Panda is a brilliant move. Here is one BIG reason why

[My_Media]

Hi All,
My site consider midsize to large with pre-panda traffic of 60k unique/day. Am sure that most of us feel the frustration. My site now hangs on life support of 18k/day and declining weekly.
But here is one reason why Google make a brilliant move that I give them a A++.

1. After months of searching and fixing my site. Boom! All of a sudden I figure out that is what Google want all webmaster do and be serious about your site and make it as perfect as possible so that the internet will have more and more great content.

Smart move Google. If Bing and Yahoo still sit around and not force webmaster to put some work on this site then they will lose the search war.

[crobb305]

But how about telling Matt Cutts why he isn't ranking for his own blog post [google.com...] ? He is #5 on my browser and we need an answer other than Google is broken, that's impossible. I took a screenshot just in case. #1 and #5 is a lot different in clicks as we all know.

Wow, and it's a recent article (early April). Before, we were talking about an article from February, so clearly scrapers continue to rank better and better for MC's content.

[tedster]

Well, he still ranks #1 for a more likely search, such as [30 day challenge going offline]. I'm not sure what it means when you get outranked on long exact quotes but not realistic searches. It does look bad for Google, but it doesn't really affect their user base.

Still, I'll bet there's something their team can learn by studying that situation. At least we know Matt's blog gets ranked by the same algorithm.

[walkman]

@crobb305
Yep, Matt Cutts a real 'brand' if you ask me: PR7, 270,000+ links (from over 5000 sites!) according to Yahoo, very busy on Twitter, Google mentions, linked from a gazillion blogs and newspapers, dozens of comments in his blog posts yet he can't rank for his articles. So how are we going to do better than him if Google is working fine?

In another post I mentioned that Aaron Wall (Seobook) had his guest's article scrapped and he was #4 or 5. He had over 170 tweets to it so I imagine dozens of people linked to his blog post yet he can't rank.

Edit: true Tedster, but not everyone searches for the title and we don't how this effects his traffic. Looks like Google long tail search has gone caput, and yes there's a lot of money in long tail.
I added even quotes and he still is #3 for me
[google.com...]

I always assumed that the longer the sentence the better the likelyhood of getting it right. I search for shorter sentences and 3-4 words, and he ranks. But I don't know if personalization has kicked in or not by now.

[brinked]

walkman, actually the page that should be ranking #1 for that search is on page 2 (the actual blog post).

What is more alarming about that example is the sites that rank 1-4 are complete crap and useless.

I do agree with the direction google wants to go in, but panda was just released and its obvious it has many bugs. Hopefully they will correct them sooner than later.

[tedster]

Looks like Google long tail search has gone caput, and yes there's a lot of money in long tail.

That's a potentially interesting focus for understanding what's going on with the algo.

The so-called content farms are all about exploiting the long-ish tail, filling in gaps where there are no exact matches and things like that. But most of the Panda analysis we've seen can only focus on the short, big money searches.

[mslina2002]

But how about telling Matt Cutts why he isn't ranking for his own blog post

MATT CUTTS:

Why does Matt Cutts not rank?

Perhaps he does not care to rank. So what if he does not rank. That blog does not make him money or cost him his J.O.B.

My thoughts why he doesn't rank because:

- the blog has canonical issues. The www and non-www both resolve and no redirection taking place. So he has dupe content on his blog.
- I didn't look in detail but he may not have put much effort into SEO, site structure, theming, etc.

I too noticed that SeoBook doesn't rank #1 for their own content. They even have 150 Tweets to this story. [google.com...]

SEOBOOK:

My thoughts are that SeoBook may also be dealing with dupe content issues due to it showing the full blog post on the main page and also on the single post itself. Note that the post on the main page is ranking (Title- SEO blog) while the original single post does not show (Title- Damned by Google). Both are the same content:

-Main blog page: www.seobook.com/blog
-Single blog post: www.seobook.com/dmd-damned-by-google

[edited by: tedster at 6:51 am (utc) on Apr 26, 2011]
[edit reason] make URLs fully visible [/edit]

[tedster]

I appreciate your thoughts, mslina2002. Let me first comment here that we do not normally do analysis of specific websites in public on WebmasterWorld. There are too many potential problems if we go down that path.

-----

In the case of SEObook, it is an almost universal trait of blogs to reproduce recent articles on both the Home Page and also on the permalink. I'd be very surprised if this situation is creating Google the ranking problems we see with scrapers outranking the original.

And though it is definitely a best practice to utilize the "www" canonical redirect, I rarely see a real problem from that in practice today - not unless somewhere on the site there is unique content on one and not on the other.

The with and without "www" canonical situation certainly used to be a major problem for websites, but it was also a major problem for Google. In the real world, there was just no way they were going to get all site owners on the canonical bandwagon. So rather than letting that issue continue to degrade their search results, Google has adapted and they usually compensate well for the situation.

There are other common situations that used to make trouble but where Google now accommodates the "errors". A lot of common IIS issues are now handled by Google very well - inconsistent use of case in the URL, and "Not Found" URLs that return 302 to 200 are two of those.

So while I would never recommend ignoring those technical issues, I also do not think that either Matt's blog or SEOBook are being outranked by scrapers because of that kind of thing.

[migumbo]

Google isn't rewarding good content, there have been countless examples of this over past 3-4 weeks (but no real world examples on WebmasterWorld for smaller sites).

My experience:

All websites 6-7 years old (about 10 in each category)

First ~10 Websites that were pandalized; quality content, well researched and written, with continuing effort and improvement to the site. Analytics installed on all sites. have been destroyed by Panda (more than 50% drops). Feel like shooting myself in the head.

Websites that haven't been touched for 3-4 years, that have html pages instead of a CMS, a confusing and crappy looking design, poor html, no analytics. up 50% or more. all of them. This is no coincidence.

I haven't been Pandalized, I've been Pancaked !

[pawas]

I believe that Panda is a completely business move by Google. Offer less relevant results and more people will click on their ads.

[koan]

Offer less relevant results and more people will click on their ads.

Ok can we stop reading that every time there's a change in the algo? It's old and busted.

[TheMadScientist]

So Google can do no wrong in your eyes.

Google's going to do what they do, whether I think it's right or wrong.
I don't concern myself with whether they got it right ... I try to find a way to rank.

I'm simply pointing out flaws. There is not one site you can point to that is "perfect", you can always find something wrong with it. Even yours that haven't been affected by panda.

Exactly ... That's why I don't sit around and point out Google's flaws ... It's a pointless waste of time and NOT my job ... (My sites and finding + fixing their flaws are my job ... And yes, I know of some flaws I haven't gotten to yet.) ... They deal with a billion queries a day ... They couldn't care less about an individual site (for the most part), because they have to look at the big picture and overall satisfaction ... If they are right 99% of the time they still miss on 10,000,000 queries a day ... When they make an update they look at the overall number, and yes, where and how they 'miss' shifts, but they can't 'micro-analyze' on a per-query basis like a webmaster does ... Why doesn't Matt Cutts page rank #1 for a long-tail exact match in quote? WhoTF cares? Not the average searcher by any stretch of the imagination, imo ... If you want to find the page that bad and know the exact title of the page you're going to find it, relatively easily, even if it's not the first *bleeping* result or even on the first page.

When my 3 month old site out ranks my 9 year old site, and my 3 month old site has one link to it from a directory (BOTW), something is wrong.

Yes, what's wrong is you're complaining about it rather than learning from it ... It should tell you something.

CainIV I usually agree with your logic, but your reasoning seems off a bit to me:

Google's market share drop was almost immediately following Panda, but as you say, users do not change immediately, which means to find the most likely cause of a drop due to dissatisfaction we would have to look at the months pre-Panda, not at Panda itself.

The alternative I see is what I posted earlier in this thread (msg 16): People who find what they are searching for stop searching ... If the market share drop was nearly immediately following Panda and the cause was actually Panda, imo, it's likely because people are searching less to find the answer they were looking for, which gives the opposite interpretation of what the numbers seem to indicate ... If the market share drop is indeed related to Panda, imo, it was too soon for a shift from dissatisfaction and is more likely because people don't need to make as many queries to find the answer they were looking for ... Think about it for a minute.

[sanjuu]

I'd bet that 99.99% of Google's users are completely unaware that there's been a change in the algorithm Google uses to create its SERPs.

On a subjective level I've seen the results for some phrases that I've been monitoring deteriorate with the Panda update, and the results are subjectively poorer in my opinion (because I know what was there previously and can see what's there now).

A user goes to Google, carries out a search, and gets three or four 'big' brands, followed by a variety of other sites. Even if none of these really match what they've searched for, they might be able to find some answers or something close - or they might even refine their search.

What the user can't do is 'miss' a result that no longer shows. They simply don't know it exists, so if they're unable to compare the current results with what was showing previously how can Google say the 'quality' of the results are better/worse than previously, without using some arbitrary indicators?

Google accounts for about 80% of search in the UK. I can't see that changing any time soon.

I see lots of searches full of non-UK sites (for example in automotive-related searches) in the top 10 results, or results that don't match the 'long-tail' search I've done, etc. But a user doesn't really know if the results are good or bad, they don't care if some site owners have been hit by Google's update, so why should Google?

[chrisv1963]

If the market share drop was nearly immediately following Panda and the cause was actually Panda, imo, it's likely because people are searching less to find the answer they were looking for, which gives the opposite interpretation of what the numbers seem to indicate

Is the market share based on the number of searches or on the number of visits to Google, Bing, ... ?

[TheMadScientist]

Search query percentage if I remember correctly ... I don't have time to go look right now and double check + cite, but usually the market share numbers are % of overall queries that took place on each engine. IOW: Unless this one was a 'special' market share report, it's % of queries.

[falsepositive]

I use classic Google. I go to Bing. :) It works pretty well for me!

[asabbia]

I am noticing improvmennt.. Up to like +10% today. has something chagned again?

[walkman]

I am noticing improvmennt.. Up to like +10% today. has something chagned again?

Something has happened /is happening. I can't tell at the 10% level due to adding /removing pages but something's up in the air. Also noticing heavy and deep crawling this weekend and today.

[rlange]

OK. Newbie question... (I hope it's not too out of line to ask it here.)

asabbia wrote:
I am noticing improvmennt.. Up to like +10% today. has something chagned again?

walkman wrote:
Something has happened /is happening. I can't tell at the 10% level due to adding /removing pages but something's up in the air.

How are you folks checking this? I know that Google Analytics allows you to check today's numbers in real time (more or less), but is there some way to easily compare, say, "12:00am through 12:00pm today" to "12:00am through 12:00pm average of the previous 30 days"?

Or are you using some other tool altogether?

--
Ryan

[walkman]

I have an internal referral thingy, I can see numbers and keywords from Google, Bing, Yahoo and all. I also compare raw log size, most of the time (barring a major scrapper) is about right.

I got excited last week too though, so I'll wait.

[asabbia]

@ryan: for real time stats (free but without advanced features) just go with statcounter.com

[tedster]

I am noticing improvmennt.. Up to like +10% today. has something chagned again?

I'd like to bring a bit more organization to our discussion now that Panda has been with us for many weeks. Google makes changes daily or better - so we maintain a regular thread for discussing those changes. Google Updates and SERP Changes - April 2011 [webmasterworld.com]

I copied the last four posts from this thread over there, too. When it come to watching Google's moves, having them scattered through a couple dozen Panda threads is not going to be good for future reference.

So let's take the discussion of daily changes to the dedicated thread [webmasterworld.com] and leave this thread for discussing whether Panda was a "brilliant move" or not. Thanks all.

[apauto]

statcounter isn't very accurate... overinflated numbers...

[netmeg]

I use a boatload of different analytics and stats programs for different reasons, would be happy to discuss relative merits - but in the Analytics forum, if you wanna start an item on it.

[webmasterworld.com...]

[Rlilly]

Any move which de-stabilizes businesses and creates so much uncertainty on such a large scale, cannot be a good move PERIOD.

The amount of hate towards Google resonating now and witnessed by so many replies on Webmaster World in different thread will have a long term effect on that company.

[netmeg]

As a percentage of total people online who use Google, we're a pretty small group. Even if you toss in the users on the Google forum, we're still a pretty small group.

[econman]

leave this thread for discussing whether Panda was a "brilliant move" or not

My take: the new focus on "quality" rather than an exclusive focus on "relevance" was a smart move, but I wouldn't call it brilliant.

I think it was a smart move because the explosion of scraped and junk content has gotten so bad it was adversely affecting Google's profitability and reputation. They needed to do something about this problem, and with Panda they are at least trying.

But, the problem is that they aren't achieving a high enough rate of success. Perhaps the actual execution was badly handled, or perhaps they are counting on the "machine learning" process over time, and in the meantime things are pretty messy with lots of collateral damage.

But, I don't buy the notion they got this one right out of the box. I'm seeing way too many examples of lower quality and/or scraped content moving above higher quality and/or original content -- with the notable exception of big brands -- but it doesn't take a team of brilliant Ph.D.s to push familiar brands to the top of the SERPs -- the challenge is figuring out the right way to rank the stuff published by the long tail sites.

[Rlilly]

@econman "the new focus on "quality" rather than an exclusive focus on "relevance" was a smart move"

If that is the case G is not a search engine anymore. There has to be manual reviews of sites and input.

[zerillos]

When I search for widgets that I sell that I am now on page 2+ for, the first page is nothing but forum posts, review sites, and nothing from an ecommerce site until page 2. Even articles rank higher.

why do you consider people use the internet only for buying stuff?

[tedster]

When I search for widgets that I sell that I am now on page 2+ for, the first page is nothing but forum posts, review sites, and nothing from an ecommerce site until page 2. Even articles rank higher.

Google generates automated taxonomies for the user intention related to various query phrases. In broad sweeping categories, they are something like informational, transactional, navigational, locational - and various mixtures.

Google also has taxonomies for various types of web pages and websites. Part of their technology also involved matching the right mix of destination URLs with the user intention behind the query.

It sounds like this query term has been slotted into an essentially informational taxonomy.

[shallow]

My site now hangs on life support of 18k/day and declining weekly.
But here is one reason why Google make a brilliant move that I give them a A++.

Your site, like mine, now hangs on life support and you give Google an A++?!?!?!?

Go figure.

All of a sudden I figure out that is what Google want all webmaster do and be serious about your site and make it as perfect as possible so that the internet will have more and more great content.

Many of us small to medium sized site owners (individuals not corporations) have been doing these things in earnest for years. Then "Boom" we suddenly get punished after all our hard work.

The only things I'm going to change going forward is to put less time and effort in my site, and certainly not pump the money into it as I've done in the past.

Why should I? To all of a sudden get punished again some time in the future? No thanks.