Google Updates and SERP Changes - February 2011 part 2

[boirun03]

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

But, I believe that Google now prefers sites with a much higher percentage of "valuable" pages. If you don't meet the percentage value determination, whatever that is, you get whacked.

I tend to agree with Fred. This is what I am finding as well.

[edited by: tedster at 8:00 pm (utc) on Feb 26, 2011]

[ckissi]

My question is ... If we will make changes to our sites... Have we any chance they will move up ? As I don't really think its just so simple even we change almost everything from content to design , including incoming link structure (almost impossible as I do not buy links myself)

[netmeg]

Do you know how they are scraping you? Most common and garden scrapers use your RSS feed - set this to summary or short and they just pick up the first few sentences, which is "fair use", rather than your full articles. Is there any reason why you need your feed to be set to "full"?

Well I come down on the side of full feeds myself, just because I hate the partials in the various readers I use. But I make sure everything in the feed has links and references back to the original source site, and for one site, I modified Yoast's RSS attribution plugin to say something like "This content stolen from example.com; if you want to see the real thing, go there"

I actually get visits from it, and that site no longer ranks first for any of my content.

[Rhonie]

ckissi, Before I would consider doing anything with your site, I would wait for awhile, and let the chips fall were they may. You may be very surprised in a week from now. Keep watch on the post here and other places you can find. You see that websites are gaining and losing search rankings on almost an hourly basis. More are losing, but the thing is, that if you touch it, you may gain a spot or two now, but you may lose 5 spots tomorrow. This will take some time to settle.

[edited by: Rhonie at 9:22 pm (utc) on Feb 28, 2011]

[tedster]

We've got thousands of mostly negative posts here about the latest update.

That's a long standing webmaster tradition ;) Seriously, check out past major updates [webmasterworld.com]. If you only go by complaints, you'd think that Google should be on death's door by now. I can't remember any update where we were flooded with positive comments - and I bet it stays that way as long as I live.

I'll bet they would have been outweighed by positive posts had the job been done by a human.

The number of URLs and the number of unique queries that are in play today is immense. The scale of the job is extremely far out of reach for any manual effort to handle.

The challenge of doing business online has also scaled way up. The easy entry days are gone - despite what you might hear on late night infomercials. For me the constant message from Google updates is "Step up your game!"

Along those lines, I was happily surprised that an e-commerce client of mine shows a 19% increase in Google traffic beginning Thursday - spread over a couple hundred thousand keywords. They're a new client, so I certainly don't claim credit of any kind, but I thought I should represent the positive side of this update.

Sites that show improvement rarely hurry to post the good news anywhere.

[Rhonie]

tedster, Way too many are worried about this, and if they take drastic steps now, it may hurt them until the next update. You will see legit sites rise in the ratings, but it takes time. Any algorithm that has ever has been written needs a tweak here and there, and just the number of sites it has to run is simply unbelievable. Googlebots are spidering the web sites at record speed and numerous times a day, but they need to share the information also across the internet. One bot in one server farm could cause the rating of a website to zoom to the top or crash to the bottom, until the others share the information. I really think that this will have a positive effect on quality websites in a couple of weeks.

[vordmeister]

The number of URLs and the number of unique queries that are in play today is immense. The scale of the job is extremely far out of reach for any manual effort to handle.

My point was I did ask humans and they gave me decent results. The google algo didn't get close.

While I take the point that it would be tricky for Google to employ enough humans to deal with my obscure searches (if they were not obscure searches why would I be searching?) my results on things I don't know about are becoming increasingly unhelpful. Google used to be useful to me, not so much these days.

I don't know what their answer is. I'm a searcher not a search engine algotithm writer. I can only highlight the problem.

I'm actually finding things in AltaVista when I can't find them elsewhere.

[tedster]

It's a good point - but I'm not saying it would be tricky to do the job manually. I'm saying it would be impossible. What was that number from last year - 30% of Google queries every day were never seen before? Trillions of pages?

For those wondering about sites that gained, I tracked down some data from SEO Clarity [seoclarity.net] on the big winners over a 60,000 high-volume keyword dataset

BestOnlineCoupons.com - 88.6%
Answers.Yahoo.com - 25.6%
eHow.com - 19.7%
Etsy.com - 19.5%
Sears.com - 18.7%
DrFosterSmith.com - 13.2%
Target.com - 11.5%
Walmart.com - 11.2%
NexTag.com - 7.0%
Amazon.com - 6.7%
Wikipedia.com - 6.2%

[vordmeister]

Sorry, differences in language from the other side of the atlantic. Tricky does mean imposible in that context.

On the other hand it wouldn't take a lot of humans to take a look at your list. :-) I see dudes in there that have been polluting my obscure results.

[TheMadScientist]

Percentage Increase:

eHow.com - 19.7%

ShakeMyHead ShakeMyHead ShakeMyHead

It's really tough to not get all conspiracy theoryist when I see numbers like that. It's just really really really tough.

[chrisv1963]

ShakeMyHead ShakeMyHead ShakeMyHead

Altavista is doing better than Google right now. Better results and less eHow-type copied junk content.

[gyppo]

Interestingly one of the sites that gained the most is doing some very questionable link building using Hit Counters *sigh*

[kd454]

"A person who works with Cutts recently told us "there's a department full of Ph.Ds at Google that exists for the sole purpose of getting Demand Media out of the search results."

Read more: [businessinsider.com...]

Moral of the story, if your worth billions, Google will cut you a break and actually let you slide right up top, but if your trying to just pay the bills,,ehh not to worried about you.

[Reno]

Sites that show improvement rarely hurry to post the good news anywhere.

As you & I & others have said, when a system is wholly complex then a small tweak can have massive repercussions. And that's who we're hearing from here ~ professional webmasters who know what they're talking about are seeing their sites getting hit hard, and they're/we're at a loss to explain the reasons. I'm sorry to say it's going to be like this from now on, because Google will never stop adding new spice to the soup. They serve it with a "take it or leave it" attitude, and so we come to Webmaster World, hold our collective noses, and take it.

......................

[tedster]

Moral of the story, if your worth billions, Google will cut you a break and actually let you slide right up top, but if your trying to just pay the bills,,ehh not to worried about you.

I'd say that's an unfair assessment - even though I do see it quite often. It's not even accurate in this case because Demand Media is a high profile public company and if the anaonymous report is accurate, then Google is going after them.

I personally know of many cases where Matt Cutts, John Mueller and other Google staff have done a lot to help a small scale webmaster with problems.

[dataguy]

Interesting data from SEO Clarity, Tedster. With few exceptions, the winners don't use AdSense and have few comments, the losers use AdSense and have comments.

[Whitey]

For those wondering about sites that gained, I tracked down some data from SEO Clarity [seoclarity.net] on the big winners over a 60,000 high-volume keyword dataset

Doesn't make sense to me. Here's a list of those reported to have dropped from the same source :

TheFind.com -51.6%
BizRate.com -36.8%
ShopWiki.com -68.0%
EzineArticles.com -71.7%
HubPages.com -68.7%
Buzzillions.com -62.0%
Shopping.com -64.0%
Suite101.com -77.1%
Kaboodle.com -71.0%
AssociatedContent.com -68.6%


Can anyone throw and light on why some of those with strong brand presence and supposedly good SEO would have dropped.

More analysis needed .... less bleating :)

( sorry dataguy - you beat my post with some analysis feedback )

[edited by: Whitey at 11:15 pm (utc) on Feb 28, 2011]

[kd454]

tedster,

Yeah I have seen them help people as well, this update is just a bit hard to swallow given what I have seen thus far. Guess when your number is up it is up. Live, learn and move forward.

On another note, I just received email from Adsense saying now is the time to sign up, never in the past have I gotten one of these. Houston do we have a problem (all the sudden millions/day are gone)?

[edited by: kd454 at 11:34 pm (utc) on Feb 28, 2011]

[Industry]

yes...I just received that Adsense email as well.....never saw one before...sort of "ironic"?

[SEOPTI]

Their adsense earnings crash so they need to 'fish' for new publishers.

[walkman]

Go google!

Tedster Just did a quick backlinks search: #1 on that list is using counters to get free links and #6 is there by...ironically having their articles published everywhere (with several links purposefully placed in them)

[mslina2002]

It is ironic that the poster child for Adsense "Askthebuilder.com" also got hit. 76% loss in keywords according to that list by Sistrix.

Haven't looked at the site closely or the niche but to me that does not seem to be a low content site with images, video, and text. Thought they do have a zillions ads.

[TheMadScientist]

I'd say that's an unfair assessment - even though I do see it quite often. It's not even accurate in this case because Demand Media is a high profile public company and if the anaonymous report is accurate, then Google is going after them.

I personally know of many cases where Matt Cutts, John Mueller and other Google staff have done a lot to help a small scale webmaster with problems.

Yeah, that's true, but I can see where people get that idea from: It's because of how it looks ... And when you just look at the results and the numbers like the ones you posted it LOOKS like they're in bed together.

I know I've been going pretty hard on G for not taking care of eHow, and I think it would be different if any one of us couldn't click 'view source' and see their external links are all 'fake' knowing the only reasonable reason for it is to artificially increase the PageRank of their internal pages by tricking GBot into thinking they don't have any links out on the pages ... That's a penalty any way I look at it, and hiding the attribution from non-javascript enabled visitors, bots included, just plain smells wrong.

I mean I'm with ya on the size of the job they have to do and the help they've provided webmasters (they used to do a WAY better job of it, especially here, imo) but those stinking 'fake' links are screaming 'manipulation', to me at least.

If they're not penalized, does that mean we should all start doing it? I don't think I would, but it sure is tempting if that's the 'cool' thing to do these days and not a violation of G's TOS...

How much more obvious does it get?

[falsepositive]

More thoughts on this and I've come up with yet another theory. So Google is looking for the long term view of the web. They want to get rid of crap. So what do they do? They look at those people who are ranking. That is the reason why we notice the pain, we were getting traffic for somewhere. We were doing well.

Now, they evaluate certain niches where scrapers, crap sites, auto blogs abound. They evaluate the top players. If you are somehow not unique enough, you will be demoted, allowing your competitors, much of them may be crap, up the ladder.

What do you do? You will CLEAN UP YOUR SITE. You will make sure you are unique. You will do whatever it takes to improve. Okay, so now you do that. The next time around the algo runs, the penalty is lifted, you are back up and your link profile/authority is restored. The scrapers have been identified, everyone scrambles. The people above who continue with less unique material will now in turn be demoted. They scream murder. They clean themselves up or go out of business. Rinse and repeat.

That means, if you hold top spots on some important search terms, beware. If this algo hits you, then you could get the dreaded new penalty if you trigger it.

And I don't think this is to do with style of writing or anything like that. It's to do with duplication, plagiarism, scraping. They are hitting niches where the heavy scraping tends to be -- that is where I think the "12% of searches" is coming from.

So they are doing this on purpose perhaps? To indeed get some scapegoats to start changing things. They are putting the onus on us to clean our sites. When we are indeed unique enough, they promote us back up where we should be. Those people who were once crap on top won't be there for long, if we adhere to these standards.

It just sucks we have to do so much work to regain what we once had.

[dickbaker]

Falsepositive, that's a really interesting theory. I'd question if Google would want to serve shoddy results while webmasters clean up their sites, but who knows?

[falsepositive]

dickbaker, they may want to give us time to absorb the blow. Perhaps their algo is not that 'perfect' and needs some external verifications. DMCAs will help your case. We are basically helping G vacuum the internet for them, while paving the way for some kind of cyclical penalty strategy.

In the end, we will all end up with a more unique internet. Now besides the other factors like SEO and such, we will all be battling over who has the most unique + relevant content!

And the devaluation of links here in favor of content uniqueness is interesting also. This will bury duplicators, plagiarists, mechanical turks, scrapers, even link buyers/sellers if your link buying/selling can't support a unique site.

It is sending a message that your site's age, authority and link profile don't matter as much if your content is not protected or unique enough.

If I ensure my site is "clean" and truly entirely unique, next algo push, I will be made whole, and may potentially spike higher in results when that penalty goes away. The less unique bunch above will fall and that cycle starts again. So does that result in a healthier web?

I'm trying to see the justification in using scapegoats here and allowing for the crap to filter up.

[aakk9999]

People are talking about penalty. I do not think it is a penalty. It is a new algo - this is not the same as the penalty.

For the sites that were hit - the rules on scoring have changed, and these sites have not scored enough to be ranked where they were before.

I think that with penalty it is slightly more simple - it usually points to one (or few) main reasons why the site was "hit" (or penalised or filtered out). It is easier to dig yourself out of penalty because it is easier to deduct why it happened.

With the new algo - the calculation rules have changed. For all what we know, it could have been (arbitrary) 50 factors that were all dialed up or down + perhaps (likely?) new factors introduced. This is why it is harder to figure this one out.

[tedster]

I'm with you on that. A penalty will usually be lifted after some period of time out - Matt's recent video made that quite clear. This is a re-ranking. If the algo gets tweaked you may recover some rankings, but there is not going to be an escape from the penalty box - because you're not in the penalty box to begin with.

That said, I'm sure there will be changes for this newborn component, Quality Score. It's just a baby and it hasn't learned to walk that well yet. It's older siblings, Relevance Score, Trust Score and Authority Score have over a decade under their belt.

[Whitey]

Quality Score

Do you want speculate on the most recent components of the on page factors? Is anything more obvious than others yet from what folks are saying ?

[tedster]

I'll guess it began with semantic analysis scores of several types - reading level, semantic variation, maybe even sentiment analysis.

Couple that with a machine learning algorithm that started with a hand seeded group of obvious junk for training. The machine algorithm was originally given wide range to grab almost any signal that Google has been accumulating, whether currently used or not. That search was looking for pages that clustered tightly with the seed group. Maybe two seed groups - one for great content, too.

If I'm right, then reverse engineering this baby will be very tough. General advice: don't cut corners in any way, shape or form.

[universetoday]

I'm on that top list of 300 sites. And according to that Sistrix list, my keywords went down an average of 71%. But from what I've seen in my traffic, I'm only down about 10%. Like many of you, my articles are completely original - total scraper bait, and reappear in many places on the internet without attribution. I haven't been very agressive going against scrapers, but I'm wondering if that's much of a contribution.

Anyway, all I'm saying is that the Sistrix data doesn't seem to be very accurate. I'm not seeing the punishment that data says I should be receiving - thankfully.