IndustryClaria Unveils Behaviorial-Based Search Ranking

Claria Unveils Behaviorial-Based Search Ranking

Claria hopes its new RelevancyRank system will let it improve the results of major search engines or perhaps offer its own improved search engine by mining behavioral activity.

Claria, the company behind the Gator eWallet software, has released new search relevancy ratings today examining how the top search listings on Google, MSN and Yahoo compare to pages the company says its research shows are actually most relevant. More important, the ratings mark the first use of technology Claria hopes will let it improve the results of major search engines or perhaps offer its own improved search engine.

At first glance, the scores gained by the search engines were pretty dismal. For the ten travel-related terms listed below, no search engine came anywhere close to earning a 100 percent perfect score. Shown for each term is the search engine among the three tested that performed best for it:

Term

Search Engine

Score

Cruise

Google

68

Travel

Yahoo

62

Airline Tickets

Yahoo

62

Jamaica

Yahoo

62

Mexico

Google

60

Florida

Yahoo

56

Las Vegas

Yahoo

55

Hawaii

Google

52

Cheap Tickets

Yahoo

50

Hotels

Google

47

Average Of All Top Scores

57

How does Claria even know what the 10 most relevant listings for any query should be? The company is making use of user behavior that it tracks to determine this. Among the things its tracks through its software applications and gathered through partnerships with publishers are:

  • Which pages are viewed
  • What people have searched for
  • What people click on in search results and other click choices they make
  • Time spent viewing pages
  • Number of pages viewed
  • Number of return visits to a page or a site
  • Purchase and conversion rate data

These factors are combined into what Claria calls its RelevancyRank system. The date means that for any term, it can compute what it thinks are the best pages for that term. In particular, it is doing this by monitoring searches that happen at places like Google, Yahoo and MSN and seeing what searchers ultimately seem to like.

A key problem in the ratings above is that it’s assumed that for all the terms, there was a “travel” intent. But someone searching for Mexico might not actually want travel information.

Of course, if you dig deeper on the Claria site, you can see exactly the pages it considers best for the terms and find that not all of them are necessarily “travel” pages. For example, the mexico query has a listing for the CIA World Factbook, providing encyclopedia-style information about Mexico rather than being a travel guide. To play with all the queries shown above, see the hard-coded answers posted at the RelevancyRank demo page.

While it might be debatable whether Claria really does know the “perfect” pages for each query, what’s more important is that the behavioral metrics of something like RelevancyRank might be another factor that the major search engines could mix into their existing ranking algorithms for gaining another leap in relevancy.

It’s another form of personal search in that it leverages the individual actions of many people to tweak results. Or for those with long memories, think of this as a third generation or level of clicktracking, one that’s now factoring in behavioral data.

First generation? That would be Direct Hit, as covered from this article from way back in 1998: Counting Clicks and Looking at Links. Direct Hit is first generation in that it’s the first time clicks were seriously used to refine search results results. Direct Hit even offered “Top 10 Most Popular” results at place like MSN but died off because owner Ask Jeeves didn’t further refine the technology, which was susceptible to click spam.

Second generation? These include Eurekster and to large degree the other personal search features rolled out last year at places like Ask Jeeves, Yahoo and A9, as summarized in my Looking Back, Looking Ahead: Developments With Consumer Search article from earlier this year. These are second generation in that clicktracking is being used, but by and large, post-click activity isn’t being monitored. And calling them second level rather than second generation may be more fair, because “generation” implies they aren’t as useful as a further generational jump. That’s not necessarily true. The data gathered to personalize results may be very helpful when it is finally used to a large degree.

That brings me to the idea of third generation or the third level of clicktracking. Claria is watching post-click activity and seeing how it can be used to refine results. The company’s plan is that the technology will either be licensed to search providers looking to use its data or it may release its own search engine powered by clicktracking and behavioral data itself. It already operates the SearchScout service. That really only lists paid results, but it could be switched to show a combination of editorial and paid listings.

Similar to the AlmondNet post-search ad targeting rolled out earlier this year, behavioral tracking like this raises some privacy issues. Unlike AlmondNet, results aren’t being delivered based on someone’s personal profile. Instead, Claria says it makes all search profiles aggregated and anonymous.

Still, the idea that your behavior is being tracked somehow may worry some. To help on this front, Claria announced the formulation last month of a team of privacy, security, public policy and consumer protection law experts that it wants to help them develop policies and practices around the behavioral tracking space.

“The mission of Claria’s new Vista Marketing Services Division is to create a more personalized web experience for consumers” said Scott Eagle, Claria’s chief marketing officer. “By obtaining permission from users to observe their anonymous online behavior, Claria and their partners can provide a better search, content and advertising experience for consumers.”

Resources

The 2023 B2B Superpowers Index

whitepaper | Analytics The 2023 B2B Superpowers Index

8m
Data Analytics in Marketing

whitepaper | Analytics Data Analytics in Marketing

10m
The Third-Party Data Deprecation Playbook

whitepaper | Digital Marketing The Third-Party Data Deprecation Playbook

1y
Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study

whitepaper | Digital Marketing Utilizing Email To Stop Fraud-eCommerce Client Fraud Case Study

1y