Effect on CTR Through Combining Organic and Paid Results
In 2009 Dr. Bernard J Jansen or Pennsylvania State University and Prof. Amanda Spink of Queensland University of Technology, Australia, co-authored a research paper focused on “Investigating customer click through behavior with integrated sponsored and non-sponsored results.” The purpose of this research was specific—to discover if web searchers were being deprived of opportunities to make relevant searches because of a perceived negative bias towards sponsored links and their obvious placement on search engine pages. Would link-integration increase searcher click-throughs of sponsored links as the two categories became less distinguishable?
Certain premises were acknowledged at the beginning of this research, based on earlier reputable studies:
• Web searchers care about relevant results.
• A genuine negative bias exists towards sponsored links
• Separate listings emphasize the difference between organic and sponsored links and may indirectly discourage searchers from accessing relevant links which is a disservice on the part of the search engines.
Jansen and Spink collected significant data from Dogpile, a meta-search engine that collects information from four other search engines and integrates the data into its listings using an algorithm that eliminates duplicates and includes both sponsored and non-sponsored links. The transaction log file examined on 15 May, 2006, contained 7,142,874 records which were sorted and identified by time of day, new user ID, cookie and query.
Once the applicable information was extracted, it had been reduced to 666,599 users, 1,874,397 queries and 4,201,071 interactions which constituted queries, page views and click throughs. This data was then sorted into categories and classified as informational (texts, documents, data, multimedia), navigational (en route to another site) and transactional (to purchase a product, download multimedia, execute a service). The 35% of views that led to no click throughs were discarded. The remaining results showed that the vast majority of click throughs were for informational purposes (80%). The remaining searches were equally divided between navigational (10%) and transactional (10%).
Although Dogpile has a higher click through rate (65%) than the average search engine (45%), the results indicated that 84% of the web searchers still chose non-sponsored sites and only 16% used sponsored links. This is a much lower figure than the advertised 30% that is attributed to sponsored links, but it seems to agree with earlier research. The low figure indicates that there is no substantial improvement in click throughs when sponsored and no-sponsored links are integrated.
This carefully researched and documented report has far reaching implications for both advertisers and future search engine implementation and development. Since both sponsored and non-sponsored links can be relevant, finding a way to overcome negative bias on the part of the searcher may be critical to better service. Trying to “trick” by integration of the two varieties only creates increased suspicion and may drive the web searcher from that search engine completely. On the other hand, sponsored links are the financial backing behind the “free” searches millions of web users enjoy daily. By studying user behavior more thoroughly, search engines may be able to leverage that information to create algorithms and interfaces that serve the user, the advertiser and the host site better.
By combining Google Ads (paid) and Google’s Organic Results (free) advertisers can create higher rankings both immediately and in the long term. Pay-per-click ads are rotational, guaranteeing immediate, temporary high ranking; and organic marketing, when aimed at a clearly targeted audience can produce lasting results over months.
Further real world testing based on this important research has been scheduled by Dejan SEO for the near future. Look for the results on our website and our upcoming research on concept of ASEO (academic search engine optimisation).