Google Updates and SERP Changes - September 2012

[backdraft7]

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

Historically we never saw drastic drops due to holidays...except this year, when the now daily everflux drops just happen to coincide with a holiday.

[edited by: tedster at 5:33 am (utc) on Sep 1, 2012]

[Martin Ice Web]

ohno, i know what you mean and i although think that a shareholder profit driven business is not willing to give money away. You can see that in introducing the new shopping fees or the fess for using the google maps. It is also clear the a "little" factor in ranking is adwords earnings. But as i stated before, i think the rules are new and old site with panda sitting on them are not going to get back until something changes. It is to long now for them to get back.
There is truly some false positives and i see all the things like traffice shapping, throtteling and even sales throtteling like some others here. The heck, i told it my wife, she was rolling her eyes and said she canīt believe it because it would be to conspiratorial. I still think goog found a way to get some more data then we know.
But this is the point once you have this opportunity you use it.

But i like more the idea that user metric is only the verification of the algo that is certainly base upon backlinks. Whereas this algo seems to have some severe problems. As crobb305 is observing the rotation of some sites I think this is the reset of "user metrics confirmation". All pandalized sites have this but on a different day and i a more or less heavy way. To escape from this you have to find claaarkys threshold. Finaly something tedster was talking about shallow or low content was something that hit me like a train, itīs smattering what we are adding as new content, because you canīt push something up that canīt be pushed. And there is already a non pandalized site on the top with the targeted content. The key is to invent the wheel new.

[crobb305]

As crobb305 is observing the rotation of some sites I think this is the reset of "user metrics confirmation". All pandalized sites have this but on a different day and i a more or less heavy way.

I definitely think this is realtime evidence of testing or machine learning. I've watched this rotation for several weeks, and it is predictable. These sites never seem to move above page 3 or 4, but they may eventually. They seem to be cresting a little higher with each iteration.

[klark0]

Anybody seeing movement on Penguin-hit sites ?

[Martin Ice Web]

@crobb305, but you need a signal to test something where you start from NULL?! You know like to push on a button to say go for it. And this is the reset on thursdays.

[Jez123]

Anybody seeing movement on Penguin-hit sites ?

Nope. What are you seeing?

[ohno]

The only movement I see is bloody down again. Abysmal day, I'd swear in was Thursday!

[gehrlekrona]

@klark0, lower than usual, but what's new these days...

[themistral]

Just an quick update on another site I'm working on. From Sunday until today, a UK based ecom site is seeing a massive increase in US traffic - now 55% of all traffic - previously it has been around 2%!
All useless traffic as it's not likely to convert for the products this site is selling.

[Splugged]

Uh...

Something is happening.
Since this morning (local time) I see HUGE movements in my SERPS.
My main keywords are moving up and down like hell: +368, -310 two/three hours later... +137, -220... +423, -238... etc...

Anyone is experiencing the same?

[ohno]

@themistral, same here, seeing loads of US traffic to a site that sells goods specific to the UK! Still it beats traffic from India, oh, hold on, that pops up now & then too! This foreign traffic lark is getting silly now, soon it will be all we have.

[Jez123]

Nope - where are you splugged?

[crobb305]

Something is happening.
Since this morning (local time) I see HUGE movements in my SERPS.
My main keywords are moving up and down like hell: +368, -310 two/three hours later... +137, -220... +423, -238... etc...

Anyone is experiencing the same?

I sure wish. The litmus test in my industry are the ridiculous number of press releases in the top 100 and number 1 site with foreign backlinks that was just registered 60 days ago (very competitive phrases). I do see some sites rotating upwards and downwards from page 15 to page 4, but this takes a few days to a week to complete. Just my observations.

[Splugged]

@Jez

Site is hosted in US, i'm in Italy.
I'm tracking KWDs on G dot com domain.

[brianp]

Something is definitely in the works...here's what I'm seeing. All our related to e-commerce searches in an established industry.

For keywords that have been fairly stable for a number of years I'm seeing major players being downgraded to page two and three. Most of these major players product pages have been pretty static for a year or more.

Products that we have recently added that were ranking well individually for their keywords are now being outranked by category page that has the product listed.

Many many keywords are seeing upward movement through every method used to check them (goosh.org, logged in, logged out, mobile devices) with few staying stable and a few dropping a position or two.

Either something is being "tested" right now, or this is a preview of an update that is in the works. Not sure which I'd prefer at this point.

[Jez123]

Hi brianp. Where are you? I don't see any movement in my SERPs

[klark0]

@splugged, that's exactly what I'm seeing on a Penguin hit site. Keywords are up and down. Mega Sitelinks appear and disappear for brand terms.

[brianp]

@jez123 in the USA

[SnowMan68]

I am in the US and on our main ecom site I am seeing a slight increase in G referrals. We had actually seen a slight decline 5-10% over the past couple weeks, with an increase in our bounce rate during that time frame.

With the increase in traffic today I am noticing a decrease in the bounce rate of 5-7%. Weird.

[SnowMan68]

In reference to my last post.

I am also seeing about a 7-10% increase in both time on site and pages viewed.

[backdraft7]

I definitely think this is real time evidence of testing or machine learning. I've watched this rotation for several weeks, and it is predictable.

Still seeing a definite daily pattern since the early part of September. One morning conversion, usually around 7 am, then nothing all day until maybe 5pm, then a few conversions trickle in. The lunch crowd is no more. By the end of the day it usually ends up with a strictly limited number of conversions, what's funny is that if I'm short a sale, it magically pops in at around 11:45 pm.

Definitely clamped traffic.

[Wilburforce]

I don't see how Google could be influencing time on site and/or bounce rate (or even referral rates, for that matter) directly.

These effects must come from relative phrase placement. Any "choking" of referrals could only come from stopping all listings for a given phrase (all phrases?) in search results after a count of returned listings, and high bounce rate and/or low time on site must be the result of good placement for phrases the user does not find relevant (coupled with loss of position for phrases the user would find relevant).

That Google's once superb grasp of relevance has suffered through Panda, Penguin, and the other countless algorithm tweaks is not much in doubt on this thread or my own experience.

Trying to keep people on site who are not looking for your product is pointless. Trying to get people on site who ARE looking for your product is very difficult if there is a mismatch between Google's idea of relevance and the end-user's experience.

[chalkywhite]

Im noticing time on site is up by 25% and bounce rate has dropped by 10%, looking good so far.

[topstar]

I am slowly and officially abandoning my main site. Weeks turn into months as I'm waiting for something positive to happen while doing small improvements trying to figure out a reason for a minus 950 by Panda and the dropping positions. It has wasted a huge amount of my time and money. If G decides it doesn't like your site anymore (Pagerank drop, rankings drop, 950 filters to content that ranked ok for years etc.) then there's not much you can do to get the love back.

I feel for everyone of you in the same boat.

[Bewenched]

Well, it looks like were back to spammy results again for ecommerce ... sigh.

facebook, blogspot, newegg other shopping sites all showing really high in the results.

[Splugged]

A lot of crap in SERPS right now, at least for my niche.

Ah, no. Sorry. I forgot. My site is crap. The #1 is a high quality website.

<h1>Welcome</h1>
Banner
Banner
Banner

<p>Red dress is red and good.</p>

Banner
Banner

.......

[xcoder]

WTH, geo targeting is simply off the charts.

e.g. from a site on a "com.au" domain clearly targeting local customers. Last ~60 visitors are from any other country in the world except for Australia.

To name a few:
Morocco
Romania
Argentina
Colombia
Malaysia
Greece
etc. etc.

Stupid as stupid is!

[tedster]

xcoder, if we can track down which version of Google is doing that there might be more of a clue. I assumed it was google.com being geolocated by IP address - is that the case?

[xcoder]

@tedster

It looks like they are coming from their own country SERPs

eg.
Argentina = www.google.com.ar
India = www.google.com.in
etc.

Not google.com

[1script]

@topstar:

reason for a minus 950 by Panda

I never heard of minus 900+ penalties mentioned in relation to Panda. Usually its effects seem to be less drastic (position-wise, not traffic-wise, -50 would already be dead enough) and much less specific - you'll be fluctuating between some low SERPs but not all the way back to 900+. And it would be really hard to give it a particular "minus xx" number - yours seems different.

Are you sure your issues are Panda-related at all, have you checked the -950 timing against the Panda refresh schedule posted here at WebmasterWorld?

[zeus]

1script - my panda hit site from april 2011, there I also see my frontpage swing around in the 90s pages, but I also thought maybe im out of panda and in something els.