Big changes in Google content scoring

Great great stuff.

ipopbb:

What's your opinion on the factor of the PAGE age. You mentioned that domain age is NOT a factor at all in another thread, and you believe the "first date of indexation" of the domain is the factor to consider for google, so what about, let's say, you have two pages on THE SAME domain, (doesn't matter if the domain is old or new) one page is created several months ago, and one page is created just recently, would that be a factor for google to consider in ranking them?

Thanks

fpi2009 said:

Great great stuff.

ipopbb:

What's your opinion on the factor of the PAGE age. You mentioned that domain age is NOT a factor at all in another thread, and you believe the "first date of indexation" of the domain is the factor to consider for google, so what about, let's say, you have two pages on THE SAME domain, (doesn't matter if the domain is old or new) one page is created several months ago, and one page is created just recently, would that be a factor for google to consider in ranking them?

Thanks

Click to expand...

This will certainly start some fights because many people are financially invested in Age as factor with either buying or selling domains.

MY OPINION (not proven fact):

Sites that rank well tend to be older. No one should find this surprising. It makes a lot of sense. However, having an old site eliminates you as a certain kind of web spam. Beyond that it doesn't equate to relevance or authority of any kind for keywords. You still have to succeed on merit in all the other areas. The kinds of web spam it distances you from tend to be less than 10 months old. The only noteworthy change in statistical advantage I have ever found in analyzing the SERPs (data is over a year old now) appeared to be the lack of a penalty once your website has been indexed for about 10 months.

I wouldn't say (or possibly shouldn't have said) age is not a factor. I just don't think it is a major factor. I think there are less than about 8 factors that explain 90+% of the SERP rankings and I would guess that Age is one of the hundreds of slight tie breaker conditions that Google implies with its "hundreds of factors"... It is really something like 8 factors and a bunch of tie breaker conditions and known spam penalties and specific niche related factors.

If you buy aged domains be sure to use wayback machine to make sure they were index by search engines... That wont guarantee advantage but at least you know google could have indexed it as a website way back then. Otherwise its like the hundreds of 15 year domains I have that have never been used as websites and have no measurable advantage in the SERPs over a new domain.

I think Age is popular as a factor because it correlates so strongly with sites that rank well that people tend not to see how many old websites don't rank well... which is most of them. As any good economist will tell you... Correlation is not causation. But the jury is out, and I am wrong more often than I'm right, but I am also right more often than most... wow! how does that work? Oh yeah, and the average online shopper has 1 mammary and 1 testicle. You online shoppers are a freaky bunch! But you hopefully see my point about reading too much in the data. It is certainly one of my most common mistakes.

Ted

ipobb any news this year?

wow thats crazy

H1 and H3 but no H2?

ipopbb i have several questions...

1. if let's say the website was indexed a year ago but then the site became inactive for 6 months before it gets updated again, does it have any impact on google algorithm on "site age"?

2. Similar question. If the website was indexed a year ago but then it lost its web host (dead) then a month ago let's say I revived the website with different content, how google sees this?

3. What kind of link building do you recommend that works with today's climate on Google SERPS considering i think link push/link wheel and scrapebox dont really bring a lot of effects anymore?

Rubbish. No offence, but you simply cannot desypher Google in this way by comparing a bulk of 250 websites against your "secret"keyword. There are way to many variables to thow into the equation to be able to reverse engineer Google with 250 websites.

If you could reverse engineer Google's algorythms in this way you wouldnt be working for anyone.....you would have people working for you.

hamd01 said:

Rubbish. No offence, but you simply cannot desypher Google in this way by comparing a bulk of 250 websites against your "secret"keyword. There are way to many variables to thow into the equation to be able to reverse engineer Google with 250 websites.

If you could reverse engineer Google's algorythms in this way you wouldnt be working for anyone.....you would have people working for you.

Click to expand...

Some people really have a preference to work for someone else. Especially those that like to setup labs and do tests and all othe sorts. They don't need/want the clutter of figuring out where the money is coming from and/or who is doing what.

People serve different functions at different levels... smarts and success doesn't mean you automatically now command other people. I don't believe Einstein was ever published as the centuries greatest employer. So please don't debunk what ipopbb has to offer or other members of this board because it doesn't fit into your elitist views.

ipopbb said:

I think Age is popular as a factor because it correlates so strongly with sites that rank well that people tend not to see how many old websites don't rank well... which is most of them. As any good economist will tell you... Correlation is not causation.
Ted

Click to expand...

Nailed it right there. I have said the same thing about a number of other popular 'myths' about what matters in ranking. Ironically enough, it is on a factor that you think otherwise on, namely exact keyword match of TLD URLs. It's the branding that correlates to the domain that matters more so than then name match itself.

Google for a number of years now has gravitated towards the branding method, and it is only getting stronger. Just because you register a domain called mypinkshorts.com and you rank for 'my pink shorts' doesn't mean it was because you registered that domain. It is because of all the links to your site that branded it as such.

If I were to take the same domain, but linked it with 'Orange Army Boots', and the content to match, then that domain would become branded as such, even though my domain name is something else all together. Showing that the correlation of an exact URL match is not a causation of SERP results.

I think there are only a handful of people who agree with me on this... but meh.. everyone else aren't the ones making me money.

I'd view the "Age of a website" and the "Freshness of the Content" as two separate factors. Freshness of a website is a top ten factor in my book.

I have made the mistake of pulling the plug on a website by deleting everything and letting googlebot find the 404 not found errors. Unfortunately google kept the results in the SERPs! 6 months later I tried to repurpose the domain for a new project and I found myself competing with the old dead links once I had a 301 catchall redirect to the homepage. It was awful since the old experiment reached PR6 on many pages.

Tune your site navigation. If your niche is health care and you have an "About" link change it to "About our "Health Care""... see how I tuned for an exact match of a specific short tail phrase. Link out to relevant pages internal or external. Outbound linking is a proven power play. Use paragraphs... not just the tags, but actual groupings of sentences. Industry keywords should be links in these paragraphs. Measure the SERP leaders stats. That will tell you what it is going to take to be competitive on page one.

Its all tools. You can build and awesome house with a hammer and you can build a crappy house with a hammer. Don't blame the tools or claim that hammers don't work anymore... (not that you made any such claim, just speaking in general terms here)... Prove things for yourself. After all... I might be making it all up just like the flood of other faith-based SEO advice that is on the web. I admit... I am wrong more than I'm right... but I am also right more than most. It is because I do things for myself and learn. When I do an experiment... whether my idea is proven right or wrong almost doesn't matter. The outcome is that I know information with a degree of certainty that the whole rest of the world doesn't know. Being wrong in my theories is often as valuable as being right. There are entire volumes of this I don't have to explore or wonder about anymore because I spent and hour now to find out a week later in the SERPs. It takes diligence and attention to detail, but there is simply no other way to get good at SEO except to become a scientist about it.

Take measurements, use scientific method. Imagine the time you'll save when one sentence into an SEO article the poster is recommending something that directly conflicts the empirical data. You don't have to waste time and brain power on that make believe SEO religion stuff... back button... next.

I'm selfish too... I want more peers. I want review your data! hint hint! Just like people are starving for my posts. I make bookmarks and take notes on other people's posts when they use measured data. I archive it locally. Its rare though.

But most of all. Your goal should be money. rankings, traffic, conversions, etc... all important but in a trivia kind of way... Now profit. That's got my attention. As long as there is growth in profits... who cares about the rest. So be sure to measure the right things.

wokaka said:

ipopbb i have several questions...

1. if let's say the website was indexed a year ago but then the site became inactive for 6 months before it gets updated again, does it have any impact on google algorithm on "site age"?

2. Similar question. If the website was indexed a year ago but then it lost its web host (dead) then a month ago let's say I revived the website with different content, how google sees this?

3. What kind of link building do you recommend that works with today's climate on Google SERPS considering i think link push/link wheel and scrapebox dont really bring a lot of effects anymore?

Click to expand...

MisterGemini said:

Ironically enough, it is on a factor that you think otherwise on, namely exact keyword match of TLD URLs. It's the branding that correlates to the domain that matters more so than then name match itself.

Google for a number of years now has gravitated towards the branding method, and it is only getting stronger. Just because you register a domain called mypinkshorts.com and you rank for 'my pink shorts' doesn't mean it was because you registered that domain. It is because of all the links to your site that branded it as such.

If I were to take the same domain, but linked it with 'Orange Army Boots', and the content to match, then that domain would become branded as such, even though my domain name is something else all together. Showing that the correlation of an exact URL match is not a causation of SERP results.

I think there are only a handful of people who agree with me on this... but meh.. everyone else aren't the ones making me money.

Click to expand...

I don't know that I disagree with this... If I did in the past I might not have understood. This an interesting and subtle distinction... Are you suggesting that google is using a behind the scenes "tags" field for web sites and/or pages? That could certainly explain some search artifacts I've seen recently if they are rolling their own "meta keywords" field on their end.

Damn, what an interesting twist. I don't know what to think about that.

hamd01 said:

Rubbish. No offence, but you simply cannot desypher Google in this way by comparing a bulk of 250 websites against your "secret"keyword. There are way to many variables to thow into the equation to be able to reverse engineer Google with 250 websites.

If you could reverse engineer Google's algorythms in this way you wouldnt be working for anyone.....you would have people working for you.

Click to expand...

LOL. Did you read the posts? No offense, but it is data. Its empirical. Google sorted the pages, not me. Its value and meaning are certainly subjective. Where do you think revenue comes from? executives are answerable to more people than peons. The CEO answers to customers, investors, employees, government, board of directors... If I had all the answers I'd be happy (wealthy too most likely)... being an executive doesn't mean you know more than everyone.

PS... I'm pretty happy and I'm in the highest tax bracket in my country. That doesn't make me right either.


But, opinion noted... one more in the disagree column of which there are many. And thank you for rebuking my data with something as strong as "your personal beliefs". That will teach me to use data!

Mate, i would like to know something...

how do you differ "sandbox" from "penalization"?

i have several sites which all were ranking in top 10 of their keywords but now they are ranked 400-500th (a sudden drop in a day)...is that sandbox or penalization? How to differ these?

wokaka said:

Mate, i would like to know something...

how do you differ "sandbox" from "penalization"?

i have several sites which all were ranking in top 10 of their keywords but now they are ranked 400-500th (a sudden drop in a day)...is that sandbox or penalization? How to differ these?

Click to expand...

Terrific Question! I get this one a lot. Here is my "BELIEF" ... its not a fact but it is a hunch that comes from suggestive observations...

This behavior has been seen by tens of thousands of webmasters in nearly all industries. It has many names. -30 penalty -500 penalty -900 penalty... I don't believe they are different "things" and I am not convinced they are penalties. I'll explain.

Many of the reports are from white hat sites that don't break the rules. Surely some of the claims of white hatness are false but when when you search for these posts on the web there are just so many from all walks of life that some must be honest too. This smacks of systematic process more than judgement. The other issue is why so much variance in the amount of penalty... its as if instead of be assigned as quantized amount of pain for doing something bad you are simply not receiving a factor as part of your ranking score and different websites have differing amounts of that factor.

When caffeine launched my PR8 when from 9.5 million backlinks in google to 2.5 million links. The rumor then was that google !@#$ed up and accidentally destroyed some data. I don't think this was the case. I think google has been purging low/no value backlinks and pages from the calculation process to improve overall performance of the system. I am probably wrong on the specifics of that, but something like that is happening and it is continuing to happen as a function of time.

Another possibility is that people are spiking a factor and when the change doesn't trend with history maybe you lose the factor until either a new trend forms or the old trend resumes. Or maybe you lose the factor until it can be fully recalculated by the overall process... Again I am probably wrong on the specifics, but possibly right on the idea that this is systematic process of some kind and not penalty.

That doesn't mean it doesn't suck. That doesn't mean it will ever get fix. For nearly all practical purposes it might as well be a penalty. But the good news is that it is so widespread that you and your website are most likely not getting specific scrutiny from Google. Odds are you are being injured with the "group" by google's changes last year and this year.

If you dropped after link building then stop link building for a while. If you dropped after big content changes then stop make changes and see if it comes back. If it doesn't come back try rolling back changes you made when it occurred.

Don't give up. The other important to note from these observations is the notion of "function of time". This is a long standing pattern with google. They have statistically determined that most new websites will give up and disappear in ten months or less. There is a very good likelihood that time heals all wounds if other actions can't. I killed some nice experiments out of panic. I regret it now... would have been just fine to let them ride. Don't pull the plug... let it ride if all else fails and start a new sister site to fill the time and hedge your bets. Odds are... everything will be OK if you just don't take too many reactionary measures out of panic. Breath and reflect. See if there are any clues that might help pinpoint the trigger and the factor you aren't getting credit for and compare notes... many many people are going through the same thing. Once we figure it out as a community we will simply dismantle it as a threat in our own practices.

Hope this helps!

ok thank you mate. another question, let's say i stick with one web design and structure for months. After i get good ranks in my keyword i use different web theme and structure for my site (let's say I use wordpress and i simply use another theme for the site), does it have any effect?

wokaka said:

ok thank you mate. another question, let's say i stick with one web design and structure for months. After i get good ranks in my keyword i use different web theme and structure for my site (let's say I use wordpress and i simply use another theme for the site), does it have any effect?

Click to expand...

Yes it most likely does.. the theme change likely also includes a change in your navigation structure on the page. Your website information pages might be at the top instead of the bottom.. navigation links might have moved from the left to the right, and thus spit out the content sequencing in a different order, and thus leading to a different content scoring on the pages.

To answer your question simply, yes, it has 'any' effect. Good or bad though is another question that can only be answered by what happens when you do it. I know some people who have switched themes and dropped, while others made gains. It really depends on the new layout and if it actually offers an onpage bonuses compared to your previous layout.

This might sound like a silly question but ok. When I write an article, it automatically shows up on the frontpage in a set font, like most CMS. The titles are just marked up text. If I want to make them <samp>, could I do this:


Which comes out at a better looking samp font. Is that legit? Does Google see it as samp? Just askin'

Hey,
based on your post I tried what you told!
Added:
02 samp
03 acronym
07 pre

And lost now around 3 to 4 places which means I lost a page one listing. I tried 3 Blogs and 1 site and all lost their rankings.
Do you think this is just temporary or does this method do not work for Europe (Germany)?

1.The content isn't scored but the tags are scored. Search engines are about content.
And providing the best content with the query.
Ditching content for the use of tags goes against there very nature.
2. The whole idea of someone sitting in a office deciding that the following tags are crap.
Doesn't go down well with me.
Especially in the way that you state,

Click to expand...

I think the op is refering to what to do when you have relevant and good content on the site.
From what I read he works with trusth worthy domains, I doubght there is no valuable content there

I think ultimately what people fail to realise here is that to actually validate whether this list or not is accurate, we would need to test it thousands of times and see whether this is true. we would need to have pages that do not link to each other and have the same link ratio from the homepage or external pages, and standardize the entire process to make sure that we can accurately track the effectiveness or lack thereof - of these tags.

Don't get me wrong it is extremely kind of him to have shared such a list but we need to all collectively work together and pool our findings to see whether this has an impact - baring in mind that Google very often change weightings and ranking factors.

In my personal opinion, the one thing you can take away from this list is that using clean and correct code on your websites, use the best formatting where possible - e.g. <address> .... </address> for instance or <samp> where relevant, which is really a tag code used to display output from a program or script or string of code; it will likely have a positive impact on your rankings overall.

Consider a site with no formatting tags whatsoever - just <html> and </html>, and little or no formatting. Is this a site a visitor or crawler would be impressed by or find useful?

Unlikely. Just my two cents.