Google Updates and SERP Changes - June 2010

[Hissingsid]

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

I've been trying to work out what happened on May 17th and May Day. I seem to spend a lot of time analysing on-page factors and back link profiles for my own ranking pages and those of competitors but have fallen into the trap of focusing on a few big traffic volume terms.

Recently I've been looking at some second tier terms and this has been much more enlightening. This has also made sense of what is happening for the big volume terms.

In site linking seems to be out ranking external back links if done in a particular way.

The pattern I'm seeing is this, when all other factors are roughly the same, pages on sites where the site is organised in such a way that there are a large number of back links containing the keywords in anchor text within the site. ie Sites with many pages each page focusing onto one page with the same anchor text and very few back links from external sites are out ranking those that have a better external back link pattern.

So sites that have Javascript or CSS drop down navigation systems with the same links on every single page seem to be doing very well.

The other thing I'm noticing and this may confirm what I'm saying above. I'm seeing SERPS loaded with home pages.

Anyone else noticed similar patterns?

Cheers

Sid

[edited by: tedster at 11:32 am (utc) on Jun 1, 2010]

[asusplay]

One of my sites actually experienced a downwards trend beginning June 1/2 where it lost many of the long tail keyword traffic that it used to get. The site was almost unaffected from the May update, traffic down by about 50% since June 2.

Whitehat site with lots of useful unique content. Not sure what to do right now.

Same thing happened to me. I started to see a slight decline in traffic around the 31st May, but nothing out of the ordinary and in the next 3 days a steady decrease so now it is half what I used to get. I have noticed most of the log tail phrases my site used to get hits for are no longer there in the stats.

No sure what to do now but sit and wait...

[AG4Life]

The same site that I posted about above seems to have gone back to *post* May 17th numbers again today, ie. the bad set of numbers. No changes have been made to the site throughout May and June.

I have to post a correction on this. Google referral traffic is down, but only moderately, and still above May 17th levels. However, non SE referrals have dropped dramatically, almost from every website that is linking to my site. Everything in the top 10 (that's 10 different websites that link to us), have dropped on average 20% compared to even the May 17th week, 30+% compared to pre-17th. I have confirmed this with my own on-server stats software, so it's not a GA bug or whatever.

It looks like other sites are being hit hard today by another round of traffic drops.

[vandread]

I have the nagging feeling that something has been completely broken in that update. When I post a blog article now my root domain ranks for it when I make a title search (with and without ""), not the article page itself. Even stranger is the fact that the root domain does not usually rank first in Google, scraper sites do that have copied the article from by blog. That cannot be right, can it? This changes once the articles reach page 2, they are then found under their article page.

[vandread]

When I check other blogs in the niche I notice that their articles appear right away. I also noticed that my posts do not rank well for partial keyword searches. The site ranks for the post title "company announced new red widget" but not for "company new red widget" while the sites that have copied my work rank just fine in Google.

[mantucket]

Is there a place where common symptoms and likely causes are being accumulated and compared systematically, so we can look for correlations? I have a hard time piecing together info by poring over these discussions and digressions...

Regarding the last few observations, my google referrals did see another drop starting at midnight on May 30. The first decline was from about 1300-1500/day to about 375-400/day, where it seemed to stabilize a bit - but then May 31 dropped again, to about 130/day. So over a 90% decline overall, a steady, sickening trend towards zero. And this is for a site that is very different from its competitors, offers a unique price-compare/alert service, and gets lots of deep links to target pages from authority sites, gets mentioned in blogs, mainstream media, etc. Basically I follow the advice matt cutts is *still* giving to small sites, to think of less-common keyword variations. I provide high-quality content (which is aggregated but cleaned and augmented) that gets syndicated.

[tedster]

mantucket wrote

If 400 of 2000 links suddenly disappear, how do you 'diff' the GWT spreadsheets to identify what's missing?

This is something like de-duping any list, except when there are duplicates you want to eliminate both of them. Then you're left with a list of only those links that have vanished. I personally might do this in some other application than a spreadsheet - but that's because I'm a bit weak on spreadsheets syntax to use in this case.

I heard a rumor several months ago to expect a major Google shift in the backlink area somewhere around June - and from some of these reports, it sounds like that rumor may have been on the mark.


AG4Life wrote

Google referral traffic is down, but only moderately, and still above May 17th levels. However, non SE referrals have dropped dramatically, almost from every website that is linking to my site.

This might point to a re-weighting of backlinks that target THOSE sites, which then creates a secondary cascade effect to your site.


mantucket wrote

It's almost as if some adaptive alg is testing my pages by exposing them on certain days, then shutting them down if they don't rank well enough or get clicks.

You may be on to something here. In the Webmaster Tools query term report, I am seeing an assortment of different PAGES being ranked for the same keyword - a term that historically had just one "winner" page on the site. This is in addition the the ranking position shuffle that get resported.

Having these other pages rank, even some of the time, has impacted the "winner" page's dependable performance for the site involved. It's not "strong AI" as in Skynet, but it surely does look like an adaptive algorithm. There seem to be waves of changes that are undulating through the SERPs, and it could take a lot of data, Stieltjes integrals, Fourier transforms and the Reimann zeta function to analyze it all ;)

[sGroup]

we are still hit hard and no sign of any shift upwards.

[proboscis]

If it was a major shift in the backlink area for those hit around June 1, then what is the next step to regain traffic?

Or, how do we determine if it was a backlink shift?

[mantucket]

tedster

I am seeing an assortment of different PAGES being ranked for the same keyword - a term that historically had just one "winner" page on the site.

I am seeing that too, in fact I'm seeing something very unusual in how goog is matching keywords against cached content and I'd like to know if any others can confirm. Here's an example.

Normally google was very good at returning the best match on my site for a search involving my (unique, non-keyword) domain name. Now I get 3 or 4 less-relevant pages from my same site before the direct-hit inventory page appears. I am wondering whether the value of keyword-in-URL has been cut for certain cases.

So for domain 'example.com', the search:

"large blue widgets boston example"

Would almost always return this URL first:

widgets.example.com/?item=large-blue-boston

But now, it returns several 'parent' pages or geographically adjacent pages which link to the item page.

Now one really odd thing is the matching I see when I click "cached":

It says "the following terms appeared only in LINKs to this page: 'example' " ...

I.e., it's not matching 'example' as part of the URL, and not highlighting it!

As best as I can recall, the cache-match used to count and highlight keyword matches in the URL... or maybe not? Can any of you remember for sure?

Now the other odd thing - the domain name, 'example' - is not a keyword, and does NOT appear in any link text, except perhaps where a raw link gets posted. I'll double-check that - but the in-page matching which determines the second-order ranking seems 'broken' to me.

Thoughts?

[edited by: tedster at 7:13 pm (utc) on Jun 5, 2010]

[edited by: mantucket at 7:47 pm (utc) on Jun 5, 2010]

[mantucket]

@ohno, re site speed - there is a thread in the google forum where devs working on the page speed feature are responding to user input. I too am seeing very bizarre ratings from WMT - it says we're slower than 68% of sites based on google toolbar response times, but alexa says we're faster than 80% of sites!

I've moved to a faster server, removed about 4 .js files, moved graphics to CDNs, etc - things that could only speed up a page - yet the GWT rating got worse!

[outland88]

It's almost as if some adaptive alg is testing my pages by exposing them on certain days, then shutting them down if they don't rank well enough or get clicks.

If you analyze the complaints for the past few months this almost becomes a central theme whether it be in the natural results or just reading the Adsense threads. Basically a feast or famine schedule. Then again people have willingly fed Google so much private data it’s entirely doable. In fact Google has probably developed the most advanced system for generating income the world will ever witness. I wonder how much of that data is up for bids that nobody ever hears about. Except for the webmaster most searchers probably never realize the difference between many sites in the top 100 except for the branded ones.

after 9+ yrs with the first year being the slowest, we have not had a sale since 5/21.

we are done.

I find this post particularly disturbing in that it doesn’t sound like one of these greedy behemoth sites “bum rushing” the serps with thousands of pages to “get rich quick”. It shows me Google is hurting those who simply want to engage in business. It’s a very bad sign for the economy if you see a lot of these.

[sean22]

My website that was affected April 28 is enjoying pre-April 28 traffic levels. This is positive. I am cautiously optimistic about my other site too, but I won't confirm until I scrutinize the data on GA.

Here's the thing. Previously the rankings were devalued across the board (all of Google), and this was made evident by scrutinizing the referrals and serp pages they were coming from. Now the pre-April traffic is coming from some data centers but there are still post-April 28 ranking on other DC. Please don't ask me which ones because I am not technical in this regard. I am able to identify from State + start=page #.

Let us be patient with Google. I think we are not forsaken but maybe just maybe we were part of the collateral damage that happens with Algo updates.

Can someone please also confirm there own improvement in traffic so we know it is comprehensive. Which would mean it is a very very positive thing.

[sean22]

This is for the past 20 something hours and counting.

[mantucket]

Sean22, how much of a drop did you see, for how long? Are you saying the site just sprang back to life over 1 day? I'm not seeing any sign of recovery yet, nor can make any sense of the drastic drops in link count, impressions, why one page ranks and not another, etc. I would love to see any kind of bump up or to understand how this new secret definition of 'quality' differs from the old one, so I can adapt to it.
Time is running out here.

[sean22]

mantucket,

I saw roughly a 60% drop. These were an array of long tail Kws that were not related to my trophy KW. Very good Kws that I thought Google didn't want me to rank for. I really don't know if all of them have returned yet but I am seeing traffic and serp ranking for some of these Kws that I haven't seen since April 28th. I am also seeing pages that disappeared altogether now bringing in very long search terms that just come naturally. I relied on those pages for a while and since April 28th they very non-existent.

What I'm trying to say is my pre-April 28 ranking spots for these long tail Kws are on some data centers and so are the pages. Example Tennessee (whichever data center that belongs to).

I am aware of Caffeine rankings but since April 28 the rank of these Kws tanked across the board. Google UK, Google.com and I searched through proxies, but now the pre-April ranking is coming from certain DC and hopefully more are to follow.

Something is up.

[sean22]

My 24hr GA period will soon be done and the findings are that only one site rebounded and not both.

I only hope it is not a high feeling today followed by a low feeling.

[vandread]

No changes yet for me. Am I the only one who has a blog that was affected? My site cannot even outrank scraper sites anymore that copy parts or all of the content that I post in Google which suggests that a) the update is not complete and this is a temporary state that is going to change again once it is or b) Google ****ed up badly

I for one am done if nothing changes soon (with a top5k Alexa, Compete blog that is going down since June 1)

[ohno]

@mantucket, thanks i'll serach out the thread.

[anand84]

Am I the only one who has a blog that was affected?

I have a Wordpress blog that is affected too.

[Hissingsid]

This UK morning my main site has gone back to #1 for the main terms I target having been at #2 or #3 for 5 weeks. I am not able to say if this is because of a change in Google or, more likely, because Google is picking up the last months worth of mad on topic backlink building.

Cheers

Sid

[backdraft7]

Can you imagine the financial toll this sudden algo update is going to take on webmasters around the world?
It couldn't happen at a worse time. The global economy is tanking, and now without their usual income, NOBODY is buying. Our slow sales could be the result of YOUR slow sales as everybody stops buying while they "sit tight" and wait for the other shoe to drop. Google's revenue won't be spared either. If they think squeezing their Adwords customers into upping their daily budgets will work, I've got news for them, it won't.
This whole situation is a "perfect storm" of conditions, including seasonal low sales, the Gulf mess, an already struggling global economy and a general doom & gloom spreading over the planet. I sure hope Caffeine does what the name implies, because we all need a shot of adrenaline right about now.
Hang in the brothers & sisters, it may get worse before it gets better! IF it gets better.

UPDATE: In a very disturbing development, we have lost 50% of our site traffic over night. I still rank #1 or #2 on over 1800 long and short tail searches, yet it is as if the entire world has turned to zombies, and I was watching "I am Legend" last night! No kidding! How could this be happening?

[mantucket]

backdraft - Thing is, for each business shut down due to radical changes like MayDay, some other business picks up that traffic. It wouldn't be a problem if we had healthy competition among search engines, and google were say 1 of 100 equally carrying the ecommerce communications of the nation.
You are correct that it probably disrupts the economy to the tune of 100M's of $s - and right now nobody is totalling or tracking the impact. Because of their dominance, google can cause such disruptions with impunity. Without any means of recourse, biz owners are still forced to prioritize spending on google, who keeps the rules secret.

I have a few friends in DC who say google is of extreme interest right now as far as potential antitrust regulation, and especially with this MayDay update. What needs to happen somehow is for impacted businesses to band together and total up the financial impact of this change. Anyone care to create a site with a big common spreadsheet? Oh I know - just use google docs :-)

[backdraft7]

Mant: I'm not sure many business ARE picking up. It seems most of not all are suffering because the new sorting is taking longer than BP's containment effort in the Gulf. Not to digress, bit it still baffles me on how I can still rank highly, yet sales conversions dropped like a rock over night on May 17th. It's not so much that my listings are bad, it's that the data centers seem to be doing something other than serving results.

As far as antitrust action, it's probably way over due, but who will pay? Probably all the little guys, like us in the form of another extreme shuffle session.
As far as losses. I'm a very small company and might do around 100k annually. Estimated losses so far could be approximately 4k to 5k per month based on a previous 5 year trend. If this keep up for the next 10 months, a total of 40 to 50k. Multiply that by 10M sites (only 5% out of an estimated 200M sites in 2008) for a conservative figure and you've got 500 billion dollars. (I'm just whipping these numbers out there, so don't whip me). I'm sure the government think tanks could pull a better estimate, or not...
I prefer to remain positive and hope that this is only a temporary situation, but the implications of a monopoly are interesting if not shocking.

[mantucket]

Temporal or RSS-generated links?

I came across a blog post from an SEO type which mentioned something about 'temporal' links having been discounted in this MayDay debacle - does that resonate with anyone else who has been impacted?

My site generates event-related feeds which get syndicated (by quality, authority sites). These links are transient and point directly "deep" to specific pages. The links are more than editorial in nature; they are functional in that the event pages allow users to instantly set reminders, alerts, etc.

So one theory I have is that google suddenly reversed its view of of highly specific links which are transient ('temporal') or composed in a regular way, as with an RSS feed which generates titles.

Do any of you who were impacted have similar structures?

My links generate real visitors who use the functionality- but if google reversed their rules on this kind of link, I'm in a tough spot.
There are solutions, such as to

- direct all syndicated event links to one 'parent' page
- attract lots more links via syndication.

Either way the effect on how I appear in the SERPS would be 'spammy' - i.e. less-relevant pages would rank, instead of the highly specific 'long tail' ones.

Thoughts?

[mantucket]

backdraft - I think the moderators want to direct meta-google discussions to another thread, how about starting one on monopoly-related issues ie whether companies who provide a public service must also be accountable about disrupting commerce, etc. FWIW, my small biz sounds like about the same size as yours. Smaller enterprises like us cannot afford to inve$t months of time diagnosing the workings of one traffic channel... large companies can. So the secretive policies favor larger companies - what we need is 'search neutrality'.

Thing is, if google (or any search company) were required to at least respond/defend/explain a major disruption to impacted business, it would be a financial liability and would cut into their profits.
Yet smaller search companies, which need to compete, do provide more transparency. Microsoft always flags sites which it penalizes, for example.

[mantucket]

HissingSid

It doesn't sound right that impacted sites would suddenly recover after a month, after what matt cutts said regarding the mayday update being permanent, etc. I suppose one might ask matt whether the mayday update is finished rolling out, on the theory that the rollout itself temporarily trashed everything.

It's sad to think that the main way to counteract the MayDay changes will be massive linkspam, yet the competing sites who still rank all seem to be doing just that.

[backdraft7]

Man: I don't want to be an instigator on the monopoly issue, so let's leave that to the powers that be.
As far as the update - I concur with the "temporarily trashed" theory. I've been through so many of these, the last bad one being in 2005 (can't recall the name) you'd think I'd have more faith. Usually the recovery is a month, but could go on for 6 months or more. These shuffles can be a good thing, as it makes us examine our sites which usually results in improvements. I won't be doing any black hat in an attempt to regain ground. I'm just re-reading the GWMGL's again and looking at my site as if I am a visitor, rather than a webmaster looking for high rankings. The rest may then just fall into place.

[mantucket]

Ha, if everyone is afraid to talk about the elephant in the room, we will continue to be crushed every time it sneezes :-) But the issue is already well past being instigated - my friends said there were recent articles in the NY Times mentioning potential antitrust action etc.

At any rate I'm doing the same as you, focusing on improving my site - but I've never seen a decline so severe that was such a steady trend before. 1 year ago I had a similar problem but it turned out to be a bug on my end where site was sending blank pages to googlebot! That was a sudden turnoff and just as sudden recovery, but I spent 3 months assuming it was some kind of 'penalty'.

So I have a plan, and I hope google will regain its regard for my pages - but if I guess wrong about google's sudden re-definition of 'quality', the impact is huge. What I need is more guidance from GWT about what caused the 90% drop in traffic in my case.

Also, I'm confused by the conflicting messages goog is putting out about this update. On one hand, matt cutts says MayDay was an alg change by the Search Quality group that impacts ranking of long-tail sites by 'quality', a vague and subjective criteria. On the other hand, cutts is still recommending to smaller sites that they cultivate traffic from long-tail keyword variations (see the recent google IO site review session video). It's as if goog is too big for one hand to know what the other is doing...

[Hissingsid]

@Mantcket

It's sad to think that the main way to counteract the MayDay changes will be massive linkspam, yet the competing sites who still rank all seem to be doing just that.

I'm seeing sites with well structured internal link structure using target anchor text doing better.

Also backlinks from on topic sites still weighing higher than garbage backlinks despite what open site explorer says.

Cheers

Sid

PS I still think there's a way to go in terms of re-instating human applied penalties. I'm seeing a few sites that previously had -6 or -10 penalties back in the top 5.

[mantucket]

Sid -- what do you mean by 'human-applied' penalties? I understood that the mayday debacle is strictly an algorithm switch, not targeting individual sites but reversing the definition of 'quality' based on certain specific characteristics or matching rules...