EcommerceBytes-NewsFlash, Number 2913 - October 15, 2012     1 of 5

eBay Poised to Turn Search on Its Head

Email This Story to a Friend

eBay is testing a brand new way to serve up search results on its marketplace in advance of a full rollout of Cassini. The new eBay search will be based on personalization, and when implemented, has the potential to disrupt eBay sellers' businesses in several major ways.

eBay's testing ground includes features that lets buyers personalize their search results depending on what they're looking for and what is most important to them, and potentially rates sellers not only on raw feedback scores, but also on buyer sentiment pulled from a multitude of sources including the actual text of feedback comments left by buyers - and even potentially from all buyer-seller communications including those sent through eBay's messaging system.

Letting Buyers Influence the Order of Search Results
One major feature being tested by eBay is called "TVS Beacon," which lets shoppers rank search results by a combination of four criteria: Trust, Value, Diversity and Relevance, depending on which of the four criteria is important to them. For example, collectors are less price sensitive but might want to see more diverse results - Harry Potter fans might want to see all kinds of merchandise, not just books and DVDs, so they'd dial up the Diversity factor. Bargain hunters, on the other hand, want to see low-prices, so they could dial up the "Value" factor.

eBay has been testing the TVS Beacon for some time, and in research published in 2011, eBay said users exposed to the ranking tool were pleased to have more control over the listings that appeared in their search results.

If the ranking feature is an indication, eBay may significantly cut back on the number of results returned in search results. And if eBay doesn't bring back all listings that match a search query, the marketplace would be hard-pressed to continue charging sellers listing fees. An eBay seller who's been using the tools, Ty Neil, agrees.

Mining Feedback Comments to Get a Truer Picture of Buyer Satisfaction
Another feature eBay has been testing is a buyer sentiment tool called Emosi Sosial, a tool eBay had tested in 2010. As described by Wall Street Journal India personal technology columnist Amit Agarwal, the feature allows users to check an eBay seller's history through a tag cloud of the comments on that user's profile page. (A tag cloud is a weighted list in visual design.)

But eBay is going beyond tag clouds and is working to mine the text in seller feedback comments looking for patterns - both positive and negative - to get a more accurate picture of the level of buyer satisfaction each seller is providing.

Many buyers are loathe to be too negative about a seller, even if they feel dissatisfied with a transaction. So some will leave a positive feedback rating for sellers but will hint or complain in the feedback comments.

This has enormous implications for sellers, because eBay can mine every single feedback comment looking for root causes of pleasure/displeasure, and use patterns it finds to influence sellers' exposure in search results. Not only that, eBay will likely use this data in its Fraud/Risk models as well.

What We Know about Cassini
In 2008, eBay CEO John Donahoe and CTO Mark Carges expressed their dissatisfaction with eBay's Voyager search engine. They set a plan into motion to launch a new search engine in a project called Cassini that was at one time slated to launch in 2012, but was postponed until 2013.

Former Microsoft exec Hugh Williams, who worked on Bing, is charged with building Cassini. What little we know of Cassini comes in part from a Reuters article in May of this year:

"Cassini will trawl full product descriptions, rather than just the titles of listings, and match search queries to photographs of products, while taking into account information about the seller and the buyer.

By crunching data on what shoppers have bought and browsed on eBay in the past, Cassini search results should be more tailored to their intent. For instance, if a shopper types in "HP," Cassini will know if the person means horsepower or Hewlett-Packard Co, (Hugh) Williams said.

"Voyager is pretty literal. It takes a query and matches it faithfully against the title of items. It's not intuitive," he said. "Cassini will take the user's query and understand that.""

And in August of 2011, Williams told The Register he wanted Cassini to be a "world class" search engine - "something capable of understanding precisely what it is that users want and doesn't shower them with heaps of tenuous results."

Goodbye DSRs, Hello Free Listings
Free listings would be necessary if eBay is going to display only a subset of listings in search results that are unique to each searcher based on what eBay knows about them, rather than returning all potential matches.

Along with listing fees, eBay may also be saying goodbye to Detailed Seller Ratings (DSRs) as we now know them.

While eBay hasn't communicated to sellers and buyers its plans for its new search engine, its scientists have published research and made presentations about the projects they've been working on - including on the eBay Tech blog in this post from 2011 on personalized search. In part two, the author said he would find it hard to believe that anyone would seriously argue that two shoppers with different preferences should see the same search results. The author also wrote that while the current Best Match does not do any personalization, "we hope to add it in next year."

Ty Neil, an eBay seller who has used the search tools in eBay's testing ground, provided EcommerceBytes with a seller's perspective on changes coming to search.

Comment on the EcommerceBytes Blog.

About the author:

Ina Steiner is co-founder and Editor of EcommerceBytes and has been reporting on ecommerce since 1999. She's a widely cited authority on marketplace selling and is author of "Turn eBay Data Into Dollars" (McGraw-Hill 2006). Her blog was featured in the book, "Blogging Heroes" (Wiley 2008). Follow her on Twitter at @ecommercebytes and send news tips to

You may quote up to 50 words of any article on the condition that you attribute the article to and either link to the original article or to
All other use is prohibited.