2: Google MAYDAY Update - SERP Changes May 2010

[cangoou]

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

Puh, this is really strange: I still see 2 resultsets - and after 7 days I can say that SERPs #1 are shown from right at 0am to about 5am local time and SERPs #2 the rest of the day. I thought this was some kind of dancing around again but it seems to be quite stable now for 1 week.

Funny thing is: SERPs #1 are better then before, SERPs #2 are worse (in rankings of main keywords).

Google is no longer following my robots.txt file. It has indexed hundreds of pages that I have disallowed.

Same here, got a complete set of new pages with robots=noindex in the index.

Generally it's best to make any changes when SERPs are stable, never within an update due to panic!

In general I would agree, but I haven't seen something like "stable SERPs" since January... This makes it quite hard not to panic ;-)

[edited by: tedster at 7:43 pm (utc) on May 13, 2010]

[cwnet]

dog training: from Berlin, Germany with Opera 10 on Windows 7 -

Google Language English

1.) dog-obedience-training-review
2.) pupdogtraining
3.) dogproblems

5.) wikipedia

Google Language German:

1.) dog-obedience-training-review
2.) dogproblems
3.) wikipedia

5.) pupdogtraining

[anand84]

Dog Training : Chennai, India

1. Pupdogtraining
2. wikipedia/dog training
3. drilldog
4. dogtrainingways
5. schoolingdogs

34. dog obedience training review

101-dog-training-tips not in the top 100!

[TheMadScientist]

Fascinating... Thanks to everyone for sharing and huge thanks to 'TheTedster' for making it possible to do here.

I am starting to wonder how long it is before this thread starts to out rank some of the sites though, which might be interesting to see itself.

[Robert Charlton]

Location is the San Francisco Bay Area.

I tried the [dog training] search on two different browsers...

- one with long-accumulated Google cookies but not signed in
- the other with cookies flushed, also not signed in.

On this search, I get same results in both browsers, and, for a half-hour window, anyway, the top three were extremely consistent, as were the others I'm noting...

1) dog-obedience-training-review
2) pupdogtraining
3) dogproblems

7) wikipedia

26) 101-dog-training-tips

[seolearner]

Results from India,

After logging off, personalized disabled, cookie,cache deleted. Results checked using G*.com version.

Browser: Chrome,
OS: Vista
Keyword: dog training
Total results: 48,200,000
--------
1.dog-obedience-training-review
2.pupdogtraining
3.dogproblems
4.cesarsway
5.en.wikipedia.org/wiki/Dog_training
6.dog-behavior-training.co.uk
12. barkbusters.ca
19. 101-dog-training-tips.com
24. trainpetdog
--------------

Results from my regular browser-Firefox where I never delete any cookie or cache, Without logging in. Since it's my regular browser, results came from G*.co.in

Total Results: 47,700,000 results
--------------------
1.pupdogtraining
2.en.wikipedia.org/wiki/Dog_training
3.drilldog
7.dogproblems

All other sites not in first thirty results.
----------------------

Results using my regular web proxy which is located in Chicago-USA.
Total Results: 52,200,000 results

-------------------------------
1.dog-obedience-training-review
2.dogproblems
3.101-dog-training-tips
-Local Maps
4.pupdogtraining
27.trainpetdog
29.barkbusters
-----------------------------------

Results using chrome, after appending '&gl=US'
Total Results: 24,600,000

------------------------------
1.dog-obedience-training-review
2.pupdogtraining
3.dogproblems
24.101-dog-training-tips
All others not in 100 results.
----------------------------

This experiment is very useful but there are so many ways to get results. There are so many factors which affect the SERPs you get. There so many niches and so many keywords. In each keyword, G* will arrange it's SERPs differently. Finding out WHY G* shows different results to each user, only using the one keyword, will not be possible. The only conclusion we can draw is that G* shows different SERPs to different users.

Edited: To add 101-dog-training-tips in first set of results.

[gethan]

Malaysia (IE8, Win7, .com, inPrivate)

1) dog-obedience-training-review
2) pupdogtraining
3) dogproblems

5) wikipedia

22) barkbusters
28) 101-dog-training-tips
42) seefido

[tedster]

< I moved a side discussion into its own thread:
How can we do SEO if Google rankings are never the same? [webmasterworld.com] >

Here's my results for the [dog training] search - from the Boston area on Opera:

1. dog-obedience-training-review
2. pupdogtraining
3. dogproblems
---
6. wikipedia

[seolearner]

If we are not able to at least get a rough idea of how we are "really" ranking across the board then how can we know how much work needs to be done to move up the SERPs.

As many members said before on this, We can not really know how our site is ranked in other people's computer. The only way I can think of knowing exact position is to trust on WMT's average position. When new WMT interface introduced for search queries, there was no average position of the keyword. Now recently WMT has introduced the average position column for the search keyword. I think variation in the ranking is the main reason WMT had to introduce average ranking position. So webmaster knows the average ranking.

I would think that if you built a site with great on-site and off-site SEO (along with Google trust)then your rankings should not be all over the place compared to other sites. From this simple testing it seems that at least the #1 site is the only consistent site for pretty much everyone.

Yes, this is true and it seems that #1 site is consistent in performance but I am sure there are many more keywords where even #1 is not consistent. As I said, finding HOW & WHY using only one keyword's ranking will be difficult. It might be possible that consistency depends more on off-page SEO rather than on-page SEO.

[john1975]

The more results I see, the more I understand that stale results are a thing of the past. I can see where google are going. This is the future of seo.

[Shaddows]

Results from UK (South East) using IE8

Google.co.uk:

1. dog-behavior-training.co.uk
2. k9obedience.co.uk/
3. wikipedia.org
- LOCAL MAP
4. dog-obedience-training-review.com

Google.com

1. dog-obedience-training-review.com
2. pupdogtraining.com
3. dogproblems.com

5. wikipedia.org
6. dog-behavior-training.co.uk

Notes:
The US #1 gets inserted at 4 in UK (5 if you include map)
The UK #1 gets inserted at 6 in .com results
.com mantains the same #1 in all searches performed so far, from any part of the world, but not in google.ccTLD

-Edit: I notice TLDs have been shunned so far in this thread, but that's not going to be helpful in comparing non-US results

[edited by: Shaddows at 8:07 am (utc) on May 19, 2010]

[Plan_D]

Can someone please help me out, I'm getting confused (or being a bit dim).

Is the problem we have observed:

1. G* has lost some data and is reindexing
2. G* has changed her algo
3. G* is now delivering different results to different people/browsers/areas
4. Something else.

(80% drop in G* traffic as of monday - aaaahhhhhhh!)

[Shaddows]

I would say
1. Google MAY have lost data, but is certainly reindexing
2. There is an algo update. It is post-caffeine, which is now live.
3. Geo- and other- targeting is now prevalent. It is core, rather than peripheral. It is no longer results-modification, but results-building
4. Possibly other things too.

[Andylew]

I just had an interesting call from a client, non comuter literate who thought that there was something wrong with his computer because he kept getting adult and spam sites up on google. That is proof enough for me if people who are completely oblivious to google updates are now calling me out of the blue as they have noticed something isnt right that google must be broken.

[anand84]

Geo- and other- targeting is now prevalent. It is core, rather than peripheral. It is no longer results-modification, but results-building

How should one look at this? Definitely, it is helpful if your website caters to people from one particular location since you no longer get non-targeted visitors. But is Google intelligent enough to understand that some websites cater to international audiences and hence need to be shown the same way everywhere? "Dog training" for one is an international theme. And ideally there should only be one particular set of results.

Also, I'm intrigued by the 'CTR' value on search queries in WMT. If Google thinks my link gets the maximum CTR when it is on the third page, it probably pushes my site there even if it is the most targeted result.

[dertyfern]

I'm seeing the exact same results as Shaddows here in London using FireFox 3.5

Google.co.uk:

1. dog-behavior-training.co.uk
2. k9obedience.co.uk
3. wikipedia.org
- LOCAL MAP
4. dog-obedience-training-review.com

[dertyfern]

Incidentally, left margin is showing me these results (which change by query but I suspect may change for each user):

o The web
o Pages from the UK

o Any time
o Latest
o Past 2 days

o Standard view
o Related searches
o Wonder wheel
o Page previews

[BeeDeeDubbleU]

This reminds me of that old Sinatra song.

Altogether now...

"That's life, that's what all the people say.
You're riding high in April,
Shot down in May
But I know I'm gonna change that tune,
When I'm back on top, back on top in June."

[ohno]

Haha, like it, you gota keep smiling :) (hard as it maybe). We are seeing more hits for niche products, products with brand names have all but disappeared........

[MacSeth]

I start to think that Google may be telling us something: "we are killing SEO seriously and the only thing we want YOU webmasters to do is create quality websites and stop fooling around with your technical "shizzle""

I keep hoping I am wrong... but I am not sitting still here either. I just keep improving linking (not for serps alone but for quality traffic and focus more on social behaviour).

[anand84]

I start to think that Google may be telling us something: "we are killing SEO seriously and the only thing we want YOU webmasters to do is create quality websites and stop fooling around with your technical "shizzle""

I believe a majority of sites that have lost traffic are those that were actually creating quality sites. A few spammy sites have risen to the top of SERPS in this meantime and they are kind of winning at the moment

_{Edit : Reworded the post to mean what I intended to}

[edited by: anand84 at 10:19 am (utc) on May 19, 2010]

[ohno]

/\ yep!

[john1975]

@anand84 and @ohno.. we create high quality sites and gained traffic on more than 1 site so it kinda defies what you are saying

[chrism]

@anand84 and @ohno.. we create high quality sites and gained traffic on more than 1 site so it kinda defies what you are saying

I think anand84 meant sites that have lost traffic tend to be quality ones, not that all quality sites have lost traffic.

[MacSeth]

I am not saying they were not quality sites or something... I am saying Google fights the technical things SEO's are doing. There are a lot of quality sites without good technical SEO-"tricks".

Also: there seem to be (a lot of) websites which had a penalty in the past and got back into the serps, thus increasing the amount of competition. I do not know if this is temporary or whatever but I strongly start to think this is NOT temporary.

I also noticed a weird error on the results maybe indicating things are still not well enough at Google:

[edited by: tedster at 3:34 pm (utc) on May 19, 2010]

[Plan_D]

What I am finding is that a large number of pages that used to get traffic because they were near the top of page 1 of SERPS have dropped out of the index.

This has meant that the pages that remain have a lower value because of all the link power that has been lost in the dropped pages.

This then leaves space for all the new pages that are appearing in the SERPS (especially if the same has happened to my normal competitors)

So surely this should sort itself out when the pages get re-indexed.

Or will they not get re-indexed...!

[ohno]

Webmaster tools shows indexed URL's going up slowly for us..........

[walkman]

Very strange!
Clicked on site:name.com
Using left side G options, looked at new pages added to the index on the past 24hrs, week, month and I see dozens upon dozens of new pages. But they have been on my site for many years and were indexed in Google before

A re-index can explain the flux and drops

[edited by: walkman at 10:39 am (utc) on May 19, 2010]

[Andylew]

Something interesting ive just found, id be interested to hear whether others can identify the same,

WMT search queries, produce the graph for mid april to now.

Then filter by different contries, for all of them we have a decline to virtually 0 between late april and eary may. However the interesting thing is several contries have now recovered beyond what they were in may, the recovered countries include:

Germany - traffic up consistently since 10th may
Sweden - traffic double april since 13th may, went up overnight
Canada - traffic back to april levels on 10th may overnight
switzerland - traffic back to april levels overnight on 10th may
austria - 10th may (im begining to see a pattern!)
australia - full recovery - 10th may
poland - full recovery - a ha 10th may again!

Partial recovery:
US - traffic to 50% of april since 10th may

The non recovered countries are:
uk (main market) down to around 30% of april levels.
netherlands
ireland - pancake since may 1st
spain - pancake since may 1st
france
denmark
belgium

The ones that havent recovered are consistent in daily impressions (although down) like the recovered ones were between 3rd may and 10th may before recovery.

[Plan_D]

@Andylew have you found that pages have been dropped from the index and then come back?

[Plan_D]

@walkman just tried the left side time options and find that on the normal search now have 10,800 pages indexed but if I click on 'past 24 hours' I get zero pages, but for 'past year' it is 38,000.

Odd to have more pages in the last year then overall.