Study: Display Ad Exposure Boosts Search Conversion

A new study from Atlas Institute proves what many marketers have intuitively known: Running display ads can help improve the results of a concurrent search campaign.

The study, which looked at 11 direct response advertisers, found a 22 percent lift in conversions when users were exposed to both search and display ads from the same advertiser, compared to being exposed solely to a search ad. A lift of 20 to 65 percent was seen in eight of the 11 advertisers, while three others saw no positive or negative effect.

“It’s long been thought that the sum [of search and display ads] is greater than the parts, but traditional metrics have focused on single silos,” Esco Strong, research analyst at Atlas Institute, told ClickZ. “Advertisers want to know if there’s a benefit to doing both concurrently.”

The study examined the behavior of 1.8 million users during April 2006 and found a 44 percent overlap of users seeing both search and display ads from the same advertiser, so the effect is significant enough that marketers should be paying attention to it, Strong said.

Users were segmented into exposure groups according to whether they were served display ads, search ads or both. The study found users exposed to search ads alone converted at three times the rate as those exposed to display ads alone. Users who were exposed to both search and display ads from the same advertiser converted at four times the rate as display-only users, a 22 percent lift over those exposed to search alone.

One possible explanation is display advertising‘s ability to generate brand awareness and increase purchase intent, helping to win over users who are on the fence about a brand or product they have previously searched on, Strong said. Another factor is the likelihood of display ads to drive brand searches. In addition, many consumers tend to use search as navigation, so users who first visit a site via a banner ad may later return to the site via search, he said.

The problem with identifying this synergy lies both in technological limitations and a business’ organizational structure, according to Young-bean Song, director of analytics for Atlas Institute. If a marketer is measuring display ad campaigns separately from search campaigns, he might not connect a user’s display ad impression history to a conversion that takes place as a result of a paid search ad. The same thing will happen if search and display campaigns are being handled by separate departments.

“As a result, Google looks fantastic, while other touch points are ignored,” Song said. He suggests tracking all media buys centrally, to ensure that cross-channel impacts can be measured and optimized.

“The ROI model that has been used for so long, where the last transaction gets credit for the sale, is a very simplistic view. It’s really a multi-dimensional problem that requires looking across different channels,” Strong added. “People have been complacent with this simple view, but there’s growing interest in measuring across channels.”

That interest is sparked by marketers’ desire to justify spending on multiple channels, which can be a challenge when search seems so easy to justify on its own, but marketers lack data to prove the direct-response benefit of display ads, Song said.

Exposure to both search and display ads can improve the search click-to-conversion rate, so individual search clicks are more valuable. That opens up the possibility of bidding higher on keywords, perhaps moving up to a higher position, Strong said. “Measuring this synergy opens new avenues for campaign planning and optimization.”

The study was not designed to find the optimal frequency of display ad exposures to improve conversion, but Song said that number is consistent with other research on optimal frequency. The point of diminishing returns will vary by advertiser, but he suggests it generally lies somewhere between 3 and 8 exposures.

That number will be affected by how targeted the inventory is, since less-targeted, run-of-network ads tend to burn out quicker and stop converting after one or two exposures, while a more-targeted ad on a niche site can continue converting even after 4 to 10 exposures, he said.

Conversions can be increased by choosing display media that reaches users who have clicked on the advertiser’s search ads, and by maximizing reach of display campaigns while minimizing frequency of exposures to the same users, Strong said.