Microsoft says Brand Search Ads a Waste of Money – Show Me the ROI!

I have to chime on this since the timing of this Microsoft release and SES is not coincidence. It’s certainly attention grabbing but I think it’s almost irresponsible to make these claims without the right data to support them.

In an article backed with little ROI data Microsoft/ Atlas is saying that it is a waste of money to put ads on your branded queries. This runs counter to my own extensive research that I’ll be presenting at SES and I’ve posted before. My research is backed up with conversion data and RPV. Nowhere is this mentioned in the AdWeek article leading to believe it was either not measured, does not support their story or AdWeek missed the most important aspect of the report. I’ve yet to read the report but I hope to get more color on it tomorrow as I’ll be on a panel with Nico Brooks from Atlas.

Posted

August 20, 2007

A/B & MVT Testing, Analytics, Branding

Tags:

Comments

4 responses to “Microsoft says Brand Search Ads a Waste of Money – Show Me the ROI!”

Nico Brooks

August 22, 2007

Jonathan,
As the author of the research refernced in the AdWeek article, I’d like to make a few comments:
I did not say nor do I believe that it is a waste of money to buy branded search terms. What I did observe is that a significant percentage of search activity appears to be navigational, and I do believe that the vast majority of branded searches fall in to that category. Look at Hitwise top search term data for an example:
http://www.hitwise.com/datacenter/industrysearchterms/index.php
These are clearly people intending to visit myspace, eBay, etc.
You may argue with my percentages, but there is no question that search engines are being used for navigation. To calculate the return on investment for a navigational click, one must also factor in the other ad/brand exposures that preceded the conversion, since they also contributed to the intent to purchase.
Most advertisers are attributing conversions to the last ad clicked, which is a flawed methodology if the last ad clicked is only part of the story.
I agree that in most cases it probably does makes sense to buy branded search terms, but I also believe that the click price should be comensurate with what the advertiser is getting in return.
I hope that provides the additional color you are looking for.
Kind Regards, Nico

LikeLike

Reply
Jonathan Mendez

August 22, 2007

Nico,
Thanks for your comment and how ironic that we shared the Ad Testing panel at SES!
I think as an industry we’ve known for a long time that navigatinal queries make up a giant chunk of user search behavior. So with all due respect in that regard that data is not new or interesting.
I guess it was Young Bean’s comment that really struck me:
“Should you be spending half your search budgets on those [branded] terms?” Song said. “Probably not.”
I cringe when I hear a blanket recomnendation like this, especially from Atlas. Every vertical and every business is different. There is no best practice with branded keyword buys. you need to test and evaluate ROI (yes, includive of other media).
In my presentation you saw for that major retailer branded kw buys were huge wins in every engine. If the volume of queries was there to to spend half their budget on brand they would!

LikeLike

Reply
David From About Results Marketing

September 5, 2007

John,
Did you have a comparison of natural vs. paid clicks on branded terms?
As an ex-affiliate marketer, I used to happily bid on brand names for similar or competing products, snapping away conversions at very low costs per conversion.
Have you created an evaluation of how many people are lost to competitors by having someone else in #1 position for your brand term? For example, compare natural vs. paid conversions on branded terms, when you are number one in paid.
Than pause your campaign for some time, and measure the increase in natural visits and conversions?

LikeLike

Reply
Kathy

July 14, 2008

You guys do a wonderful job! Keep up the good work!!!

LikeLike

Reply