No fluff - just the best news in paid search marketing every week.
eBay’s “Paid Search Is Ineffective” Study Gains Buzz And Skeptics, Again
By claiming that paid search is largely ineffective, eBay has attracted a lot of attention and buzz with media outlets repeating the findings.
Anyone who isn’t familiar with eBay’s history of throw-it-at-the-wall-and-watch-what-happens paid search advertising strategy may be able to take the company’s final release of a report that first made the rounds last year at face value. Those familiar with eBay’s paid search efforts (i.e., “Bid on Cancer now”) have taken the study with a grain of salt, if not discounted it outright.
The problem isn’t with the research or testing, it’s the fact that the study relies on eBay’s paid search campaigns and then proceeds to extrapolate the findings to the effectiveness of paid search for large brands in general– as if eBay’s paid search results are a proxy for all major brands.
A look at the historical ad archive in SEMRush provides a sampling of the types of keywords eBay’s ads were showing on and the Dynamic Keyword Insertion at work in both the ad headlines and ad copy in 2012, when the study took place. (Note that SEMRush displays archived ads the way they would appear today with the yellow “Ad” icon.)
These ads below — for Winning Lotto Numbers, Scanners Police, Pink Roses In A Vase and Base Ball Babe Ruth — are from the January 2012 archive.
The study’s authors – eBay’s data scientist and researchers at UC Berkeley – contend that, “Because eBay bids on a universe of over 100 million keywords, it provides an ideal environment to test the effectiveness of paid search ads for non-brand keywords.”
The problem with this reasoning is that this assumes ad copy and keyword strategy doesn’t play a role in paid search.
As Matt Ackley, now CMO at Marin Software and formerly at eBay in the early aughts where he oversaw the pause of Google AdWords for a period in 2007 to test effectiveness, points out the researchers don’t seem to distinguish the value that paid search can offer in attracting new customers. Referring to this quote in the study:
“We ﬁnd that SEM accounted for a statistically signiﬁcant increase in new registered users and purchases made by users who bought only one or two items the year before. For consumers who bought more frequently, SEM does not have a signiﬁcant effect on their purchasing behavior. We calculate that the short-term returns on investment for SEM were negative because more frequent eBay shoppers are accountable for most of paid search sales.”
Ackely responds in a blog post, “It is important to understand that paid search is a great acquisition channel. When considering the ROI of your advertising spend it is important to attribute some percentage of the Customer Lifetime Value to the acquisition channel. We did that very early on at eBay and it had a large impact on our ROI. This was the right move. However, a company the size of eBay has less Americans to acquire and more of the paid search activity becomes retention related. It is at this point that pure transaction value may not meet certain ROI hurdles when compared to other channels.”
Additionally, Google now offers a way to bid differently on searches from past or frequent visitors with Remarketing Lists for Search Ads, as Ackley also addresses.
So, is there an ulterior motive for eBay to gain from dismissing paid search – and Google in particular – and suggesting other big brands do the same? Certainly eBay would rather be the place where consumers first turn to “search” for products. Both eBay and Amazon have had their tiffs with Google, particularly as it has become much more aggressive in its product search efforts and transitioned Google Shopping to a paid platform. Amazon refused to pay for product ads that had once been free.
Interestingly, this study doesn’t mention Google Shopping at all, but the study took place before it switched to paid, so the impact of turning off paid search for eBay could be much different today now that there are no longer opportunities to get free traffic from organically-generated product listings.
An interesting study would be to look at the impact on Amazon’s sales when eBay turns off its paid search campaigns.