eBay: Paid Search Is Worthless

A study done by eBay on the effectiveness of paid search for established brands has found it to be worthless.

The eBay study called Consumer Heterogeneity and Paid Search Effectiveness: A Large Scale Field Experiment (no, I'm not kidding) concluded...

• There is no short-term value in brand keyword advertising.

• Stopping brand keyword advertising resulted in no detectable drop in traffic and sales. 

• The non-brand keyword experiments show that SEM had a very small and statistically insignificant effect on sales.

According to the study...

"eBay historically managed over 170 million keywords and keyword combinations using algorithms that are updated daily and automatically feed into Google’s, Microsoft’s and Yahoo!’s search platforms."

Presumably, eBay commissioned the study to figure out how much of their search money was being wasting. Their conclusion seems to be: all of it.

Now before you read the entire study, just a warning. It is written in that academic style that makes your brains fall out, and includes sentences like this... 

“The instruments isolate the exogenous experimental variation in spending to estimate the causal impact of spending on changes in revenue.” 

Which I think means something like this...

"We wanted to see whether paid search is worth anything or is just another online whack-off."

It also has formulas like this...

 

I don't have room for the rest of the formula, but I'm sure you can work it out for yourself.

Also, the author seems dangerously fond of the word "endogeneity," which I have always felt is inappropriate in polite company.

After wading through several pages of this stuff, I think the most useful piece of information is this -- eBay concludes that people who click on a paid search listing are just looking for a shortcut to a website which they would find anyway. So why pay for it?

I have no idea whether this study is worth anything or not, but I'm hoping it starts a great big shit fight between eBay and Google, which would be tremendous fun to watch.

Near the end of the paper the authors pose this rhetorical question about paid search:

"...why do well-known branded companies spend such large amounts of money on what seems to be a rather ineffective marketing channel?"

And they answer thusly:

"...Google offer analytical advice that is not consistent with true causal estimates of ad effectiveness."

Which I think means something like this...

"Google is full of shit."

Big thanks to Dave Trott for sending me this study.