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]

[trakkerguy]

Perhaps "Mayday" is not the best name?

Payday?

[backdraft7]

Google is actually taking advantage of the bad SERPs by sending-out AdWords invites right now?

A premeditated plot to upset everyone's business in order to force their services on us is beyond evil and should promote the government to initiate anti-trust action against such a company. If this turns out to be true, then I've lost (even more) faith in mankind.

[outland88]

I definitely second the motion of renaming it to “Payday”. It sends a stronger message pro or con about what people think of Google. In fact from this point on out I will call it just that.

As for things getting better many of those Google fellows vacation for much of July and August so I was predicting something would get settled. After all it's payday. Then you can wait for September when they permanently put you out of business with the holiday changes. I’d start ramping up for Bing and Yahoo plus other traffic sources. Google’s volatility has become to much except for the strongest brands, 100,000 plus page web sites, niche areas, those engaged in full time link buying and building, and heavy Adword spenders it’ll be hard for many small businesses to survive.

[ohno]

I looked at adverts on Bing & Yahoo, both look reasonable? Although not sure if any use in the UK. One things for sure, i will not be doing Adwords again. I must be getting paranoid too as right now i can't help but feel analytics has given them the data they wanted and they will use that to shaft us.

[pontifex]

Something happened in the last few days... I am up in traffic a bit and revenue even more from the drop on the 28th. Summary: I got 3 mails from fellow WebmasterWorld members directly, telling me they recovered to around 90% of pre-slaughter and so am I. So, there was a little recovery for SOME. Not sure if you folks can second that?

[globglob77]

@ohno

I think so too in regards to analytics. The question is: How many of the websites which have lost traffic have an analytic account ?

[tedster]

Welcome to the forums, globglob77

You know, even in recent days, Matt Cutts has flat out denied that kind of connection - even saying that he recently double checked just to be 100% certain. See this video [youtube.com]. Now I've known him to use some cagey wording at times, and we could say misdirect our attention, but I've never known him to flat-out lie.

I also don't think it would make any sense to have a ranking factor that was only available for some of the index. So I definitely don't by the idea that Google Analytics data is part of the organic ranking algorithm.

[aspdesigner]

Perhaps "Mayday" is not the best name?

Payday?

@trakkerguy: I was actually referring to SON of FLORIDA, see my latest post here why -

Is History Repeating Itself at Google? [webmasterworld.com]

or perhaps the more high-tech name...

FLORIDA V2.0

But I like your idea too!

The Google PayDay Update

[drall]

Analytics has nothing to do with it in our case. The only site we have that really took a hit from all of this has never ran GA and only runs the latest version of webtrends which I find to be far more robust for this particular site.

So in our case at least GA data being used against us doesnt hold water.

But one sidenote, we do have adsense on this site which probably could be used by Google for the same purpose if they so desired.

I am seeing a slight recovery but still down some 25% pre mayday numbers and this is factoring seasonality in. Also traffic patterns are completely out of whack.

We are ranking #1 for 2 single terms that we have no business ranking for, not even loosely related to our site in any regard.

[webby282]

Well Sure something happened in the last days, Cutts said Caffeine has been completely released so maybe what we've seen, is the final process of releasing this G* update.
I had a big competitive keyword that was ranking 1 for a couple of years and all of a sudden at the mid of may something happened and JUST this keyword has dropped to 6. The maintance work I've done on this one can probably lead to the conclusion that there is some kind of OOP (Over Optimiz.)penalty going on, but it's weird that happened just in that days, now one month ago more or less and JUST for that specific keyword.
Any idea apart from the OOP? Could be something else? How could i get back where i was?
I have analytics by the way...

[mantucket]

Could some of you please perform the experiment I requested and tell me the results?

Again, google for something that is usually a direct match, including your domain name (but not the site: command) then click Cached - I was seeing about 80% error rate on Cached page links, trying to figure out why and if its' related.

[aspdesigner]

"I looked at adverts on Bing & Yahoo, both look reasonable? Although not sure if any use in the UK. One things for sure, i will not be doing Adwords again."

@ohno: One of my clients that is angry about his rankings, is also a big AdWords customer. Perhaps I should explain to him what Google is up to, and advice him to re-direct all his PPC budget to Yahoo or Bing instead?

[Robert Charlton]

scottsonline - A follow-up question on this comment....

Our competitors are listing 100 products as "related" per page to jam the brand name in there and Google seems clueless that this is not user friendly and is really just a new way to keyword stuff.

Just to clarify... I'd been thinking that MayDay is reacting positively to co-occurrences, but negatively to occurrences, so I've reread your post over a bunch of times.

Is the brand name stuffing boosting just the co-occurring product searches... or is it boosting searches that include the product name?

[Robert Charlton]

And PS to the above... that doesn't sort out the repetition of "custom widget" 30 times on the homepage, which you report as ranking for "custom widget". Anything otherwise exceptional about this page?

I'm trying to play devil's advocate with other theories here... not questioning what you've seen.

[aspdesigner]

@Robert: What scottsonline reported is the same sort of thing we are seeing, and which was also being seen back in 2003 in Florida.

It appears that a full keyword match in a relevant Title (i.e. - white hat SEO) is now penalized, while those same keywords repeated numerous times in the body (i.e. - spamming) is now rewarded!

This is the exact same thing that was also being reported during Florida, with several members here angrily complaining that Google was "slapping-down" those who had followed Google's recommendations for SEO, while rewarding those that deliberately broke the rules!

BTW, the title penalty is also keyword-specific - more like Google is simply evading giving you those results that are the best match for your search.

Imagine the "boost" you would normally get from a complete match in the title tag, being flipped to a negative, and the "penalty" you would normally get for massive keyword-stuffing, being flipped to a positive.

It's like Darth Vader got his hands on the control knobs at Google!

[mantucket]

aspdesigner, I'd love to see some specifics on this somehow or verification from several sources.

I tend to doubt that an exact-match title would be discounted, because competing sites with the same titles are still ranking while mine is not, for those keywords. In my case it would be fairly easy to switch
to regular repeated text in item descriptions, so I'd really like to know if there is something to this theory.

I did notice something similar that seemed a bit odd though, which was that when you click 'Cached' to see what it matched, the keyword matches are highlighted in page text - but NO matches are highlighted in the URL itself. I thought URL matches counted for a lot, but can't find a good example of how it used to work for comparison.

Perhaps the new mayday rules only apply to pages that get classified as 'inventory' pages?

[mantucket]

I just did an experiment which disproves the title-exact match theory. I still have a page that ranks #1 for:

"blue widgets electrical"
The first 3 words of title are exact match.

A competitor ranks #2 for the same phrase, but his order is "Blue electrical widgets"

If I switch the keywords in my search to "Blue electrical widgets", he now ranks as #1 and I have #2.

So this is a very isolated case where google is clearly paying attention to title exact match for both #1 and #2 slots.

[mantucket]

Here's another video from MattCutts from today talking about MayDay [webpronews.com...] he gets a bit more specific, saying it impacts auto-generated content, content farms, etc. I'm more confused than ever since the sites I see still ranking are all auto-generated or templates...

Mod's note: Fixed link... removing shortened link to frameset that was framing the page and made direct link to webpronews.com.
With current demand, video may still take a while to download before it plays.

[edited by: Robert_Charlton at 7:26 am (utc) on Jun 10, 2010]

[cien]

I never believed a word from MC's mouth. He is a BSer. He just uses scare tactics to keep the sheep at bay. He is been proven wrong many times.

[tedster]

I found those comments to be the most helpful "official" remarks so far. The newly mentioned concept for understanding the Mayday update is "content farm" -- Matt repeated the phrase several times and that's apparently one of the key targets. I get it. I've been at meetings with other web marketers in the past months where the topic came up... a lot. We wondered when Google would put content farms in their sights.

But there's still some trouble in paradise. I'd agree with others here that this new algorithm element is still a way off from being perfected. However, since May 3 we have been seeing ongoing tweaks. Some rankings that were incorrectly nailed, false positives in other words, are now seeing their traffic rebound. In addition, I still see some content farm style sites getting way too much traffic.

Matt mentioned some other targets of Mayday - such as sites with too many stub pages. So there is some new detail and food for thought in this interview.

[Robert Charlton]

aspdesigner - Thanks for the clarification. I haven't run across any examples of pages with numerous repetitions ranking, which is why I'm asking about them.

Though in our discussion in the Repeating History thread I noted that I was seeing pages lacking search terms in body copy, I should add that I myself wasn't seeing any predominance of pages that were missing search terms in titles.

Those pages that I am seeing rank without the full three words of a three-word search in the title in competitive searches are generally pages that would also rank for the core two words of the three word phrase. As an example of a three-word search that doesn't exclude title terms, try a search for the title you suggested, which... contrary to general forum rules about specifics... I'll let stand as an example.

I'm sure we're talking about different kinds of pages here, and I think that mantucket's phrase "inventory pages" may be hitting the nail on the head...

Perhaps the new mayday rules only apply to pages that get classified as 'inventory' pages?

I've called this "thin content", and, along with many others, I've been warning against this kind of page for quite some time now. Google's been warning against it too. This shouldn't be coming as a surprise.

I agree that Google has turned the "knobs" way too far. In many searches, where the results are bad, they're very, very bad. I don't think it's intentional, and I don't think it helps Google to be returning pages as bad as they're returning.

As to why the apparent keyword stuffing might be working right now... algos tend to grasp blindly when there's very little that matches the criteria they're searching for, so MayDay could be cutting so deeply when looking for co-occurring content that it's not leaving much of anything, and then deciding that those pages with multiple repetitions... spam... is better than nothing. Just conjecture that this might be what you're seeing.

PS: I haven't been able to get the video to play... too many folks here logging on at once. ;)

[ohno]

Hmm, our .co uk site saw a spike in traffic yesterday which was the highest since May 17th (Mondays are usually our busiest day). Our .com site saw a spike on Monday which was the highest THIS YEAR if not ever! (I've gone back a fair bit). The .co.uk site seems to be showing an upward trend while .com site is downward. Sales were still slow so the quality of traffic must be poorer.(time on site has increased massivly too?)

RE:analytics, I wasn't meaning sites with analytics taking the hit more the (majority of ?) sites out there WITH analytics has given G a lot of data on user behaviour? Like telling your "best friend" everything only to find out he was out to shaft you from day 1?

[ohno]

Oh, & the page speed tool is rediculous, we've gone from being faster than 92% of sites to being slower than 68% in a week!

[Top_Hat]

Just to add my experience to the discussion.

We have a .co.uk shop that has been running for over 10 years, we've never spent much time worrying about seo or google, we've never had a sudden dramatic change in traffic, we just build our pages and sell our products.

Looking at our traffic we had no shift in traffic mayday or throughout may, but on June 1st we lost a huge chunk of traffic coming from google (40% - 50%)

Is that part of mayday or something else?

We also sell our products via amazon uk, we've see a large jump in the number of sales coming from this source (not enough to fill the hole thought)

[Robert Charlton]

Top_Hat - Welcome to WebmasterWorld. Sorry about the loss of traffic.

We also sell our products via amazon uk, we've see a large jump in the number of sales coming from this source...

Are any other sites using your product descriptions (including amazon uk)?

Does amazon uk have product reviews and other text on its pages that you don't have on yours? Is your text rich and descriptive, or sparse?

[ohno]

Isn't this the point though? G wants rich content, say you sell screws, there is only so much you can say about a screw! We sell boring products, our customers like(d!) our site because they could find what they wanted easily, I'm sat here now thinking I will have to empoly a writer to make padded out descriptions(maybe an estate agent would be a good start?!).

[Robert Charlton]

...there is only so much you can say about a screw...

Agreed. That is a real dilemma, particularly if you have resellers who have more link juice than you do.

I've been adamant with several clients that they did need to add something extra... and they not only had to contractually specify to resellers that their web copy had to be different, but needed to create a different version for resellers. Fortunately, the products had more to be said about them than screws do.

My guess is that if your category pages are differentiated enough, your parts number pages are still likely to do well, but I've never been confronted with such a lean topic.

[sean22]

You can have an imaginary 250-300 word review for each screw. Strip as much code as possible from the page to elevate your content to code ratio, and if G doesn't like it, try something else.

[aspdesigner]

"aspdesigner, I'd love to see some specifics on this somehow or verification from several sources."

Well mantucket, unfortunately we're not allowed to post specific searches here...

(QUICK NOTE TO ROBERT: Responding to your post shortly, but wanted to clarify, that was NOT an example search in my prior post! :()

...but I will provide what I can. Let's see...

First, there is the test I described in detail in my first post in this thread -

Is History Repeating Itself at Google? [webmasterworld.com]

We were kicked-down to the fourth page on an exact match. Examining the sites above us revealed that out of the Top 30 sites, only 2 of them even had all 3 keywords anywhere in the title!

And by De-Optimizing the match (via changing words in the search) so that we were no longer an exact match, jumped our page up to #2 out of the Top-10!

I then started looking at various SERPs for other clients (whose prior results I was already familiar with), and then doing "test" searches, trying to see exactly what the "payday" update was doing.

I discovered that competitive SERPs, that had previously been dominated by full match titles, now had the majority with partial match, 1-word match, and sometimes no match in the title at all! These were all competitive Top-10 listings.

In searches where Google decided to use a not-quite-synonym, the Top-30 listings (that's as far back as I checked for each) were dominated by sites with the synonym in the title, rather than the actual search term I entered.

Some SERPs had more full matches than others. But often the keywords were broken-apart or mixed-up. Or they were an exact match, but often were spammy sites, with URLs like -

www.keyword-keyword-keyword.com/keyword-keyword-keyword-keyword.html

There are still some decent sites with an exact match still showing-up at the top (we have a few). But they are now in the distinct minority.

And I'm not the only one reporting this. See also, for example, scottsonline's posts on Page 5 of this thread. Others here have also reported seeing this.

This same thing was also reported by numerous people in Florida.

" I just did an experiment which disproves the title-exact match theory..."

mantucket, if you take a look at my previous post, you will note that I was careful to indicate that this applies to a FULL title match (vs. partial match), NOT an exact match.

You tested something else.

But if you want to try an experiment - see if you can find a search which is a full match for which you used to come-up Top-10, where some of the results now include a synonym for one of the keywords you entered. Now, try doing variations of that search, but replacing that keyword with the synonym Google used instead, and see if you can get your site to come-up!

That is how we managed to get our client's site to go from fourth page, to #2!

You can also try adding an extra word to the search, which is present in the body text, but not the title.

Once you know what works, you will have a clue as to how you can modify your site! ;)

[Top_Hat]

Top_Hat - Welcome to WebmasterWorld. Sorry about the loss of traffic.

We also sell our products via amazon uk, we've see a large jump in the number of sales coming from this source...

Are any other sites using your product descriptions (including amazon uk)?

Does amazon uk have product reviews and other text on its pages that you don't have on yours? Is your text rich and descriptive, or sparse?

We do have good descriptions on our product pages, we also use those descriptions to feed our amazon listings, we have been doing that for over a year.

Lots of our pages have reviews, but not all equally amazon have a few reviews (but not many)

Do you think it might be a duplicate content issue?