Google Updates and SERP Changes - April 2010

[wmxpert]

< continued from [webmasterworld.com...] >

We are in the same situation, since March 15, 2010 our traffic is continuously declining with the average of 6% every week. In current week it gets worst, approx. 15% decline!

Overall, 30% to 35% traffic has been reduced compare to peak week of January 2010.

"site:" is also showing huge drop in index pages, though it only provide the avg. not actual.

Don't know what's going on, Is it really the effect of Caffeine roll-out?

[edited by: tedster at 9:46 pm (utc) on Apr 1, 2010]

[internetheaven]

30% to 35% traffic has been reduced

Can we get all these moved to the "I need a shoulder to cry on" forum?

If you are experiencing dramatic fluxuations then posting such is only helpful if you tell us what is unique to your site that might be causing it so we can pick up a general theme. e.g:

I don't know whether is caffeine or not but my vbulletin forum with 120K site:mysite.com pages on Google is dead in the water. Have around 60 Google spiders constantly in my forum at the same time all day, but no true visitors for 20 days now. Grankings dropped overnight.

[tedster]

We've now got a thread where everyone who lost traffic [webmasterworld.com] can compare notes.

[Pass the Dutchie]

Google clearly stated that caffeine would be rolled out after the (Christmas) holidays. They now only have a few days to go before Easter. If they don't roll it out by then they either misled the public or they have to admit they have a problem.

Hoping for a happy Easter!

[sjxs]

I'm for the first time seeing Caffeine results on google.dk and that's both from my home connection and on my HTC hero cellphone (android).. I'm getting the Google IP's: 74.125.77.104, 74.125.77.147 and 74.125.77.99

Furthermore I'm starting to see some traffic from keywords where my sites ONLY have rankings on Caffeine, i.e. not within top50 on non-caffeine results.

[sandboxsam]

They said the rollout would be after the holidays.
Now, the question is; which year were they referring to?

[dusky]

If they meant after the holidays as X-MAS and Easter, they are bang on and the roll-out should be steadily sweeping the net as I remarked in my previous thread. I expect it'll take few days to level out on all DCs worldwide, my guess would be, the 15-16th April days should be a significant time span.

[dusky]

"as I remarked in my previous post" I mean..

Further to that, I think before that happens properly, massive temporary gains and losses are going to be noticed even more than ever before, I say this from my recent observations of some of our own sites being caf, then decaf since September and that increased on the 15th of March, some of which lost 90% of traffic for few days then returned to above normal (20% +).

I expect also those sites to suffer for few days again and revert hopefully to even better ranking. I heard and read also that some sites are still waiting for the revert-back situation, take 1script's posts in the previous pages and others on various threads here and on other forums

[restless]

on the 15th of March my traffic jumped nearly 70% to 80% til the 23rd then after that steep decline back to traffic levels prior 15th perhaps a little worse 5% down

i was on a high for 8 days!

[gouri]

I think in February Google said that it was hoping to do a full rollout of Caffeine in a couple of months.

Do you think they meant Memorial Day?

[andy_boyd]

Something's definitely cooking ... anyone else seeing a lot more results showing just URLs-only, as if they've been blocked by robots.txt?

I've checked several sites that are are showing up as URL-only for very competitive phrases and strangely they aren't blocking Googlebot.

It's a weird one. :S

[gouri]

I saw this on the Caffeine datacenter for a few long-tail phrases a couple of days ago.

Are you seeing this on Google.com or on the Caffeine datacenter?

I think the fact that you are seeing this for very competitive phrases makes this more significant than when I saw it.

[andy_boyd]

@ gouri - I'm seeing URL-only in the results mostly on Google.com, but I've also seen it on Google.co.uk. It's probably just a bug that'll be fixed fairly quickly, but I thought it was something of interest worth pointing out. :)

[gouri]

@andy_boyd - Do you think it is a bug or maybe something new in the SERP?

[tantalus]

Just throwing this out there but I had this problem with msn and yahoo a couple years ago when they changed the way they handled robots txt. All my traffic completely vanished and I thought I had been banned at the time, it was only when yahoo went the same way I figured it out. Google was always fine with the robots txt that I had. Is Google following suit?

From a quick look what I am is seeing is no "allow directive" ie...

User-agent: *
Disallow: /administrator/
Disallow: /cache/
Disallow: /components/
Disallow: /images/
Disallow: /includes/
Disallow: /installation/
Disallow: /language/
Disallow: /libraries/
Disallow: /media/
Disallow: /modules/
Disallow: /plugins/
Disallow: /templates/
Disallow: /tmp/
Disallow: /xmlrpc/

Instead explicitly allowing as in ...

User-agent: Googlebot
Allow: /
Disallow: /cgi-bin/
Disallow: /shop/
Disallow: /dev/
Disallow: /admin/
Disallow: /private/


Just a suggestion.

[1script]

@dusky

I expect also those sites to suffer for few days again and revert hopefully to even better ranking. I heard and read also that some sites are still waiting for the revert-back situation, take 1script's posts in the previous pages and others on various threads here and on other forums

Dude, your posts remain the only glimmer of hope in an otherwise hopeless situation: just today I lost another 50% of the remaining 10% of Google traffic before March 15th. Same sites affected again.

[tantalus]

1script

just took a look at your robots txt and its exactly the same as above.

Same signals as I had, slowly being delisted. Not saying its the answer, but you may want to look into it.

[1script]

@tantalus:

Dude, that's the first time I hear that you need to explicitly allow anything in robots.txt

After all, it a Robots exclusion standard! I think you are digging in a wrong direction.

Cheers!

[SEOPTI]

LOL, you really don't need to explicitly allow Googlebot.

[marketingmagic]

Is anyone seeing anything different on google.ca? Not seeing any changes for my categories...

[arizonadude]

The Allow tag is not a valid markup for the robots.txt file.

You're looking in the wrong place if you think you have to add the allow tag so Google will index you. I think your problem lies elsewhere.

None of the hundreds of sites I've got have that tag in there and they are indexed fine.

[internetheaven]

Ooh! Anyone else seeing lots of snippets being URl only even though there is a web page there? The #1 ranking for one of my terms just went URl only (very weird seeing URl only ranking #1!) but that site has ranked at #1 for nearly 2 years -- bet they are wetting themselves tonight seeing themselves listed URl only despite not having change their page or allows!

Glitch or indication?

[gouri]

I also saw today a URL only for the second listing in the SERP but not the first one in the Caffeine datacenter.

I think this makes this a little more interesting.

[wneennsdjj33]

@internetheaven
Definitely! That's what i wrote yesterday... All my pages' search results for one of my site has turned into URI only, and the exact page is now listed about at least 20 pages back... I wonder what the f*k going on... I'm going crazy...

[wneennsdjj33]

URI only i meant = URL only...
And this whole change happened on the 22th of March... for all of my top search results :S

[internetheaven]

Matt Cutts just updated his blog:

[mattcutts.com...]

now reads:

Things are on track though, and we expect to roll out Caffeine to all data centers in the coming weeks/months.

Edited April 1 to add “weeks.”

Is that a hint? Or another April Fool? The original message was posted 10th March and just said "months". But whilst the word "weeks" has been added to the post today, the word "months" was not removed ...

[internetheaven]

Also found this:

[itproportal.com...]

I found nearly every paragraph interesting! Especially this one:

There were a lot less pages (roughly half the original number) but at least they were pertinent.

[SEOPTI]

The itproportal information is more than interesting if it is valid. Thin URLs or duplicated content will become supplemental with caffeine.

Therefore I think sites which see less crawling could have less unique content but this is just speculation.

[tedster]

internetheaven, I'd find the article even more interesting if it didn't begin with "According to snippets of information gathered from sources at Google UK's HQ in London". That sounds like rumor journalism and possibly complete FUD.

But I have no doubt that Google has been struggling against "scrapper" sites and other kinds of worthless pages for a long while. Maybe they will make a breakthrough on that front, at least for a time - we can only hope!

The article also says "Bizarrely, we heard that some price comparison websites or sites that only rely on third party feeds could be affected by this." I have seen some evidence that results showing price comparison sites are being "adjusted". I'm not sure why the writer would find it bizarre - how many of those sites do you need in any one SERP, anyway?

[walkman]

Looks like this is it, especially after reading Matt's 'weeks' comments.