flinky

Did Yahoo know (and tell us) that Google would report a positive surprise?

noreply@blogger.com (flinky) — Sat, 19 Apr 2008 18:46:00 +0000

We now know that concerns about Google's growth and monetization (on the whole) were a bit overblown after Google reported strong earnings significantly ahead of expectations after the market closed on Thursday April 17th. (My discussion in a previous post that better monetization would offset lower clicks was partially right. It seems paid click growth rates did not fall as much as the market had expected though.)

I remember thinking on Thursday during the day that it was strange that Yahoo chose to disclose on that particular day that they "were close to an outsourcing deal with Google". In fact, on that day before Google reported, Rob Hof at BusinessWeek.com wrote:

"Oh, please. An initial test of outsourcing Yahoo’s search to Google showed positive results, according to an account in the Journal. Hold it; didn’t that test just start this week? Wasn’t it supposed to go two weeks? Didn’t we already know that Google ads work better than Yahoo’s?"

He goes on to discuss why Yahoo might have done it early in the context of Microsoft, and of course it makes sense.

But here's what I'm trying to get at: Yahoo could have waited a day. If the Google earnings were strong, they could have pushed the story to the next day, and that would have been fine I think. If the Google earnings were weak, they could have waited, or done it anyway, or messaged it differently or any number of things.

It seems that only possible benefit of doing it that day before Google reported, instead of waiting, would be if you knew that Google would beat.

comScore's interpretation of their Google click data

noreply@blogger.com (flinky) — Sat, 01 Mar 2008 02:44:00 +0000

comScore has now released a blog post interpreting the data mentioned in the previous post. They seem to be supporting the idea that (1) it's deliberate and (2) Google has likely been successful in offsetting the (potential) revenue loss with a higher average CPC for fewer clicks. It would have been nice to see the data from the graph in the previous post plotted on the same graph that comScore shows in their post (the click rate over time). Maybe comScore doesn't have that kind of data.

In any case, the graph in the previous post also seems to show two regimes of change in the year where the average CPC increases. Eyeballing it, it looks like the time periods correspond roughly to the two regimes of change in the graph from the comScore post. But here they are together:

Google data: clicks versus cost per click

noreply@blogger.com (flinky) — Tue, 26 Feb 2008 23:01:00 +0000

The (further) fall in Google's stock today seems to have been triggered by a comScore report saying that growth in the number of sponsored clicks has decelerated a lot. See the post at Silicon Alley Insider for details.

However, on the other hand, consider recent trends in the average cost per click according to EfficientFrontier:

Google's average CPC certainly seems to be trending higher, so this could be a significant counterbalance to a drop in the number of clicks. I.e. if YoY sponsored click growth is indeed flat, this data would still put revenues up 20% or so, maybe more if you maintain the slope of the average CPC. Whether such revenue growth matches expectations, I don't know (I haven't looked it up and also there are other factors that play into that).

It may be a deliberate attempt on Google's part to reduce low value clicks (some of which they call "accidental"), with the idea that they can get the average CPC up.

version 2: Reading The Wall Street Journal For Free (made easier)

noreply@blogger.com (flinky) — Sun, 10 Feb 2008 03:51:00 +0000

Ok, playing around a bit more, here's another bookmarklet that appears to work for everything (that you can find without logging in) at the WSJ. It would be easy to prevent this behavior from working since it is more of a simple technical issue than fundamentally about segmenting the user base.

gWSJ.v2

(See this post for instructions on using a bookmarklet.)

Reading The Wall Street Journal For Free (made easier)

noreply@blogger.com (flinky) — Sun, 10 Feb 2008 02:32:00 +0000

Danny Sullivan points out on his blog that (at least for now), it is possible to get the full content of some Wall Street Journal stories if you find them via Google News. These same stories are not fully accessible when you get to them via the WSJ site itself.

It's not clear how sustainable this kind of customer segmentation/price discrimination is on the web.

For example, it took me about 15 minutes to write a little bookmarklet that makes it easier to see if a WSJ story is in fact available for free via Google News.

To try it,

1. just drag the link below labelled "gWSJ" to your browser links toolbar (it should work in Firefox and Internet Explorer).
2. go to a WSJ teaser page like this one
3. then just click on the "gWSJ" bookmarklet and see if it works for you

Here is the bookmarklet: gWSJ

It doesn't work on all teaser stories because it appears that not all stories are in fact indexed in Google News.