This article is re-posted from User Centric’s blog and SEOmoz.org's blog, YOUmo.
When creating a pay per click ad (PPC) ad, it is easy to overlook the importance of the display URL. However, search engine marketers have only four lines of text to work with and should take maximum advantage of each of them. (Not including the geographic descriptor that is added as a fifth line for some geographically targeted ads.)
This article summarizes the results of an eye tracking study and a review of two tests of display URLs. The eye tracking study demonstrated that the display URL in PPC ads receive a significant portion of the gaze time allocated to the ads. The review of the two tests shows that display URLs can have a meaningful impact upon click thru rates.
Eye Tracking Study Shows that Display URL’s Attract Attention
In order to gain a better understanding of viewer interaction with display URLs, User Centric, Inc. (now GfK's User Experience group) conducted an eye tracking study measuring the length of gaze time spent looking at display URLs by viewers of Google search engine results pages. The study measured thirty individuals who participated in the research with each performing twelve searches using Google.com.
The results of the eye tracking study confirmed that the display URL is an important component of a PPC ad. The findings revealed that the display URL captured a significant portion of the time people spent looking at each PPC ad, as shown in the tables and heat map below. For PPC ads served above the organic listings on the left side of search engine results pages, the display URL garnered over three times more gaze time (0.53 seconds) than the text description (0.15 s). Further, the display URL in PPC ads in the top position on the page above the organic results received the same amount of gaze time as the headline, 0.65 s. Average gaze time (in seconds) per visit to a Google search results page:
The position of a PPC ad on the page had a significant impact on the gaze activity the URL received. Not surprisingly, the display URLs of the top three ads on the left attracted more attention than the top three ads on the right. Display URLs presented on the left received almost three times as much attention (0.53 s per URL) as URLs on the right side of the page (0.19 s per URL). The length of time spent looking at a URL also varied by the specific position of the ad. For ads on both the left and the right of the page, the URL located within the first ad was looked at the longest while the URL within the third ad was looked at the shortest. Within ads on the left, display URLs received, on average, more attention (0.53 s) than the ad text descriptions (0.15 s) but less than the ad headlines (0.70 s). This is not surprising because in those PPC ads, the display URL is located to the left of the text, directly below the headline, which makes it more prominent than the description text. URLs within ads on the right received, on average, less attention (0.19 sec) than both the headlines (0.27 s) and ad text descriptions (0.30 s). The display URLs for ads on the right are the least salient ad elements, as they are presented at the bottom of the PPC ad, below the text description.
Heatmap showing aggregated participant gaze activity during one of the study’s twelve searches (the redder the area, the more attention it received):
Test of adding descriptive keywords to display URL’s was a mild failure
Understanding that display URLs receive significant gaze time is not simply a branding issue. The display URL can be altered in attempts to improve results. The most common manner in which display URLs are altered is by adding descriptive keywords to the end of the domain name. A Google employee in account management encouraged me to test this tactic. In our experience, adding keywords in the display URL fails in the majority of cases, particularly for domain names that have significant namerecognition and brand equity. However, adding keywords at the back end of a display URL increases click thru rates often enough that it is worthwhile to test. Shown below is an example from a test of display URLs. The three ads are identical except for the display URL. The goal of the test was to determine if adding “eye tracking” to the display URL would increase the click thru rate of a PPC ad promoting User Centric’s eye tracking service. The first ad simply features the domain name as the display URL, the second ad includes the descriptive keyword term added to the end of the domain name, and the third ad includes the keywords in front of the domain name.
In the test of the above PPC ads, the version featuring only the domain name produced the best results. Adding the keyword term “eye tracking” to the end of the display URL in the second ad resulted in about a 5% decrease in click thru rate. As is often the case, adding keywords to the display URL reduced the click thru rate.
The most dramatic result of this test was that adding keywords in front of the domain name had a disastrous impact on click thru rates – it decreased clicks by over 25% (on top of the fact that our ISP charged a minimal fee for an extra top level domain name). Another drawback of adding descriptive keywords to a display URL is that this form of “keyword stuffing” may reduce the branding value of a PPC ad by cluttering up the display of the domain name.
Test suggests that the domain name in display URL has a major impact upon results
The value of an identifiable domain name in a display URLs was highlighted by a test conducted for iNest Realty. Their parent company is RealEstate.com, a much more recognizable domain name. The test was conducted to determine the impact of using the stronger brand name in the display URL of PPC ads. The test ads are shown below. The only difference between the two PPC ads is the display URL.
Utilizing RealEstate.com in the display URL resulted in a 20% greater click thru rate for the bottom PPC ad. Based solely on that information, it might appear that the bottom ad was the winner. However, the ad’s conversion rate was abysmal. In order to comply with Google guidelines on display URLs, a landing page had to be created for this ad on the Real Estate.com server. The page was co-branded for both iNest and RealEstate.com. Visitors did not like the experience of clicking on an ad with one display URL and then seeing information about a different brand. Thus, iNest was paying for a 20% increase in clicks, while generating fewer conversions.
If the test had not been ended quickly, it would have been an expensive lesson on the importance of providing users with a good experience. However, the test provided a good demonstration of the positive impact a recognizable domain name in the display URL can have upon click thru rates.