Feb 182011

What's the Matter with the Search Alliance?

The Yahoo! and Microsoft Search Alliance has received some good news in the past couple of weeks with both comScore and Hitwise showing it gained appreciable market share from December to January.  For the month of January, the two market research firms show Bing powered searches comprising 25.6% and 27.4% of search traffic respectively.

From our perspective though, the picture isn't as rosy as we see signs that the Alliance's share of paid search traffic has eroded since the transition to Bing-powered results was completed in late October, 2010.

At this time last year, the typical RKG client received 24% of their paid search clicks from Bing and Yahoo, when the two were independent.  Now under the Alliance, just 20% of clicks are coming from the two sites and their partners:

It should be noted that Bing powered searches on Yahoo began well before the transition was completed with around 10% of Yahoo ads served by Bing in September 2010 and close to 50% by mid-October.  In other words, we wouldn't expect to see a precipitous drop in share as a result of the transition and we don't.

So what is going on if Bing and Yahoo are really chipping away at Google's dominance in overall search share as comScore and Hitwise indicate?

Reduction in Paid Search Partner Traffic

In the months following the transition to Bing-powered ad serving we have seen a sharp decline in the share of traffic originating from search partners:

As the transition to Bing ad serving ramped up, we were seeing around 30% of combined Bing and Yahoo traffic from search partners and that has steadily dropped to around 20% today.  For comparison, Google partner traffic currently makes up roughly 28% of their total.  Part of this decline may be due to the increased vigilance of RKG analysts in blocking the worst performing partners, but it appears Bing has simply done a better job of policing its program.

Along with the partner traffic decline, sales per click has increased dramatically for the Alliance since the transition and we now find it roughly on par with that Google after it languished around 85-90% of the value of Google for the previous year.

We've argued previously that poor quality partner traffic and inadequate advertiser control is likely to result in lower overall spend, so we would expect the decline in poor Alliance partner traffic and the increase in sales per click to lead to increased Alliance PPC share.  Why hasn't it?

Monetizing the SERP

After an initial decline in Alliance CPC relative to Google tied to the integration of lower quality Yahoo traffic with Bing's, we have been increasing our bids for the Alliance as traffic quality has improved overall.  Despite the more aggressive bids, the Alliance's share of paid clicks has continued to erode as seen above.

This is a bit anecdotal, but it appears that Bing is not monetizing its search results pages nearly as well as Google.  On search after search, we see Bing devoting far less real estate to ads, even for keywords with high commercial intent.  Take this example for the phrase "buy flowers" (with paid search ads highlighted in green):

Google SERP


Google shows 11 paid search ads taking up 571,000 pixels, when ad extensions are fully expanded, on a monitor set at 1680 x 1050 resolution.  In contrast, Bing displays a single paid search ad taking up just 19,000 pixels.  Google's result devotes roughly 30X the real estate to ads as Bing's result does for the same query.

One of the touted benefits of the Alliance to Yahoo and Microsoft was that it would draw in new advertisers who had not found traffic levels sufficient previously to justify the extra efforts involved with running on multiple platforms.  There's no reason to believe it hasn't, so it is odd to see such a disparity in the number of ads running on Google and Bing and for our CPCs not to be keeping pace with the improvement in sales per click.

We've seen cases where we have a handful of independent RKG clients advertising on the same keywords and Bing chooses to show just one or even none of the ads despite all having fairly aggressive bids.  It's possible that the results we see in our offices are not fully representative of what the average user sees, but given our overall data, it's hard to believe this isn't a real effect.  It's possible that Microsoft is deliberately keeping its results less commercial as it works to gain overall share or they may just be having a difficult time keeping up with Google's ad innovations.

There has been some conflicting data out there on this issue from other search marketers, but our figures sound similar to what Efficient Frontier has seen, so we are in are in good company there.  Microsoft's and Yahoo's most recent quarterly earnings filings also show weakness in search revenue relative to Google with Yahoo reporting an 18% decline YOY in Q4. On our end, it does appear that the Alliance may be turning a corner in the last few weeks when it comes to PPC spend and traffic, but it's a little early to call.

We're sure the engineers at Microsoft are working hard on these issues and we hope to see traffic levels trend more in line with the sales per click improvement soon.  One change coming up in the near future that could help is the loosening of trademark restrictions to bring them more in line with Google's and we expect there will be additional tweaks to Bing ad serving as well.  Stay tuned.


22 Responses to "What's the Matter with the Search Alliance?"
Tad Miller says:
It's my understanding that if your ad doesn't meat a 0.7% Click Through Rate on Bing, that they will not show it. I think that is a good part of the reason they aren't using all of their real estate for showing ads.
Tad Miller says:
whoops make that "meet" instead of "meat".
Josh says:
Mark, it looks like you're likely cookied into an MSN experience that isn't showing you many ads for some reason (searching extremely often without clicking ads did this in the past on MSN) - This is what I see: http://i.imgur.com/cuPgR.png . On the clients that I manage, Jan/Feb of this year has about 21% of volume coming from MSN/Yahoo, vs 18-19% before the transition. It all depends on the industries, really - The verticals that I work in have seen great improvements in MSN/Yahoo volume, and no decrease in conversion rates.
Mark Ballard says:
A good point Josh and I did make sure to note that what we see isn't necessarily representative, but it does make me wonder how many other users are getting a similar ad experience on Bing. I think the strongest evidence of the weakness of the Alliance relative to Google lies more in the quarterly financials than RKG's results or any anecdotal examples. Thanks for giving us your take on the trending. I agree that results are likely to vary depending on the types of clients and industries being analyzed and our view is just one of many.
Mark Ballard says:
Tad, I had not heard about a 0.7% CTR cutoff personally. Do you have any documentation of that?
Mark says:
We have experienced the same decrease in Alliance traffic across multiple advertisers, and we have also seen this decrease in traffic despite aggressive bids. Though I think you're giving too much credit to Bing in policing partner traffic. You may notice that the dip in traffic actually starts in April/May, and becomes more severe in October. In April/May MSN transitioned to Bing. When they made this transition they instituted a policy which only serves ads to users "likely to click on the ad." Following this switch we experienced a sharp decrease in MSN impressions. The .07 CTR threshold is an interesting additional variable to the system, and is just another way that Bing has decided to shoot themselves in the foot and take money away from themselves. The Alliance makes the dip in traffic more severe, because YH had a bigger share of search impressions, and now all of those impressions are being served under this self-restricting system. The end result is an efficient engine that is difficult to scale, which is a lose-lose-lose from vendor to agency to client, but Bing would argue creates a better Search experience for the consumer.
Bob says:
Is there any reason to think that the impact of an additional match type for Yahoo traffic (from standard/advanced to broad/phrase/exact) coupled with better matching (exclusive ad service for singular/plural which can convert VERY differently) is a contributing factor in any meaningful way? What has been the impact of the combined conversion pre/post transition or YOY in your study? I've heard that there has been little parity for advertisers pre/post transition and that YOY trends have not been favorable for most. Are you guys hearing/seeing the same? Thanks
Terry Whalen says:
I think it's academic, though still interesting. At the end of the day, I don't really care too much who is gaining in market share nearly as much as I care about good allocations of our clients' specific budgets between Google and the Alliance, based on respective performance. Having said that, I'm still not sure who is more stupid - Yahoo or Bing. Both platforms were/are laughable when compared to Google's or even when just measured against common sense.
Geordie says:
In my experience I've traced the traffic drop to reduced ads showing against trademarks on Bing due to more vigilant trademark policing. Apparently that's all changing March 3rd though...
Mark Ballard says:
Good insights Mark and good questions Bob. It's difficult to tease out all of the factors affecting Alliance performance precisely, but you've given us some great ideas for further research. In the post I focused on the relative performance of the Alliance and Google, but YOY we are seeing gains for the Alliance in revenue metrics, they are just not as strong as those for Google. When it comes to matchtypes for Yahoo, I would expect the greater control we have now to contribute to the improvements we have seen in sales per click, and eliminating some of Yahoo's bizarre handling of plurals/singulars should have helped as well while potentially increasing traffic volume.
ChicagoGirl says:
The Bing/AdCenter system is just so much more difficult to use. They are probably losing Yahoo's budgets because the system isn't set up to buy/serve ads the way advertisers need it to and the helpdesk is not very well trained to fix things quickly. (one sent me his script, and two others did not know how to help) A few weird things I've been told about their system lately: 1. If you have the same keyword in more than one campaign in Bing neither get shown and you are effectively shut off. In Google the one with the higher bid/quality score gets served. 2. If you have click through rate of less than 1% Bing shuts your ads down because they feel that means you don't have a quality ad/campaign/landing page/product for that search. In Google you may get a notice that your ads don't get shown often due to low quality scores but that is at a much lower CTR than 1%. 3. Two reps have told me that you do not get phrase and exact match ad serving when you select broad match, and in Google you do get all three included in Broad match. If this is true , it is very limiting for quality targeting/CTRs/getting your ads turned off. Microsoft is shooting themselves in the foot by not making things easier for advertisers and by having improperly trained staff.
Natively says:
I've been advertising on AdCenter for 5 years. In my opinion the traffic decline is due to: 1. Narrow keyword matching: AdCenter's "Broad" matching is quite narrow (compared to both Yahoo Panama and Google AdWords). It doesn't expand plural/singular forms, less synonyms , and often disregards geo locations. The search term "cheap led bulbs des moines ia" isn't triggered by neither of the keywords "led bulb des moines ia" (singular) nor "led bulbs" (no geo location) - even in broad matches. 2. Awkward UI. 3. As mentioned by Mark, MS had to filter out many Yahoo search partners. However if you happpen to bid on a "trending" keyword - expect sudden bursts of poor-performing traffic - all because of the "Trending Now" box on Yahoo's homepage. See here: http://community.microsoftadvertising.com/forums/p/67928/95180.aspx
MrCifci says:
Maybe there is a certain trend with your clients and how their paid search program is setup
Matt says:
Your search example for "buy flowers" didn't match up with my search test. I searched for "buy flowers" on Bing and found 9 paid search results. On Yahoo, for the same term, I found 10 paid search results. On Google, I found 11. Not sure why you're only seeing one paid result for Bing.


Check out what others are saying...
[...] This post was mentioned on Twitter by Peter Young, Rimm-Kaufman Group, Matt Whelan, Tom Bates, Matt Hopson and others. Matt Hopson said: RT: @ChadSummerhill: New RKG post! - What’s the Matter with the Search Alliance? http://t.co/41fnO83 [...]
[...] This post was mentioned on Twitter by Tim Cohn, Tad Miller. Tad Miller said: What's the Matter with the Search Alliance? http://bit.ly/ia7SBa We are seeing the exact same thing with our clients. [...]
[...] What’s the Matter with the Search Alliance?, Rimm Kaufman [...]
[...] Rimm-Kaufman Group recently posted some fascinating data regarding the Bing / Yahoo PPC trends since the merger (and in fact since a bit before the merger, [...]
[...] Bing consistently monetize considerably less of the SERP area than [...]
[...] unlike for Google, expectations are understandably low.  Earlier this year, we pointed out some troubling trends in Search Alliance traffic and we offered some anecdotal evidence of the Bing powered ad listings [...]
[...] million page views from December to January while Bing.com gained 138 million.  Around that time, RKG noted that paid search traffic for the combined Bing/Yahoo Search Alliance appeared to be turning a [...]
[...] though combined Bing and Yahoo revenue per click (RPC) has improved since Bing took over ad serving, CPCs have not kept [...]

Leave A Comment