just served targeted ads to a custom-designed population and, assuming you have a big enough budget and your competitors haven’t tied up the appropriate inventory, you can be pretty sure you’re hitting most of the people who fit the bill.

Chances are, however, you aren’t doing behavioral targeting or even online media in isolation, so we will need to appropriately attribute credit for that program and will likely be using some version of third-party ad tracking to do that. Campaign results could be measured by a variety of conversions, but to keep things simple let’s call them sales.

Clicks Are Easy

A profiled audience member is served an ad, clicks on an ad, and completes a sale. Click sales may be easy to track and understand, but they’re also rare. If broad consumer CTRs (define) hover at less than 0.25 percent and you apply site conversion rates once a user gets to the site, you see how the click sale numbers quickly diminish. With a 0.25 percent CTR and a generous 10 percent conversion rate, you’re translating just 0.025 of your ad-served population into a click sale. That’s slim pickings to justify a campaign’s costs, so why are advertisers still flocking to behavioral targeting?

View-through sales.

If only 0.25 percent of the ad-served population are clicking on your ads, 99.75 percent aren’t clicking — but they still may be converting or buying. Anonymous cookies track sales and other conversions to ad exposures during a defined period, even if users never clicked an ad. This allows advertisers to better understand the ad exposure’s value.

People surf for content, not ads, but absorb ad information both knowingly and unknowingly. Companies like Dynamic Logic and InsightExpress have repeatedly proven the brand impact of ad exposures online. Likewise, view-through conversions are a testament to the closing power of ad exposures, even without the click.

Counting Is Hard

But if the networks blanket the Internet and you count view-through conversions from those network ad exposures, then theoretically the networks will get credit for most of your sales by virtue of sheer size. This has multiple implications. Bigger networks get more credit than smaller networks or individual sites. Other factors, including creative and messaging, that may be driving site visits or conversions can get lost in the noise. Double-counting plagues multichannel programs for many reasons, including parallel but separate tracking systems.

Double-counting media and search can be avoided through utilizing a third-party ad server to measure your results. This is a major benefit of using a third-party ad server like Atlas or DoubleClick to serve both your media and search programs. Though these results won’t match Web site analytics, they will help your programs perform better because the media programs won’t take credit for search conversions.

The Consumer Is in Control

In their consideration cycle, consumers tend to stop at a search engine last. Ads play into consumers’ conscious or unconscious choices. When people are ready to buy, however, chances are they’ll return to sites of interest by way of a search. This makes search conversion rates look fabulous because they occur much closer to purchase intent, but it discounts earlier exposures’ value. Double-counting occurs because search gets credit for the close and networks get credit for that same consumer when they touched her earlier in the consideration cycle. Both are important touchpoints that support the advertiser’s goals, but which program gets the sale credit?

Advertisers with extensive media programs tend to count view-through conversions fully because they are either the only audience touchpoint, or make up an overwhelming percentage of their spend. Others retroactively apply sophisticated site-side analytics across all their programs to try to attribute a contribution percentage to each channel and tactic and to eliminate the double-counting that often results from multiple tracking tools.

Whatever approach you take, it’s important to set guidelines so everyone understands how view-through conversions, often a significant proportion of sales, will be counted. Clear optimization depends on clear communication on this point.

An accurate measure of view-through contribution is in the eye of the beholder. No doubt, all those targeted exposures influence consumer behavior but the networks’ reach may lead to an overreaching share of conversion credit by assigning a view-through conversion to any user exposed to network ads. Something between 0 and 100 percent credit is appropriate, depending on the advertiser’s unique environmental, programmatic, and analytic profile. Each advertiser has to find its own answer.