• by Gord Hotchkiss , Featured Contributor, May 12, 2005

Television is toying around with a new pricing model. From now on, you'll only be charged for the television ads that prompt you to actually take action. If you choose not to visit a place of business or eventually buy something, the advertiser won't be charged for that ad. If successful in television, the same pricing model will likely be used in all forms of advertising, including newspaper ads, magazine, and radio.

Yeah... right!

One of the paradoxes of search is that the pricing model described above, which is relatively unique to search, has proven to be a blessing and a curse. The idea of paying just for your performing ads and the accountability that it brings has fueled search's meteoric rise as a marketing channel. Its appeal has been particularly popular with direct marketers, where every single advertising expense is measured against the return it can bring. For these marketers, a pure performance-based pricing model was a gift from on high.

But in adopting this pricing model, search has also done itself a disservice. By not putting any value on the ads not clicked on, search has implied that these ads are worthless. But as more research comes out showing that search's role in a customer's buying decision is much more complex and long-term than we thought, it's beginning to appear that the unclicked search ad could be the bargain of the century. Because search, my friends, does build awareness and those ads do have value.

The Role of Search in the Buying Cycle I've talked about where search is typically used by a prospective customer often enough. Research conducted by comScore has shown that potential customers can launch anywhere from two to six related searches in the 12 weeks preceding a purchase. That means two to six interactions with a number of search results pages. Combine that with our own eye-tracking research that shows that the top region of the search results page (referred to in our study as Google's Golden Triangle) has 100 percent visibility. This includes top sponsored ads and the top three or four organic ads. So, every eyeball for that search will see an ad in this prime real estate. But the advertiser only pays if the ad is clicked. We know that proportionately, only about 15 to 20 percent of the clicks will happen in these top sponsored locations on Google.

So let's put some real numbers to this and try to get some sense of the value provided. Let's assume you're bidding for a term that will get 50,000 searches in a month. You bid enough to capture the top sponsored spot. Your per-click bid price is $1.50. And, we'll estimate that you capture about 7 percent of all the clicks on the page. One last assumption: Every search does not result in a click-through. So let's say that 15 percent of the searchers will not find anything on the page worth clicking, and they'll either relaunch the search or click-through to the second page.

So, in a month, given the above assumptions, you would get 2,975 visitors at a cost of $4,462.50. But you've also had your ad seen by 47,025 other people, for free! True, these visitors didn't click-through to your site... this time. But remember, chances are they'll be coming back to a search engine and launching a related search at least one more time in the buying process. If your ad comes up again, the reinforced brand recognition might prompt a click-through during this second session.

We Have to Measure the Full Value of Search Search marketers are fond of saying that search is the most measurableof marketing channels, and that's true, up to a point. I believe one ofthe reasons we don't give search full value is that we're not always measuring the right things. How do you measure the value of a split-second glance at a brand name in a search listing? How do you assign a value to the cumulative impact of seeing the same site appear in four or five different searches? I know these things have value, but I'm not sure how to measure it.

We're very good at measuring the easy conversions. We can track back from a purchase or the submission of a quote request form to see which listing on which engine generated this lead. But we're not good at measuring subtleties and nuances. It's difficult to assign values todifferent patterns of site-side user behavior. It requires a conversion-tracking mechanism that extends into every aspect of the business to track offline purchases that are generated by online research activities. And theonline analytics industry is just beginning to grapple with the challenge of getting a more balanced picture of true-visitor value.

The role of search in a customer research session is much more complex than we ever imagined. As we do more research, we'll get more clarity in regards to how search helps influence buying decisions and the nature of a customer's cognitive interaction with the search results. As we find these answers,we'll get better at assigning value to our search advertising, whether the ad is clicked on or not. But until then, recognize that it has a value and enjoy the free ride!