| This 116 message thread spans 4 pages: < < 116 ( 1 2  4 ) > > || |
|Quality According to Google - Official "Guidance" on Panda Update|
Finally a post from Webmaster Central about what Google thinks about quality:
I really need to chime in here.
First of all, that is a great blog post by google and very very helpful if you know how to read between the lines. I recommend everyone read it at least 5 times.
Now, this is not there secret sauce but they do give us some tips as to what panda is about, and much of it I have greatly speculated about here on these forums.
Lets take the main points from this list, the realistic ones and not ones such as "Is this article written by an expert or enthusiast who knows the topic well, or is it more shallow in nature?" you can all but throw this out the window. Here are the points everyone should be paying attention to.
- Does the site have duplicate, overlapping, or redundant articles on the same or similar topics with slightly different keyword variations?
We have discussed this in the panda threads. Having same/similar articles with slightly different phrasing can be a major factor with panda. Everyone should read this point at least 10 times, I overlooked it until I read the article for the 3rd time and then the wheels started turning. If you overlook this point and think it doesnt apply to you, you need to really check your site and make sure this does not apply to you.
- Would you be comfortable giving your credit card information to this site?
This one is not as important to me as all my ecommerce sites are doing fine and guess what? they all use copied manufacturers descriptions, so all those who say duplicate content/manufacturer descriptions are being pandalised, look elsewhere for your problem source.. But to everyone out there with an ecommerce site, do make sure your site is SECURE meaning that all your SSL certificates are valid and all your pages that should be secure are in fact secure.
- Are the topics driven by genuine interests of readers of the site, or does the site generate content by attempting to guess what might rank well in search engines?
This falls in line with over optimizing for google. Its tempting to put your keywords everywhere, but dont do it....just dont.
- Does the article provide original content or information, original reporting, original research, or original analysis?
I theorized on these forums about many sites that write about the same story. It didnt get many responses but this further brings my theory to light. Lets take your typical gossip/entertainment site. They all mostly write about the same thing. Every time Lindsey Lohan is arrested, every gossip site has the same story written differently but with the same point. Be original and write about something nobody else is writing about (googles words not mine).
- Is the content mass-produced by or outsourced to a large number of creators, or spread across a large network of sites, so that individual pages or sites donít get as much attention or care?
A lot of people talked about syndication. I dont agree with google doing this, but from this point it sounds like they do not want you to spread your content to other sources. Big thumbs down for google on this one and not giving the proper credit to the source.
- Does this article have an excessive amount of ads that distract from or interfere with the main content?
Here is the point I have been pushing big time on here. I think I was one of the first to push this idea in the "sites that dont fit the mold" thread. If you have too many ads on your site it is taking away from your users experience. Do not use deceptive ads or ads that overwhelm your users.
As for Matt Cutts saying that pandalised sites can get released when the algo refreshes, that is great news for everyone. Everyone should assume that it will take 3-4 months, so keep working on your sites and cover all of your bases. If you have one small quality issue on your site, fix it, you never know what might help you break free of panda.
pageoneresults, "But, there are a whole nuther set of factors to look at when dealing with image optimization." Sure, but that's assuming the algo even GETS that the site is supposed to be image heavy. If an algo is focused on text content, runs a test to see how the site looks without images, determines it sucks because there's not enough text on the page, it may never get to that "whole nuther set of factors". It used to get to that set, before Panda, but I'm not convinced it does now. Or if it does, it may only evaluate that AFTER applying the "Panda, your site is low quality, smackdown", so after the ranking gets pushed back X amount, then it gets folded into the results at that point based on that "whole nuther set of factors". Now, obviously, I'm not saying this is a fact, just saying that it sure feels that way. I just think some forget that content doesn't have to equal articles. An artist, for example, who shares his art as downloadable images, for instance, shouldn't be required to stink up the site with lots of text to please the content quality filter. Optimize the images and what content there is, sure. But don't accidentally place a low-quality stamp because the algo forgot to consider that not all sites focus on articles.
I read the article about 10 times and went through all of my sites that got hit and none of them fail the test. What surprises me is that my site contains infographics which I paid more than $#*$!x per piece with links from tons of publications like WSJ/NYTimes I did not force them to link they are all editorial links we were still hit.
In regards with Ads, Google SPAMS MY INBOX on daily basis that I am not utilizing the ads across my domain. They even mailed me a layout which if you want I can show you.
And it looks 100% spammy layout but that is what Google is sending through email. So first they tell me to place ads and then another guy from another section of Google does not want that layout.
Can Amit sit down with Adsense team and make up their mind?
Aaron did a great post on his seobook site which clarifies a lot of points I am trying to make.
|Google have already moved on from Panda. It's here to stay. It's being improved as we move forward. If you've not been able to recover as of this date, there's a good chance you won't be recovering anytime soon |
|They have already run the evaluation again at least once during panda 2. |
Personally, I do feel that they have run it one more time after panda 2 as well.
OK, how do we know when /if they have run it? April 11th was for Panda to go global on other English speaking countries. They weren't touched during panda I so they are different indexes. We just haven't seen a "We're back" tsunami or even a trickle. Just, 'I'm ranking for one one or two pages,' 'lost it again'...
|For example, many of those hit by Panda are what I refer to as UA Abusive. I see sites making 300+ HTTP Requests per page weighing in at 2.5MB+ after rendering. That's huge from my perspective and something to carefully consider optimizing. |
Ever seen Huff Post or Business Insider with their gazillion outside ads, js etc? A Huff Post Bin Laden story I just checked was 2.64+ MB in total.Now they have a lot of positives so an average site would get crushed but unless Page Speed screams, most sites don't have to worry much about it. A decent, above average speed, pageload time covers all that for most people. Just checked Google.com and it doesn't validate.
Edit: I am starting to think that certain pages have improved (I see it on own pages) but the sitewide penalty /low rank score it still there. So it seems to me that the overall score hasn't changed.
|I am starting to think that certain pages have improved (I see it on own pages) but the sitewide penalty /low rank score it still there. |
This lines up with what I'm seeing, too. We still can't say for sure, but it is beginning to look like site-wide scoring is not an area that gets dynamically updated. Instead, for now at least, it will run in occasional (and very rare) batches - more like automated taxonomies, query intention types, or even "good phrase" identification in their semantic engine.
I imagine (and hope) that the Caffeine infrastructure will eventually be able to handle more frequent updates - but full infrastructure integration may also be a big job. Still, with Panda having been in the planning for more than a year, I would hope that Caffeine will handle it eventually.
|Ever seen Huff Post or Business Insider with their gazillion outside ads, js etc? A Huff Post Bin Laden story I just checked was 2.64+ MB in total. |
I have. Ever seen their server response times? Also, I don't see that Business Insider was affected by Panda. Huff Post is on the rise. I should have added a disclaimer that there are going to be exceptions to the rule, huh? ;)
|Now they have a lot of positives so an average site would get crushed but unless Page Speed screams, most sites don't have to worry much about it. A decent, above average speed, pageload time covers all that for most people. Just checked Google.com and it doesn't validate. |
How many sites fall in the "decent, above average speed"? Here's a quote from Google about Site Speed...
|At Google, we are passionate about speed and making the web faster, and we are glad to see that many website owners share the same idea. A faster web is better for both users and businesses. A slow loading landing page not only impacts your conversion rate, but can also impact AdWords Landing Page Quality and ranking in Google search. |
Measure Page Load Time with Site Speed Analytics Report
May 4, 2011 - [Analytics.BlogSpot.com...]
While the two sites you mention are abusive on the page weight side, they're able to deliver that content in an optimized fashion. Many sites I've reviewed appear to not have the backbone to support that type of payload consistently at high speeds. There's a lot at play here when discussing the speed of the site. You're using two major players as an example and they've done quite a bit from the technical side to accommodate the serving of larger documents.
And yes, I agree with you, they are both heavy sites. They've got the foundation to support the weight.
|A faster web is better for both users and businesses. A slow loading landing page not only impacts your conversion rate, but can also impact AdWords Landing Page Quality and ranking in Google search. |
This is the first thing I did after Panda. Added cdn, css sprites, gzip etc. My site score is 94/100. But, G traffic is still bad. I don't think it is one of the Panda factors.
Pageone, there's extremely fast, fast, OK, somewhat slow....I'have had enough. I doubt you can extra points by being extremely fast when 'fast' is good enough. So speed matters but not every milisecond
|I'd be hiring another set of eyes to take a close look at everything. Existing site owners typically have blinders on and miss the bigger picture. I've seen sites hit by Panda and have seen the owners complaining. One look at the sites and you can clearly see why they were hit. |
My site was hit so badly by Panda that it now takes several days to make enough to pay for just one hour of my web developers time! So how can I afford "another set of eyes?!"
I never made a huge income like many of the website owners I read about in these forums. But I made enough to pay fees in the five figures during the last three years for web development and design, hiring very good writers to write content for my site (besides my own; I was a writer long before I owned a website). I hired an expert last year to review my site for layout, ad placement, suggestions for site improvement, etc.
I even had money left over to pay some of my personal bills.
I'm an individual site owner, not a business or huge company. Panda has just about wiped me out. Income down more than 50%.
Little money to hire anyone now. Besides, I'm not really inclined to do so after this Panda experience.
I'm not sure anyone really knows what to do to improve some sites. It would be a crap-shoot, would it not?
The only way I figure I'm going to make a little money now is by not spending it for other eyes(except for absolutely necessary things such as hosting and security updates to my site).
|My site score is 94/100. But, G traffic is still bad. I don't think it is one of the Panda factors. |
What your results might conclude is that Page Speed is not one of the Panda factors that hit YOUR site. I think it's clear that with such a large number of factors involved, only some would apply in any single case.
My site score is 99! The blog at my site is graded 91. It's in the top 0.49 % of all websites.
Can someone please tell Google this.
Is this grader for real?
|Besides, I'm not really inclined to do so after this Panda experience. |
This will become the dominant attitude across many if not most ecom sites ~ as long as so many are so dependant on Google, how can anyone hire in good conscience, knowing that the rug can literally be pulled out at any time? Or at the very least, all hiring will be on a subcontracting basis, with no permanent employees. There is simply too much to lose, and it can happen overnight.
Everyone should read this reply by a guy who posted on the article but it wasn't approved I guess it made Amit look stupid..
Let's see how this blog post rates according to Google's recommendations:
* Would you trust the information presented in this article?
Definitely not. This is a public relations piece which ventures far from the truth. It is patently dishonest.
* Is this article written by an expert or enthusiast who knows the topic well, or is it more shallow in nature?
Amit may be an expert, but this post is the very definition of shallow. It presents no useful information. It's worse than an eHow page.
* Does the site have duplicate, overlapping, or redundant articles on the same or similar topics with slightly different keyword variations?
Definitely. This post is yet another regurgitation of Google's standard content-free public relations platitudes.
* Would you be comfortable giving your credit card information to this site?
It's Google -- we don't really have a choice any more. Google has handed the top SERPs to bland brand sites which lack relevant content -- forcing everyone else into AdWords. Coincidence? Cui bono?
* Does this article have spelling, stylistic, or factual errors?
This article purposefully contains serious factual errors. It is written to obfuscate the truth, which is that the Panda update is a dismal failure which has penalized quality web destinations like MerckManuals and AskTheBuilders.
* Are the topics driven by genuine interests of readers of the site, or does the site generate content by attempting to guess what might rank well in search engines?
Based upon Amit's unwilling to provide honest answers, it is pretty clear that he isn't "driven by genuine interests of readers of the site." This is a bait-and-switch page. The readers came looking for information and were delivered only prevarication.
* Does the article provide original content or information, original reporting, original research, or original analysis?
Nothing in this post is original. This is simply a mashup of half-truths and misdirection which Google's PR droids have already posted elsewhere. This page should be removed from the search engine results by the duplicate content filter.
* Does the page provide substantial value when compared to other pages in search results?
This page provides no value at all. It purports to discuss the Panda updates and instead lists a number of attributes which Google would like web pages to possess. The vast majority of these attributes, however, are not algorithmically measurable. This means that they cannot be part of the Panda algorithm, or any algorithm. This is purely duplicitous behavior on the part of Google.
* How much quality control is done on content?
Quality control on this article was performed only by the legal and propaganda departments. This article has no technical quality to measure. Simply put, the quality is vapid.
* Does the article describe both sides of a story?
Absolutely not. The article equivocates regarding Google's side of the story and completely fails to address the concerns of web publishers.
* Is the site a recognized authority on its topic?
Blogger, as a free web hosting platform, cannot be considered an authority on anything.
* Is the content mass-produced by or outsourced to a large number of creators, or spread across a large network of sites, so that individual pages or sites donít get as much attention or care?
Blogger content is mass-produced by a huge number of creators, many of which do not give their individual pages significant attention or care.
* Was the article edited well, or does it appear sloppy or hastily produced?
The article looks like a cheap cut and paste job from Google's webmaster forums.
* For a health related query, would you trust information from this site?
I would definitely not trust health information from any should who could not clearly and objectively explain their diagnosis. This article clearly fails to meet that standard.
* Would you recognize this site as an authoritative source when mentioned by name?
Definitely not. Blogger is a spam pit.
* Does this article provide a complete or comprehensive description of the topic?
Definitely not. This article is virtually content-free. It provides absolutely no useful information.
* Does this article contain insightful analysis or interesting information that is beyond obvious?
Definitely not. This article eschews analysis, and in fact recommends against analysis.
* Is this the sort of page youíd want to bookmark, share with a friend, or recommend?
Definitely not. I wasted enough time reading this drivel. I would hate to be responsible for another person being subjected to this soul-sucking post.
* Does this article have an excessive amount of ads that distract from or interfere with the main content?
For some reason, Google won't run AdSense on their own pages. Perhaps that should be a hint for the rest of us.
* Would you expect to see this article in a printed magazine, encyclopedia or book?
No one but Google would publish this article, because it is completely lacking in honesty.
* Are the articles short, unsubstantial, or otherwise lacking in helpful specifics?
There are no helpful specifics. The "content" is completely unsubstantial.
* Are the pages produced with great care and attention to detail vs. less attention to detail?
There are no details, only broad and useless platitudes designed to deflect criticism of Google's failure.
* Would users complain when they see pages from this site?
The comments are full of users complaining about this content.
Clearly, the entire blogger domain should be penalized for hosting low-quality pages such as this one.
maybe one of the reasons that some decent websites are suffering is this:
if google really are running panda only infrequently, like people are suggesting, then whole swathes of sites are going to get demoted for days, weeks or months (who knows?)
if those sites are linking to you, then presumably all those links will be devalued too.
that means that your perfectly decent site might appear to be getting a "penalty" when in actual fact its because bazillions of backlinks from so-called "poor quality" sites are no longer carrying any weight.
there would be absolutely nothing that you could do to get your traffic back, short of getting brand new links.
|Everyone should read this reply by a guy who posted on the article but it wasn't approved |
I realize that webmasters have always had a love/hate relationship with Google ~ the intensity of which has gone up & down ~ but in all my years (I was online before them!), I have never seen this level of outright animosity.
When I started my first business in the mid-70's I had an old guy tell me that one unhappy customer will do more to hurt you than 25 satisfied customers will do to help you. Under that ratio, Google is the target of one seriously vicious anger attack (and from my point of view, they deserve it).
It's just mind boggling Google puts out such tripe. Panda is a disaster because, as almost always, Google thinks they do a better job at what they do than, well, everybody else on the planet thinks.
What they write in that blog post is great, except they don't value what they say they value. What they value is middling quality content on generically reliable websites.
In contrast, what they don't value is high quality, in depth, original, expert content. Why? Because tha nature of the Internet is high quality content is stolen by low quality sites, which leads to the phenomenon of one expert site having content that is on dozens or hundreds of other sites... and Google then (stupidly beyond words) penalizes the expert site, because it appears to be (to them) a low quality site because the content is copied everywhere.
That's mind-numbingly stupid search engineering.
So, they penalize original quality, while boosting original mediocrity -- "original mediocrity" meaning accurate but not authoritative or in depth coverage.
The true secret to ranking in Panda... have 300 words of text on a page that is so tediously dull that no one would ever dream of stealing it, and then get some decent quality links.
|Everyone should read this reply by a guy who posted on the article but it wasn't approved I guess it made Amit look stupid.. |
Well, this comment is now approved and published, but it was not there this morning.
|one unhappy customer will do more to hurt you than 25 satisfied customers |
But one unhappy guy made happy by customer service is worth more than normal customers i.e. solve the problem and he tells his friends..
|if those sites are linking to you, then presumably all those links will be devalued too. |
I don't remember anything from Google about that, or anyone mentioning private research in that direction, either. Even within the site itself, Google didn't say anything about the downgrade traveling by links. And as far as I know, the downgrade is still only within that domain.
My own research suggested (but certainly didn't prove) that within a domain, the downgrade might travel by BACKTRACKING along link paths. In other words, if a page links to the demoted page, then it too gets a (smaller) downgrade. But I haven't seen anything that suggests a Panda value can jump domains.
|Does the site have duplicate, overlapping, or redundant articles on the same or similar topics with slightly different keyword variations? |
Ok, I have several ecommerce sites selling similar type specialty merchandise.
Even though these "questions" seem to mostly talk about "articles", on ecommerce sites the descriptions of our products may be similar, though not exact. For example, let's say I sell black cotton widgets. Some black cotton widgets may be black cotton widgets with purple stripes or black cotton widgets with yellow polka dots or black cotton widgets with with pink triangles. There are only so many ways to describe these widgets as they are very similar or sometimes the same type of product but varying colors, materials, etc.
There are only so many ways to describe these products when the main difference is style, color, etc.
I feel that ecommerce sites (IMHO) should not be scored the same way as an article or information site. If we have over a thousand products on our site and they are similar yet different and there are only so many adjectives to use to describe it and what it's used for, it does not mean we are trying to keyword spam anything, we are just trying to accurately describe our merchandise.
Any thoughts about this?
|if those sites are linking to you, then presumably all those links will be devalued too. |
|I don't remember anything from Google about that, or anyone mentioning private research in that direction, either. Even within the site itself, Google didn't say anything about the downgrade traveling by links. And as far as I know, the downgrade is still only within that domain. |
what i mean is, having a PR8 site link to you is going to give your site a boost, whereas having a PR1 site is going to have no effect, for example.
so if panda has demoted a load of seemingly decent sites down the SERPs, then presumabley their worth to other sites as backlinks has also dimmed.
if your site has a lot of these kind of backlinks, then your site is definitely going to suffer. its as if all your decent links have disappeared overnight. and then every site that YOU link to is going to suffer as well, and so on.
the only sites that wont suffer are ones that have huge backlink profiles -- typically the "trusted" branded ones.
so it doesnt matter how much onsite fiddling you do, and how much quality copy you add to your site -- you are still going to suffer until your backlink profile improves. people are looking at their sites thinking there is a problem with them, but there isnt -- the "penalty" that you have received is nothing more than other people's sites getting downgraded.
londrum, I haven't seen any evidence of links from other panda affected sites being devalued. I have seen a couple of pages recover and these get links from guest posts on sites which are affected by Panda.
but if the sites are getting demoted down the SERPs, how can their links continue carrying the same weight as before?
it would be like google saying "we think this site is low quality, but we still regard its links as high quality"
we're just speculating though, no one really knows. Just because it may make sense doesn't mean Google took it in consideration. We don't know how they treat pandalized sites, other than not rank them. Everything is one big black box
sure, i know its just an idea. but it might explain why a load of perfectly decent sites are getting hit through no fault of their own.
|But I haven't seen anything that suggests a Panda value can jump domains. |
I don't see this in the same context. If a page value is diminished , then I would think the value of the links from that page would be reduced - so i wouldn't pigeon hole it as a "Panda Value". Not sure if i read your line correctly , but thought i'd put this forward.
|but it might explain why a load of perfectly decent sites are getting hit through no fault of their own. |
It's impossible, google can do no wrong. Your site must suck, you are trying to cheat google, follow the guidelines, they have PHDs but your site has misspelled words :). Just joking.
It's frustrating, my worst sites got much better rankings. So chuck it up to flux, an innocent 'mistake' on your part that will solve as time goes, or google sees it differently and your site will need some fixing.
One thing it's clear, Google is consolidating, a few top sites are getting more and more and more traffic from Google and others will probably be forced to eventually close down soon. Feast or famine
Give me a break! If you saw what is outranking us in our niche you would cry laughing.
Here is what now dominates our niche
sites filled with malware warnings in the SERPS
yep I can totally see how these sites fill those guidelines and are of much higher quality then a handmade 4 scroll review written by experts and has thousands of natural IBL's.
It is ALL so clear now! lol
|if the sites are getting demoted down the SERPs, how can their links continue carrying the same weight as before? |
There is no reason to think that Panda scoring affects PageRank - no evidence of that at all. It might - we can't say because toolbar hasn't been updated. But Google certainly described Panda as a new component to the algorithm, rather than a change to any existing part.
There is a lot of talk about links being the big important thing with the Panda update. My research indicates that links have MUCH less to do with the results in our niche. Having said that, I'm not going to rule out that links from spammy neighborhoods might be a factor in the Panda-pounce; however, as a factor in ranking well (in the first positions in google), I think that the on page factors that your site represents is more important post panda.
|There is no reason to think that Panda scoring affects PageRank - no evidence of that at all. It might |
I'm seeing pages on effected domains having good backlinks from unaffected sites with original content holding. I'm seeing pages with limited backlinks and original content falling.
I'm seeing site's launched inside a week ranking and holding , with minimal links and good trust signals.
i'm not seeing any applied remedy changing either a page or a site. So I really think a time penalty factor or dampener is in place.
To me, nothings changed , except the levels of detection in the algorithmn calculating the individual and aggregate scores of each page by site and referal to a sum that effects one page or a total site.
From the remedy perspective, I'm seeing a delay in anything moving back on effected sites - so I'm assuming " trust" has to be restored.
With regards to links - ( and I've spoken with a lot of respected and much smarter folks than me over a wide range of client sites , who discount links as part of this Panda update ), i still would consider that in the case of a total site, if enough of the pages fall beneath a score threshold by virtue of a lack of links the site becomes more vulnerable. And i would strongly recommend the need to be " open minded " on the possibility that those pages have limited link referal capability, effecting the " link juice economy of the web"
With regards to the effects. what perplexes me is that Google is often only partially knocking off traffic to producing pages and i have yet to see analysis that strengthens or dismisses the theory that some sort of " traffic throttling" may be occurring.
| This 116 message thread spans 4 pages: < < 116 ( 1 2  4 ) > > |