Implication of Flattening Keyword Prices

Jason Jones:  Fathom Online reports that year-over-year growth in U.S. keyword prices remains in mid-single digits.  This means that Google, Yahoo, et al, are losing a growth-driver and must grow their paid search businesses through international expansion, breadth of keyword buys, and market share gains.  The flattening of keyword prices explains both the tepid growth of all search players except Google, and the recent pick up in M&A interest in the branded-advertising business.

Where will advertisers shift their focus as ROI's begin to stabilize in the paid search business?  Probably to contextual & behavioral advertising.

          Price    q/q    y/y
1Q07    1.46    -3%   5%
4Q06    1.51    2%     6%
3Q06    1.48    17%   3%
2Q06    1.27    -9%   
1Q06    1.39    -3%   
4Q05    1.43    -1%   
3Q05    1.44 

Peter Hershberg of Reprise Media responds (Text adapted from comment below and from this Jan 2006 piece on Reprise's site): 

[FROM COMMENT]: The fact that people continue to reference Fathom's KPI is mind-blowing to me...  While it is *possible* that keyword prices are stabilizing, there's no way anyone outside of the search engines themselves know with any degree of certainty.  Along those lines, it's worth noting that in January of 2006, Fathom's KPI suggested that keyword prices for 2005 had fallen by 16%. Google's stock more than doubled over that time period...

[FROM JAN 2006]: Fathom's sample size of 500 keywords is not representative of the entire universe of advertising opportunities available on Google. There are literally millions of unique keywords being purchased in the search marketplace today. Furthermore, the KPI doesn't include either proper or brand names -- keywords that typically deliver significant volume at higher average CPC's. (In fairness to Fathom, they acknowledge this shortcoming in each of their reports. Whether or not the press picks up on it is another story...)

Second, and more importantly, Google's stock continues to rise because it's CPC's simply ARE NOT falling. And even if they were, there would be no way for Fathom, or anyone else, to know that was the case.

How can I be so sure? Because of Google's "Quality Score," a topic I've written about on several occasions in the past.

For anyone not familiar with this concept, Google defines Quality Score as "the basis for measuring the quality of your keyword and determining your minimum bid. Quality Score is determined by your keyword's clickthrough rate (CTR), relevance of your ad text, historical keyword performance, the quality of your ad's landing page, and other relevancy factors."

This essentially means that two (or more) advertisers could be required to bid completely different CPC's to occupy the same position against the same keyword. In other words, the advertiser with the "better" Quality Score might have to pay just $.10/click for the top position against the keyword "wireless accessories," while an advertiser that's been penalized for poor ad copy, a landing page that's light on content, or some other "violation" would be required to pay $.50/click for the same position.

      

Comments

[Peter Hershberg]

There are any number of issues with Fathom's KPI -- something I wrote about over a year ago: http://searchviews.com/archives/2006/01/quality_score_k_1.php

While it is *possible* that keyword prices are stabilizing, there's no way anyone outside of the search engines themselves know with any degree of certainty.

Along those lines, it's worth noting that in January of 2006, Fathom's KPI suggested that keyword prices for 2005 had fallen by 16%. Google's stock more than doubled over that time period.

The fact that people continue to reference Fathom's KPI is mind-blowing to me.

[Jason Jones]

Peter - I agree that Fathom's KPI has to be taken with a grain of salt and I should have mentioned that in the blog post. Fathom says as much in their disclaimer: "The Keyword Price Index® should not be used to make financial projections or be considered a proxy for the financial state of the Search Marketing industry."

However, I am not sure if total disregard of their index is the solution. I use it as a singular data point when triangulating the financial health of the search industry and I think many other people do the same. I am always looking for additional data points. Is it possible to extract keyword pricing trends by using Google's Quality Score? If so, I would love to learn your methodology. Jason

[King Troll]

Henry, some interesting info I have discovered this weekend.

As you know Google search was giving me about 70% of my traffic. Sales were very low. Yahoo was about 15-20% with low sales. i been paying about $150/day between the two recently.

I submitted my items to Google Product Search. Traffic was not as much as search but I got way more buyers for free.

Let's give some basic numbers. Google/Yahoo Search =
$200/Day. About $175 from Google.

Google Product Search. Additional $100/day for free.

Yahoo Product search is a fucking Gold mine bro. The Yahoo shopping site. They charge $.25 per click, but fuck bro I've been getting another $300 average per day from it. I am about to try out Shopzilla too.

Maybee Amazon when i get time.

[Peter Hershberg]

Jason - Unfortunately, it's impossible to assess what each individual advertiser is paying for a given keyword b/c of Google's Quality Score (and the other quality-based bidding programs all major search engines are using at this stage).

For instance, Google may indicate that a keyword requires an initial CPC of $.25, but subsequent performance (or lack thereof) may cause Google to disable that keyword. Reactivating it may require a bid of $1.00, or even higher in certain instances. There's obviously no way anyone other than the advertiser and Google can have transparency into the fact that the advertiser is now paying a higher CPC for a keyword that appears to have a value of $.25. Conversely, there are ads that go live with initial CPC's of several dollars. Those CPC's can go down to cents/click over the course of the day as "value" is proven. -- Peter

[mb]

How many times does the fable of the declining keyword prices need to be debunked? The fact is, you really can't tell much of anything from the an index of prices of common keywords.

Not only are there other factors like the Quality Score, but there are lots of keywords not in the index that drive search engine revenue.

We've seen the keyword price index go negative, while at the same time Google's revenue skyrockets.

The index is worthless as a predictor of anything an investor cares about.

[aizhengw]

癌症 肺癌 胃癌 肝癌 食管癌 肿瘤 直肠癌 结肠癌 宫颈癌 乳腺癌 脑瘤 甲状腺肿瘤 胆囊癌 胆管癌 前列腺癌 白血病 鼻咽癌 肾癌 恶性淋巴瘤 皮肤癌 喉癌 舌癌 胰腺癌 膀胱癌 神话 家庭建康 心碎了无痕 pallyboy 幸福 医药类 我的快乐 胃癌的博客 天天健康