Preliminary Scorecard: How Effective Is Yahoo’s Panama Upgrade?

The world of search marketing is divided into two types of people, it seems: Those who have never actually used Yahoo’s new Panama sponsored search upgrade and never will; and those who are in knee deep working on their accounts, and want to find out how other fellow marketers are faring, so they phone me. […]

Chat with SearchBot


The world of search marketing is divided into two types of people, it seems: Those who have never actually used Yahoo’s new Panama sponsored search upgrade and never will; and those who are in knee deep working on their accounts, and want to find out how other fellow marketers are faring, so they phone me.

Never fear, we’re all in this together! And I’m happy to report that on the whole, our lives are getting easier. Based on my rough industry survey, here goes with some preliminary Panama findings. I’ll focus on the period after February 5, 2007, when the new ranking algorithm (Marketplace Design) kicked in.


Click-through rate observations

Right off the bat, click-through rates (CTRs) in Panama were much higher than in the previous system (with all factors that could affect CTRs being relatively constant). Could there have been something wrong with the calculation of CTRs in the old system? In the legacy system, I noticed some of our best converting terms had very low CTRs (sometimes as low as 0.09 %!) and the figures were very different from what I was seeing in other PPC search engines. With the new system, CTRs are in an acceptable range and more along the lines of what’s happening on other PPC engines. In the past, Yahoo! may not have worked out this bug as there was no real need for CTR data in the bid-to-position landscape.

Notwithstanding this, the new system seems to be producing genuinely higher CTRs. In our analysis of numerous accounts, we have seen increases in CTRs since the launch of Panama. There is obviously a huge disparity across accounts but the average increase for our clients has been in the 8 percent range. This data has been recently supported by comScore.

An overall better user experience seems to be contributing to better CTR figures. And better back end systems like improved matching technologies are helping this along. In the old system, terms in Y!SM accounts did not always trigger ads when they should have (like when there was ample advertising inventory available). Take a look at the following example. In March of 2006, I tested variations of the term “buy a cocktail strainer” like “get cocktail strainer,” “purchase cocktail strainer,” “cheap cocktail strainer,” “purchase cocktail strainer,” “discount cocktail strainer,” etc. (all of these examples are commercial in nature). In March of 2006, the ads displayed 50% of the time. In my recent test, the ads appeared 63% of the time.

CPC (cost per click) & CPA (cost per acquisition)

The following data is from February 6 to March 6 2007 (inclusive).

In many accounts, we’ve also noticed average CPC (cost per click) & CPA (cost per acquisition) have decreased. In most cases, CPC and CPA have decreased in the 6 to 7 and in the 7 to 8 percent range respectively. In some accounts, we’ve seen significant CPA and CPC decreases but it’s far too early to draw definitive conclusions. The initial Panama data looks promising and I hope to share more information on this in future articles. All findings seem to be consistent with data collected by AvenueA/Razorfish in their initial Panama report.

I’ve also noticed initial volume in our accounts is up by an average of 8 percent in Panama. It will be interesting to see if this trend continues.

What’s next for Yahoo Search Marketing?

The migration of accounts outside of North America has just begun. Yahoo is starting with Japan and other countries will be announced soon.

Canada is one market that Y!SM always served poorly. To buy exposure for Canadian English-speaking searches only, you would have had to use the US program, basically. The inefficiency worked both ways: Many North American advertisers were losing 12-15% efficiency because their ads were showing up in Canada (didn’t know that, eh?), but for Canadians, it was prohibitive—they would have lost 85-87% efficiency.

Now that Panama is live for Canada, Canadian companies can actually use the system the way it was intended. The ranking algorithm is already having a positive effect on the campaigns of certain advertisers north of the border. Off the record, Yahoo staff commented to me on the example of a well-known financial company with a strong brand and a heavy radio ad campaign, creeping higher in the ad rankings without raising its bid, mostly because of its strong CTR. This “AdWords-like” effect is a welcome sight for smart advertisers.

Note to US advertisers: Your ads might still be showing up in Canada. Turn off Canada unless you want your ads to show there.

With the announcement of the new Click Quality executive, Yahoo! made two interesting statements related to traffic quality:

1) Yahoo! will introduce quality based pricing. With this advertisers will pay less for terms deemed to produce lower quality traffic. This is similar to Google’s Smart Pricing technique, but it’s notable that Google only applies it to content targeting. It’ll be interesting to see whether Yahoo extends it to the core sponsored listings. As advertisers we feel that this type of discounting should be no substitute for rigorous evaluation of publisher partners, but as long as they’re redoubling their efforts to filter in multiple ways, then there is certainly nothing wrong with a discounting feature. Indeed, advertisers welcome it.

2) Advertisers will be able to exclude sites they don’t want to advertise on. Given Yahoo!’s past silence on this issue, we’ll be watching closely to see how this is implemented, and when.

Y!SM has started to roll out demographic targeting features for the content targeting program. They’re currently hand-mapping for certain verticals and hope to have the feature fully automated in 6 to 8 months.

My phone friends and I are still in evaluation mode. While Panama’s core functionality has had a significantly positive impact on the advertiser experience, it’s clear that a lot of specific features are still ramping up. And it really is too early to judge the complexities of the ranking algorithm and other finer points. I’ll tackle some of these finer points in upcoming columns.

Mona Elesseily is an internet marketing strategist at Page Zero Media, focusing on paid search campaigns and conversion improvement. She’s also author of Page Zero’s Unauthorized Yahoo! Search Marketing Handbook. She is currently working on the Panama version of the Y!SM Handbook (forthcoming May 2007). The Paid Search column appears Tuesdays at Search Engine Land.


Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.


About the author

Mona Elesseily
Contributor
Mona Elesseily writes extensively and speaks internationally on search & online marketing. She is the Vice President of Online Marketing Strategy at Page Zero Media, where she focuses on search engine marketing strategy, landing page optimization (LPO) and conversion rate optimization (CRO).

Get the must-read newsletter for search marketers.