Update Allegra - Google Update 2-2-2005

[illusionist]

My site which came back on december 26 update, seems to have disappeared again on this data center [216.239.53.99...] . Its notwhere to be found even in allinanchor, allintitle etc? I see majot change on that data center, is this a new update?

[thepast]

Rankings are returning on many DC's and also showing up for many people on the "live" G. Question is, is this pre-update data or a welcome tweak to the update graciously allowing websites to rank for their own name again?

[Kangol]

On 216.239.53.99 I see a Tip displayed on the top of the page when a search is done. It appears when you click twice on Search.
“Tip: Save time by hitting the return key instead of clicking on "search"”
Is this something new or it just arrived?

[edited by: Kangol at 12:00 am (utc) on Feb. 9, 2005]

[KrisVal]

I just had a similar thought. There has always been speculation about Google recording the clicks for a particular site. This update is one of the longest that I can remember and the results keep changing on different datacenters throughout the day. Is it possible that Google is A/B testing the update to see if CTR falls off for initial query? For example, on average how many queries does it take a user to find a website of interest? Results A = 1.5 Queries before a click while Results B = 1.75. Hmmmm.... If I were them, I would do that before finally rolling out anything. Choose the one with the best CTR.

[Kangol]

Its interesting. I do not see any tracking code but I am sure that his can be done without by G. There should be lots of data to gather and lots of calculations to be done.

[itloc]

"Its interesting. I do not see any tracking code but I am sure that his can be done without by G. There should be lots of data to gather and lots of calculations to be done."

It can be done very easily - but for sure they are not doing it now "realtime" to change the serps. If they collect this kind of information (and i would) they have done it earlier to collect a LOT of user selections which they add as a factor to rate a certain page.

One should also not forget the Google Toolbar ... this little baby delivers nearly everything you need regarding user behaviour / selections combined with querys..

itloc

[bnhall]

Choose the one with the best CTR

I bet they check their Adwords revenue stream as well. One would think higher CTR on natural SERPs = lower Adwords revenue.

[Liane]

Is anyone using <a certain update tool> to check the data centers? If so ... the results delivered bare no relationship to the actual search results performed on the individual data centers.

Thought I'd mention it just in case nobody else had. Do not trust these results!

[edited by: ciml at 9:20 am (utc) on Feb. 9, 2005]

[dfre]

mcdar has been a great help in my opinion.

2 things to consider though:

1) Sometimes you can refresh a query on certain datacenters and you will get dramatically different results with each refresh.

2) I believe the results at mcdar are based on a query that shows 100 results per page, which will sometimes result in slightly inflated numbers.

[robster124]

I get the feeling this update still has at least 48 hours to go...

[danny]

Here are the figures for referrals from www.google.com/search for my biggest site since the update, as percentages of the January average:

February 3: 21%
February 4: 13%
February 5: 9%
February 6: 14%
February 7: 19%

There's nothing even remotely blackhat about my site. It's static HTML with no frills - no Javascript, not even any images - except for AdSense. It has huge numbers of (almost entirely unsolicited) incoming links, from all over the place, and almost no SEO has been done on it (page titles are derived from book titles).

The only thing I can think of is that my AdSense alternate ads are causing a problem. For a while I had myself convinced that was the explanation, but the more I think about it the more implausible that seems.