Google Updates and SERP Changes - August 2009

[willybfriendly]

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

I just came across another anomolous result, just as I was thinkging things were settling down.

Results 1 - 10 of about 4,490,000 for widget

Only 9 results appear on page 1. We see

"Image results for widget - Report images"

Nine listings, with the ever present wikipedia as #1. (No indented results)

"News results for widget" (One totally unrelated listing)

"Video results for widget"

"Searches related to: widget"

First time I can remember seeing Google count to 9 and call it 10.

[edited by: tedster at 10:29 pm (utc) on Aug. 1, 2009]

[Shaddows]

The title element is NOT a meta tag.

Changing the Title is the single most sensitive change you can make to your site. However, unless you changed the MEANING of the title, or added or dropped keywords, I doubt that changes would be worse now than normal.

Changing back will probably hurt more. Google seems to equate fiddling with Titles to being unworthy of a Page 1 site (occasional, difinitive changes to individual pages tend to be fine)

[Nimzovich]

The last activity was a year ago, when I set up a 3-way link exchange with a few same niche sites.

GG can see much more sofisticated things than this. Remove it and wait.

[MLHmptn]

Anybody else noticing the smaller snippets being used? 160 characters to be exact...I am assuming this is new, if not I have never noticed it.

[cangoou]

I still don't now if it's a good idea to change 50% of your IBLs. Any comments on that? It seems to be really a lot.

[James_WV]

Don't know about anyone else - but I've seen HUGE improvements in my UK results this month. I've seen a load of pages that were in the supplemental come back and rank very highly and other pages that were ranked low rank really well too.

Admittedly I've been doing a lot of on page stuff this month - mainly around changing title tags / H1 tags that I thought were a bit similar - but I don't think they were enough to give me this much of a boost

[tedster]

I still don't now if it's a good idea to change 50% of your IBLs. Any comments on that? It seems to be really a lot.

It depends on what your current profile looks like. If you've been really pushing the envelope, I can see how 50% might make sense. but in the average case, I agree that it is probably not something to aim for - in fact, in the average case, it's not even possible!

[engine]

This is the second time in as many months that i've been looking at the Google SERPs and found all sorts of poor quality sites at the top.

I have given it a few days before commenting as often these things settle fairly quickly. Not so, this time.

I've had Bing as default on one of my machines, and it's noticeable that i'm getting what I want on Bing easier than the current Google SERPs.

I can't say for sure what's causing the apparent topsy-turvey SERPs in Google, however, I have noticed a good number of long tail, on-the-money pages from various sites are no longer in the index. Again, i'll put that down to some fluctuations yet to settle.

Come on Google, you can do better!

[Optimus]

I have been watching with interest the allins and their correlation to page 1 results for a primary travel term eg blue widgets. From the getgo back at the start of this update, sites 1-5 remained constant in the results and seemed immune to the algo changes (maybe google wanted to maintain some decent page 1 results - remembering that the golden triangle is what most viewers focus on).

Shifts were taking place from positions 6 - 30. My site - like others that were shifting - saw the serp result sometimes correlate to allintext, sometimes to allintitle or allinanchor. Last weekend, the allins slowly started aligning, with the top 10 sites now occupying identical positions on the serps as they do on the three allins. I have not seen this type of merge of allins before, nor such a close correlation to serps.

Anyone else seeing similar trends for primary search terms in their sectors?

[willybfriendly]

I'm going to guess that things have pretty much settled in again. The anomolous .txt page that shows up in my niche during flux times is gone and I have sites back at #'s 5, 8 and 13 (and holding steady).

There still is a 2 year old "news" story showing top 10. When that is gone I will be sure that the churn is over for now.

[eljacko]

I have just been a victim of the yo yo effect. I have a term at position 6 and all of a suden it's down to 21, can it be an update of is it really the yo yo effect?

[Shaddows]

It might be a Yo-Yo if it goes up and down frequently. If its just moved once and stuck, its probably a re-rank.

[whitenight]

I've had Bing as default on one of my machines, and it's noticeable that i'm getting what I want on Bing easier than the current Google SERPs.

Come on Google, you can do better!

lol, just wanted to highlight this from engine for posterity sake (...and future debates)

classic! =)

[tedster]

I'll bet they will do better, too - after Labor Day when the new experiments are tuned down a bit, as seems to be the tradition.

Here's what I currently think is in play (at least today) --

1) A newly tweaked taxonomy for query types
2) An attempt to find the right recipe for blending whatever diversity seems indicated into just one SERP.

For some queries, this "diversity mashup" is not a big deal. For others - those that intersect several taxonomy slots - it can be way crazy.

[dynamicwebs]

< moved from another location >

Just wondering if anyone else has noticed the algorithms change about every three months? I'm guessing the next change will occur in late September 2009.

[edited by: tedster at 3:46 am (utc) on Aug. 8, 2009]

[Marcia]

>late September 2009

>I'll bet they will do better, too - after Labor Day when the new experiments are tuned down a bit, as seems to be the tradition.

Don't you think it'll stabilize well before heavy seasonal holiday traffic starts in October. Better to have changes now than then. I sure hope so, that dreadful Florida update was in November.

[tedster]

Florida was such a shocker -- and it was also before the Big Daddy infrastructure, and that introduced us to the continual change that Google now creates. Once a month updates were all we got. But the algo was certainly easier to study back then, wasn't it?

The history of Google Updates [webmasterworld.com] makes an interesting study every so often. One thing I can barely find, though, is an update where our discussions were full of comments like "Wow, they really got is right this time." ;)

[whitenight]

Speaking of... do I get the honorary position of suggesting this update's name? ;)

The "Long National Nightmare [en.wikipedia.org]" Update?
The "Jumped the Shark [en.wikipedia.org]" Update?

One thing I can barely find, though, is an update where our discussions were full of comments like "Wow, they really got is right this time."

That's probably because the algo SERPs 3 years ago were just as or MORE "helpful to it's users" than anything they've done since.

I get the VERY STRONG feeling this algo update was/is more about money and profits than actually improving the accuracy of searches.

If it is INDEED about improving the SERPs for users, than as many people have said in this ever-running thread,
"too many cooks in the kitchen"

How many PHDs and patents and "new ideas" does it take to improve the basic design of the wheel or apple pie?

At a certain point, it's just shuffling around the same pieces over and over again until one forgets how the original pieces fit together.

[mrguy]

At a certain point, it's just shuffling around the same pieces over and over again until one forgets how the original pieces fit together.

I think they need to post that quote on the big white board in the plex. Talk about hitting the nail on the head.....

[Shaddows]

Whenever you criticize the SERPs, its worth remembering that, as a group, we are WAY more savvy than your average searcher.

3 years ago, I could find anything I wanted. Usually not on the first search. However, as I modified my search terms, I got there. That is no longer the case.

But the thing is, most searchers are browsing. They're looking for general information. The result is a dumbing down of content. As frequently comes up here, Google is less inclined to return what you searched for, but rather what they think you meant to search for. This causes fuzzy SERPs filled with irrelevant results. Very, very annoying.

There is also a confirmation-bias problem. On "our" terms, we think people making that search are looking for sites like ours. Actual analysis of behaviour might show that more people are interested in another TYPE of site. Google includes more of that type of site, SERPs looks completely different, and (as far as we are concerned), stuffed full of irrelevant results.

As for the lack of "they got this right", the fact is we all think good outcomes are our hard work, and bad outcomes are due to poor Google algorithms

[mrguy]

Google is less inclined to return what you searched for, but rather what they think you meant to search for.

There in lies the problem. Google is trying to think for the visitor.

I've had numerous clients, family members who don't know what a serp is or what seo means, or wouldn't know a webmaster if one fell in their head ask me, "what's up with Google lately, I can't find thinks like I used to."

The average user is noticing that the serps of Google are not what they used to be.

[helpnow]

How are the past 24 hours for everyone? I noticed a shuffle in rankings on August 3, then reverted to "normal" (normal being what I have become accustomed to for the past YEAR) 24 hours later.

Things stayed "normal" for the past week.

Now, again, I see today another shuffle in rankings, identical shifted rankings to what I saw last week on Monday.

Anyone else noticing a shuffle these past 24 hours?

[tedster]

Yes, I've seen some juggling in the past 24 hours. One interesting feature is that the "related searches" suggestions seem to have changed at the same time.

[helpnow]

Grrr... ; ) The question always is - is it me, or is it google? Is it something I did wrong, or are they just shuffling things around? So, it sounds like it is common for everyone, and that I didn't do anything wrong. ; ) That's the good news - the bad news is, my traffic is down and I cannot control my destiny. ; )

[eiilers]

Breaking news - Looks like google is working on a new search engine (or a major update to its current one). See the article at [bizjournals.com...]

And

[googlewebmastercentral.blogspot.com...]

Also, TEST the new search engine live at [www2.sandbox.google.com...] . I am curious to see feedback from everyones SERP's on this new engine at [www2.sandbox.google.com...] . I've lost TONS of rankings on the current google.com, but on the new test server my rankings are back where they should be. Could this be why the current SERP's are so screwy? I hope Google rolls this out ASAP.

[helpnow]

after 24 hours of some rankings lost on the old serps, my rankings are 100% back to normal. This is twice in the past 10 days that it has done this - always on Monday ; ). Mondays suck as it is, don't do this to me, google! ; )

As for the new SERPs, I have about 95% of my SERPs - I only see one SERP where I've dissappeared, but I think I can attribute that to a goo hiccup. The rest of the SERPs are about the same for me on the www2.sandbox... The only thing I noticed is that it is no frills - no images inserted in the SERPs, no Google Base, no ads. I suspect this may be the reason the new SERPs feel faster right now - it is all just basic SERPs...

(My site is an authority site - many different categories, hundreds of thousands of listings, most SERPs are double listings for our site, been around for 11 years now.)

[helpnow]

eeilers - with respect to having lost rankings on the old SERPs, I wonder if you are now back thios morning, like I am?

[eiilers]

No I am not. They are close, but my current rankings are significantly different than the test server. The test server rankings are much better and more relevant results.

[Hissingsid]

Also, TEST the new search engine live at [www2.sandbox.google.com...] . I am curious to see feedback from everyones SERP's on this new engine

Those results look like the last time Google was really bad made worse.

Cheers

Sid

[eiilers]

The current Google SERP's for me are probably the worst they can get. The new engine is much better and reminds me of the Google 2 months ago before all of these crappy updates.

[Shaddows]

FYI- There is a very active thread [webmasterworld.com] dedicated to the dev engine (http://www2.sandbox.google.com)