Search less, pay more. That’s the main takeaway of three economists who did a deep dive on 500,000 eBay users’ search habits over a 30-day period in 2014. Their recently published findings differ from previous research on search behavior and are thus worth examining.

By default, eBay displays search results using a ranking algorithm called “Best Match.” This algorithm was created to display items in the order that best predicts expected eBay revenue, maximized by increasing the probability that a product is purchased times its sale price.

The researchers from eBay and the universities of Chicago and California, Berkeley found that people searched an average of 36 times per transaction. According to their calculations published by the National Bureau of Economic Research, people saved an average of $15 on purchases for every hour they spent searching.

Among the major findings:

• Consumers save, on average, 25 cents per search page and about 75 cents for each day spent searching.
• Most users do not “unsort” best match.
• Many users drop off in the first few minutes of sessions, suggesting that many searchers abandon the site if they do not quickly find what they are looking for.
• The process starts with a “what would I like” phase, followed by more of a “find a good deal” phase. This behavior fits neither the existing fixed-sample nor sequential search model.
• Whereas a search engine like the one at eBay is tuned to encourage immediate purchase, the site might be better served if it thought holistically about this search funnel and helped consumers learn about the attributes of products in a way that ultimately led them step by step down the process instead of assuming that they are at the end of it.

Ryan Pitcheralle, VP of the Digital Center of Excellence at Acronym, whose background includes a stint at eBay, offers some observations about the research.

The thing about eBay is that it really is a truly unique shopping experience. More than half of the products sold there are actual auctions, meaning you can bid for price. So with that in mind, most of the searches on eBay come from users looking for something very particular that is no longer available in the mainstream. This notion could speak to the high bounce rates, as the product is either available or not. Casual check-ins of discerning seekers looking for the gems they seek are the norm.

There are a couple reasons for the number of searches to reach a transaction being surprisingly high. One of them being the “Best Match” filter. For every search, this filter essentially reorders the next page of results as the query is refined. This tends to lend itself to a user behavior of continuing to refine the query. The other reason is that most users seek something very particular, but this doesn’t necessarily mean they know exactly what they seek. Things can be commonly known by a multitude of labels. So with that in mind, searchers will first seed with a general query – a broad one. Then based on the results and the labels they use, the users will pick up on some additional labeling nuances, maybe a product ID or full product name, perhaps even an SKU. All of this can be interpreted by the user on the fly from perusing the results. Each viewing suggests new “seeds” to use when refining their search. But I think, above all, users want to be sure they sought the right way. As in, did they look hard enough for that diamond in the rough? Did they go deep enough past Best Match to find what’s hidden behind the algorithm? These things drive up searches leading to a sale.

With regard to the principles of intent-based marketing, the user may start with an early stage “what would I like?” and then move into the more transactional “How cheap can I find it?” With eBay, though, the behavior already starts very well down in the buyer’s journey. Remember, it’s an auction site and it’s not as if eBay merchandises its products extremely well across all the categories it actually points SKUs to. So users on eBay definitely have their hand on their credit cards.

So what can online sellers do? Research the same way as users. Find the right labels and use them all. And above all, the more product pictures, the more sales. Images sell products that you can’t touch. On eBay, the content is product content. Product content is content that satisfies transactional intent. Very little “I’m only looking” occurs on eBay. It’s a deep entry site. Most traffic enters on internal pages and not the home page as most retail sites assume.

The research can be accessed here.