Brand New Young Domains & the Google Sandbox [Is it a Penalty?]

[Chris Munch]

The question about the mysterious Google sandbox is one that pops up a lot so I though I'd post it here...

The sandbox is loosely defined at best and not largely agreed upon by SEOs.

What it generally refers to is that Google does not trust new websites, and therefore places them in the 'sandbox' and lowers their rankings until trust is established (age being a big factor in that trust.

This is because thousands of new websites appear online everyday and only a very small amount are genuine, decent quality and last for the long term, so Google improves its index by punishing new sites.

Whether an actual 'sandbox' exists, or whether it just takes time for a new site to develop trust with Google (or whether both of those are the exact same thing) is open to debate.

Having launched a number of sites and grown them to 6 figure traffic, worked with many other site owners closely who have launched successful sites, and watched the growth of a number of sites I can tell you this...

Google HATES new sites, with the exception of the homepage of keyword domains, getting new sites to rank highly is exceptionally difficult compared to an aged and trusted site. I find this lasts anywhere between 6 months and 2 years depending how well you promote your site, and the number of genuine editorial citations from authority sites. I find that as time goes on and your site grows your Google traffic and rankings will improve considerably if promoted correctly.

That's not to say you can't get good rankings and traffic in the short term, just that it does not compare to what you can achieve with an aged trusted site.

Please jump in with your own experiences and conclusions about new domains, rankings & the Google sandbox.

[mark587905]

Id have to disagree a little, I have sites indexed that a only a couple of months old.

[rajacks]

thanks again,

[Chris Munch]

Originally Posted by mark587905

Id have to disagree a little, I have sites indexed that a only a couple of months old.

Being indexed has very little to do with the sandbox. Your whole site can be indexed but still have low rankings. Often the sandbox seems to be more linked with the supplemental index.

Your entire site can be indexed (in the traditional sense) and still have a full on severe sitewide penalty where no pages rank including your homepage for its own domain.

Check out the various levels of indexing:

Source: http://www.seomoz.org/blog/crawling-...just-in-or-out

[rajacks]

Thanks for certain,saw decent first page position from article just sort fall off the page and tryin to uderstand why.then I noticed the articles description show up on quite a few other sites.

[FredJones]

In general I agree with you to a good extent although not to a 100% level. I absolutely agree that nothing like an authority site and that does build up only with time. And not only that, I also have my wonderings and thoughts (with no concrete proof) about extensions on domain names.

However, the exception is that Google does think well of exact match domain names with relevant content. In other words, exactkeyword.tld will rank well quickly enough, and can potentially start competing with trusteddomain.tld/keyword.ext relatively early in its life span. However, notverytrusteddomain.tld/exactkeyword.ext and notverytrusteddomain.tld/keyword.ext will definitely take longer. Even closetoexactkeyword.tld will take time to earn the ranks.

[samual james]

thanks chris tew to tell specifically what exactly crawlers do

[bgmacaw]

There is no sandbox...

YouTube - Broadcast Yourself.

[jenhuei]

Originally Posted by bgmacaw

There is no sandbox...

YouTube - Broadcast Yourself.

rofl!! so only ourselves? :P

[jenhuei]

Originally Posted by FredJones

In general I agree with you to a good extent although not to a 100% level. I absolutely agree that nothing like an authority site and that does build up only with time. And not only that, I also have my wonderings and thoughts (with no concrete proof) about extensions on domain names.

However, the exception is that Google does think well of exact match domain names with relevant content. In other words, exactkeyword.tld will rank well quickly enough, and can potentially start competing with trusteddomain.tld/keyword.ext relatively early in its life span. However, notverytrusteddomain.tld/exactkeyword.ext and notverytrusteddomain.tld/keyword.ext will definitely take longer. Even closetoexactkeyword.tld will take time to earn the ranks.

I support FredJones's statement. I have seen some brand new websites with exact match domain ranked in first page, and it has only 1 page of content on it...

[dburk]

Hi ChrisTew,

While some aspects of your assertions are partially true, I have to say that you have missed the mark just a bit.

First let me say that there is no sandbox, it's just a myth. This has been confirmed by Google engineer Matt Cutts. However there are factors within Google's standard search algorithm that may "feel" like a sandbox and thus the term "Sandbox Effect" was adopted to describe this effect.

While you are partially correct that age is a factor, it's not the age of your "website" that is important, it is the age of your individual web page URL within Google's index. Old websites have their pages ranked with the exact same algorithm as a brand new website's pages. So a new page on an old website is treated the same as a new page on a new website. The reason it may seem that they are treated differently is due to the internal linking that is typically done on new page when it is added to an old website. If you started a brand new website and linked to it from an established website just as you would do with internal links you would see the same result.


Google doesn't hate new websites, I would say they love new websites and put forth a great deal of effort to find and index them. Google also values trust and trust is something that must be earned over time and only pages that have stood the test of time will have earned the highest level of trust.

Older pages will generally have earned more trust and therefore higher rankings, but this doesn't mean Google hates new content, or new websites. In fact, Google has a special feature built into their standard algorithm that boosts the ranking of new content. This feature is known as QDF (Query Deserves Freshness) and is partially responsible for the feeling that new webmasters perceive as the "Sandbox Effect".

It seems that new or inexperienced webmasters often convince themselves that they have actually earned the higher rankings that they receive from the temporary boost of QDF. After this temporary boost wears off and their new page sinks to it's truly earned rankings they often imagine that they have been penalized by the mythical "Google Sandbox". It's just their imagination running wild due in part to not knowing about this QDF freshness factor.

I believe where you really miss the mark is in your assertion that search engines rank websites, they don't. Sites that list websites are known as "Website Directories" and are different in nature to a search engine. The primary difference in a search engine and a website directory is what is indexed. A website directory indexes websites and a search engine indexes individual web documents. When you don't understand this fundamental difference it leads to incorrect conclusions about SEO test data and some really wacky SEO theories.

Search engines don't assign PR, trust, relevance nor authority to websites, they assign those values to individual web pages. A new page can be ranked on a new website just as easily as a new page on an established website, if you understand how search engines really work and take advantage of that knowledge.

[Chris Munch]

Originally Posted by FredJones

However, the exception is that Google does think well of exact match domain names with relevant content. In other words, exactkeyword.tld will rank well quickly enough, and can potentially start competing with trusteddomain.tld/keyword.ext relatively early in its life span. However, notverytrusteddomain.tld/exactkeyword.ext and notverytrusteddomain.tld/keyword.ext will definitely take longer. Even closetoexactkeyword.tld will take time to earn the ranks.

Yes I completely agree, I did say so in my original post 'Google HATES new sites, with the exception of the homepage of keyword domains'.

Thanks for expanding on that.

Be aware that this almost instant keyword domain love does only apply to the homepage, and not the inner pages.

[Chris Munch]

Originally Posted by dburk

First let me say that there is no sandbox, it's just a myth. This has been confirmed by Google engineer Matt Cutts. However there are factors within Google's standard search algorithm that may "feel" like a sandbox and thus the term "Sandbox Effect" was adopted to describe this effect.

I agree, but the "sandbox effect" is a pretty clear and apparent in new sites, but it does lead to endless confusion as it is not something a site falls into, it is just based on degrees of trust, and a new site starts at the bottom.

While you are partially correct that age is a factor, it's not the age of your "website" that is important, it is the age of your individual web page URL within Google's index. Old websites have their pages ranked with the exact same algorithm as a brand new website's pages. So a new page on an old website is treated the same as a new page on a new website. The reason it may seem that they are treated differently is due to the internal linking that is typically done on new page when it is added to an old website. If you started a brand new website and linked to it from an established website just as you would do with internal links you would see the same result.

Kind of, but aged sites out perform new sites over and over again.

While it might not be that Google hates a new site per se, it could be that the age of backlinks, age of pages in the index, increased indexation that happens over time, etc. all play a factor. However, the end result is the same: new sites will struggle until they build up trust.

One thing is certain is that age matters (especially in the first year). If you add an internal page to two sites (one 2 months old, and one 2 years old) of the same standard in terms of backlinks and content, the content on the older site will rank.

Google also values trust and trust is something that must be earned over time and only pages that have stood the test of time will have earned the highest level of trust.

Exactly my point, and the big reason why new sites struggle.

Older pages will generally have earned more trust and therefore higher rankings, but this doesn't mean Google hates new content, or new websites. In fact, Google has a special feature built into their standard algorithm that boosts the ranking of new content. This feature is known as QDF (Query Deserves Freshness) and is partially responsible for the feeling that new webmasters perceive as the "Sandbox Effect".

Yes this is very true, the sandbox effect should not be confused with a drop of rankings after a few days as Google will normally rank you higher and drop you

It seems that new or inexperienced webmasters often convince themselves that they have actually earned the higher rankings that they receive from the temporary boost of QDF. After this temporary boost wears off and their new page sinks to it's truly earned rankings they often imagine that they have been penalized by the mythical "Google Sandbox". It's just their imagination running wild due in part to not knowing about this QDF freshness factor.

Yes this is very true. Also Google is generally just pretty sketchy with new domains and pages, it takes it a little while to settle on where it should rank and will move it up or down accordingly.

Also small sites with few backlinks (often a trait of new sites) can only support so many indexed pages without some of them falling into the supplemental (look up indexation cap). So what happens is they might have 30 pages doing great in Google, but then they add 100 more posts, another 200 tag and category pages, and Google drops a huge chunk of their site into the supplemental, including pages that were previously ranking well, and Google traffic drops considerably. Instead Google keeps unoptimized tag pag, archive and category pages in the main index, and these pages don't rank.

Google usually eventually figures out the right pages to rank, although if the site structure is set up poorly sometimes it won't. Either it feels like a sandbox effect but is something different.

Search engines don't assign PR, trust, relevance nor authority to websites, they assign those values to individual web pages. A new page can be ranked on a new website just as easily as a new page on an established website, if you understand how search engines really work and take advantage of that knowledge.

I get what you are saying, and maybe we agree on the same thing but describe it differently. I agree that Google does largely rank individual pages based on that pages properties itself. However...

The end result is that an authority website has internal pages that rank higher. You can orphan a page on an authority site, and do the same on a non authority site. So there are no links to those pages at all.

Google will typically still manage to index the page without any links to it (just get a few people to visit both pages with the Google Toolbar on) - that's more likely to happen on the authority site.

Or you can just post one single link to both those pages on the same page of a third party site.

The orphaned page on the authority domain will always outrank the orphaned page on the non-authority domain.

Try it and see the results, as it is a strong indication that Google values what domain a page is on, and doesn't just rank each page based on its individual traits - the domain it is afixed too matters.

[wickedt2]

How do you get out of the sandbox? I have a site that's 10 months old. I built it to PR3 and #2 for one keyword #4 for another. Today my site has dropped to #116 & #119 for both keywords & my site has no PR "n/a". How do I get out of this sandbox?

[Chris Munch]

Originally Posted by wickedt2

How do you get out of the sandbox? I have a site that's 10 months old. I built it to PR3 and #2 for one keyword #4 for another. Today my site has dropped to #116 & #119 for both keywords & my site has no PR "n/a". How do I get out of this sandbox?

Doesn't sound like the typical sandbox phenomenon. Instead that sounds like you built too many low quality links and Google has ignored them, or hit those pages with an over-optimization penalty (tends to happen if you don't vary anchor text and lack high quality links).

[dburk]

Originally Posted by ChrisTew

I agree, but the "sandbox effect" is a pretty clear and apparent in new sites, but it does lead to endless confusion as it is not something a site falls into, it is just based on degrees of trust, and a new site starts at the bottom.

Hi Chris,

To be technically accurate, what people perceive as the "sandbox effect" is not limited to new "sites" but applies equally to all new pages. Since many folks don't understand how search engines work they tend to focus on their homepage rankings as if it is the only page of their website.

Pages earn trust over time and this increased trust can be passed to other pages that they link to. A new page on an old website can be neglected just as easily as a new page on a new website. Likewise, a new page on a new website can be linked to by trusted pages just as easily as a new page on an old site. The fact that folks don't understand this is what leads to so many neglected new pages, especially on new websites created by new webmasters.

Originally Posted by ChrisTew

Kind of, but aged sites out perform new sites over and over again.

Again, see my explanation in my previous paragraph.

Originally Posted by ChrisTew

While it might not be that Google hates a new site per se, it could be that the age of backlinks, age of pages in the index, increased indexation that happens over time, etc. all play a factor. However, the end result is the same: new sites will struggle until they build up trust.

One thing is certain is that age matters (especially in the first year). If you add an internal page to two sites (one 2 months old, and one 2 years old) of the same standard in terms of backlinks and content, the content on the older site will rank.

This is not entirely accurate. Google doesn't index and rank websites, they index and rank individual pages. If you place the exact same links to both pages they will have equal trust and PR. On-page factors will be the only factors leading to higher rankings.

The fact that people seldom link trusted pages to new pages on new websites is why you are seeing this pattern. It seems you are confusing cause with effect.

Originally Posted by ChrisTew

Yes this is very true, the sandbox effect should not be confused with a drop of rankings after a few days as Google will normally rank you higher and drop you

Actually, I believe you are confusing the so called "sandbox effect" with TrustRank. The 'effect' is a perception of being penalized, when in truth you have not been penalized. You have been treated equal and simply do not understand what has happened.

Originally Posted by ChrisTew

Also small sites with few backlinks (often a trait of new sites) can only support so many indexed pages without some of them falling into the supplemental (look up indexation cap). So what happens is they might have 30 pages doing great in Google, but then they add 100 more posts, another 200 tag and category pages, and Google drops a huge chunk of their site into the supplemental, including pages that were previously ranking well, and Google traffic drops considerably. Instead Google keeps unoptimized tag pag, archive and category pages in the main index, and these pages don't rank.

Google usually eventually figures out the right pages to rank, although if the site structure is set up poorly sometimes it won't. Either it feels like a sandbox effect but is something different.

Okay, now you seem to be going into issues with poorly optimized internal link structure and the resulting duplicate content problems that arise from those conditions. These are not limited to small or new websites, they are simply common mistakes made by inexperienced webmasters.

Originally Posted by ChrisTew

I get what you are saying, and maybe we agree on the same thing but describe it differently. I agree that Google does largely rank individual pages based on that pages properties itself. However...

The end result is that an authority website has internal pages that rank higher. You can orphan a page on an authority site, and do the same on a non authority site. So there are no links to those pages at all.

Google will typically still manage to index the page without any links to it (just get a few people to visit both pages with the Google Toolbar on) - that's more likely to happen on the authority site.

Or you can just post one single link to both those pages on the same page of a third party site.

The orphaned page on the authority domain will always outrank the orphaned page on the non-authority domain.

Try it and see the results, as it is a strong indication that Google values what domain a page is on, and doesn't just rank each page based on its individual traits - the domain it is afixed too matters.

I must totally disagree with this conclusion. I have never seen evidence that suggest this to be true.

You seem to be on the cusp of understanding something that you could be of great usefulness, but you are hanging on to that "domain authority" myth. Try working under the assumption that Google does in fact index individual documents, not websites or domains, and you will discover what really works instead of laboring under false notions. Google doesn't care what domain or website your page is on only what is on your page and what is on the pages that link to and from your page.

[Chris Munch]

Originally Posted by dburk

I must totally disagree with this conclusion. I have never seen evidence that suggest this to be true.

You seem to be on the cusp of understanding something that you could be of great usefulness, but you are hanging on to that "domain authority" myth. Try working under the assumption that Google does in fact index individual documents, not websites or domains, and you will discover what really works instead of laboring under false notions. Google doesn't care what domain or website your page is on only what is on your page and what is on the pages that link to and from your page.

Google does a mixture of both in my opinion, but only Google knows for sure exactly how its algorithm works.

From your reply it still seems like you agree with pretty much everything I think, maybe just splitting hairs on terminology and how you explain it, but the reality of how it works is the same. The only thing you disagree on is that Google only looks at pages, and ignores what domains they are on. In my opinion and from the evidence I have seen Google takes into account what domain a page is on, and it makes perfect sense, as that can help, along with other factors, to decide how important that page is. Certainly though, that page by itself and its associated metrics (independent of what domain it is on) is still very important.

I suggest doing the test I explained to you and it will give you the answers you need. Its pretty easy to set up such a test if you have an aged authority domain. Then you'll know with more certainty without the need for speculation.

Whether either of us are right does not really matter much to me, as I can't see in anyway how either of us being wrong would make any different to actual strategy. Strategy remains the same for me in either scenario. So I'll leave my comments on it at that.

[dburk]

Originally Posted by ChrisTew

Google does a mixture of both in my opinion, but only Google knows for sure exactly how its algorithm works.

From your reply it still seems like you agree with pretty much everything I think, maybe just splitting hairs on terminology and how you explain it, but the reality of how it works is the same. The only thing you disagree on is that Google only looks at pages, and ignores what domains they are on. In my opinion and from the evidence I have seen Google takes into account what domain a page is on, and it makes perfect sense, as that can help, along with other factors, to decide how important that page is. Certainly though, that page by itself and its associated metrics (independent of what domain it is on) is still very important.

I suggest doing the test I explained to you and it will give you the answers you need. Its pretty easy to set up such a test if you have an aged authority domain. Then you'll know with more certainty without the need for speculation.

Whether either of us are right does not really matter much to me, as I can't see in anyway how either of us being wrong would make any different to actual strategy. Strategy remains the same for me in either scenario. So I'll leave my comments on it at that.

Hi Chris,

Not only have I already tested this, I consistently utilize this knowledge on an everyday basis.

When the notion of domain authority became a buzz word in the industry, I along with many others tested this hypothesis. What I found is that if you isolate the difference in page authority and domain authority you will find that the individual page authority accounts for 100% of measurable ranking power. In other words, models that measure page authority entirely in lieu of domain authority are consistently more valid than models that incorporate domain authority in their model.

Even Rand Fishkin, who seems to be the most vocal proponent of "domain authority" had to back way off of the importance of domain authority to get any of his test models to approach validity. He concluded that page authority is much more important than domain authority. In my opinion he did not go far enough. I believe he stubbornly clings to the domain authority notion because he has invested so much in its' existence that he feels that he needs to retain it to save face. Just my opinion.

[Chris Munch]

Originally Posted by dburk

Hi Chris,

Not only have I already tested this, I consistently utilize this knowledge on an everyday basis.

When the notion of domain authority became a buzz word in the industry, I along with many others tested this hypothesis. What I found is that if you isolate the difference in page authority and domain authority you will find that the individual page authority accounts for 100% of measurable ranking power. In other words, models that measure page authority entirely in lieu of domain authority are consistently more valid than models that incorporate domain authority in their model.

I'm very intruiged! When did you do the test? Was it recently? I found the difference to be more prominent post 2008.

Would love to chat more about this, I sent you a PM.

[gittar1122]

I think there is some penalty for new sites if you go for heavy SEO for them. If your site is and you are going to a lot of linkbuilding efforts for it then there is a threat of temporary penalty to your site. You can notice this penalty in terms of traffic going down day by day. It happened with my own domain last year.

On the other hand if your site is old then even large-scale seo efforts will not hurt your site.

[scott g]

Originally Posted by mark587905

Id have to disagree a little, I have sites indexed that a only a couple of months old.

Indexing and Ranking... I can buy a brand new domain & slap up some static HTML and have it INDEXED in 30 minutes. Ranking, however, is another story LOL!

Ranking does take some time and SEO effort (on- and off-site). But again, this all depends on the competition too.

CHEERS!

[Chris Munch]

Originally Posted by gittar1122

I think there is some penalty for new sites if you go for heavy SEO for them. If your site is and you are going to a lot of linkbuilding efforts for it then there is a threat of temporary penalty to your site. You can notice this penalty in terms of traffic going down day by day. It happened with my own domain last year.

On the other hand if your site is old then even large-scale seo efforts will not hurt your site.

I have noticed an effect, as have some friends, where a site suddenly gets an influx of links, which has not happened to that site before, that Google seems to drop its rankings and penalize it temporarily. This happens even with 100% white hat links (i.e real honest citations from quality sites).

Hasn't happened enough to confirm anything, but certainly I have seen that effect in some form.

[Mike Anthony]

Originally Posted by ChrisTew

n. In my opinion and from the evidence I have seen Google takes into account what domain a page is on, and it makes perfect sense, as that can help, along with other factors, to decide how important that page is. Certainly though, that page by itself and its associated metrics (independent of what domain it is on) is still very important.

Well of course it is. the authority domain thing is NOT a myth. Dburk often brings this up as such but never with any evidence. Of course in the serps google is ranking pages not domains but that does not mean that the algorithm does not include domain authority as ONE of the factors.

[Mike Anthony]

Originally Posted by dburk

Even Rand Fishkin, who seems to be the most vocal proponent of "domain authority" had to back way off of the importance of domain authority to get any of his test models to approach validity. He concluded that page authority is much more important than domain authority.

As you noted not the same issue. If you have evidence that proves your point that there is NO domain authority affect in the algo maybe you can finally put it up? You have never done so.

I'm very intruiged! When did you do the test? Was it recently? I found the difference to be more prominent post 2008.

Don't worry about PM. We would all like to see it. yes - no hiding - I doubt the claim.

[Mike Anthony]

Originally Posted by ChrisTew

I have noticed an effect, as have some friends, where a site suddenly gets an influx of links, which has not happened to that site before, that Google seems to drop its rankings and penalize it temporarily. This happens even with 100% white hat links (i.e real honest citations from quality sites).

Hasn't happened enough to confirm anything, but certainly I have seen that effect in some form.

Semantically speaking there isn't really a sandbox but there is and have been things in the algo that will cause a new site to fluctuate. Some call that a penalty some call it something else. In regular business terms if a site falls out of a position where customers cannot find it for an extended time it is hard for them not to characterize it as a penalty. Its a loss of income. I personally though have not seen fluctuations as much when the links come in at a slower rate.

Oh and I mildly disagree with you on the domain age thing. I have not noticed it extended that long. I have first page rankings on domains under three moths and by the fourth month I have never noticed much of a problem in ranking. May depend on the competition level.

[Fraggler]

I think Don covered the 'Sandbox Effect' as many refer to it for new domains very well. If you support a new domain's pages with quality backlinking you can almost eliminate the usual and expected drop after the page is no longer 'fresh'.

I still believe there are filters and penalities used in the index though. I have had historically strong pages being returned in the 'last pages of returned results' for specific keywords across multiple domains and multiple pages.

It didn't matter if there were only 600 or the full 1000 results returned, my page along with other previously strong pages would show up a couple of pages from the last page of results - every time. I would have other less relevant pages being returned higher in the SERPs and the filtered pages would be returned in expected positions for other terms and phrases.

Filter, penalty, sandbox? Whatever you call it, it does exist. I just don't think it is what new domains experience.

[dburk]

Originally Posted by Mike Anthony

As you noted not the same issue. If you have evidence that proves your point that there is NO domain authority affect in the algo maybe you can finally put it up? You have never done so.

Hi Mike,

Where is your evidence that it actually does exist? You don't have any.

I have searched high and low for any evidence that Google assigns any type of authority at the the domain level and have yet to find any. I have challenged anyone on this and other forums to present credible evidence of it's existence. Yet, no one has ever come forth with any evidence at all.

Are we to conclude since there has never been any evidence of it's existence, nor nonexistence, therefore it exists? To use an analogy, if that were reasonable logic shouldn't we conclude that unicorns and mermaids also exists? If you say "unicorns don't exist" do I get to say "where's your evidence?" Can I ridicule you in a public forum because you are claiming unicorns don't exist without providing proof of their nonexistence?

The very fact that it seems no one has ever found any evidence that suggests the actual existence of "domain authority" in use by search engines is evidence that strongly suggest that it does not in fact exist.

It is the fact that no one has evidence that even suggests that domain authority is ever used by search engines that I conclude, for all practical application it does not exist.

I admit that many folks seem to believe in it without any proof of it's existence. I don't have a problem with folks taking things on faith, but I choose not to look at SEO as some mystical practice that must be taken on faith.

I have studied every available logic model on every search engine that I could find. I have built working models of search engine database schemes and ran large scale datasets through the various architectures. I believe this knowledge has helped me gain some insight on the challenges that large scale search engine must deal with.

When I test an SEO concept I usually start with trying to understand how a search engine would implement a ranking factor and how it might improve search results. The Problem I see with the whole notion of domain authority is that it could only serve to dilute the relevance of search results. So why would a search engine do that? Any possible benefit from scoring authority would work better at the individual page level and everything that search engines have disclosed indicates that is exactly how they do it.

I try to understand how search engines work at a very detailed level based on the published papers of those that have actually built the major engines. There is a great deal that is generally understood about how a search engine works at the programing level. It is not the big secret that so many folks seem to imagine. What is secret is the specific weight applied to each of the known factors. Those weighting rules are constantly tweaked to maintain search result quality.

Google in particular has developed some highly innovative technologies for improving relevance and usefulness of search results. But in all this research I have never seen anything that suggests there would be any benefit to scoring domain authority and I've seen nothing that suggests that Google has ever done this.

[Chris Munch]

Originally Posted by Mike Anthony

Oh and I mildly disagree with you on the domain age thing. I have not noticed it extended that long. I have first page rankings on domains under three moths and by the fourth month I have never noticed much of a problem in ranking. May depend on the competition level.

I've found the effect definitely is more visible at higher competition and also effects internal pages a lot more than a homepage.

Generally, after running many sites, I find that the site as a whole breaks out to higher rankings as it ages and creates more links. The exact cause I can't explain, but the phenomenon seems to exist.

While I can get high rankings on a site relatively quickly, to get all pages on the site ranking competitively without a big push of links to individual pages seems to happen sometime between 6-18 months.

[Chris Munch]

Originally Posted by dburk

When I test an SEO concept I usually start with trying to understand how a search engine would implement a ranking factor and how it might improve search results. The Problem I see with the whole notion of domain authority is that it could only serve to dilute the relevance of search results. So why would a search engine do that? Any possible benefit from scoring authority would work better at the individual page level and everything that search engines have disclosed indicates that is exactly how they do it.

What do you think about these benefits of taking into account domain authroity (as well as page authority)...?

1. Not all pages on the internet attract links, in fact a huge proportion of pages just are not that link worthy. That doesn't mean it is not good content, just people don't link to it whether it be a product listing, instructions on how to do something mundane, or some stats. Whatever it is, a lot of pages go unlinked, but they still contain content that answer a searchers question.

Where on-page external links are lacking, utilizing domain authority can help determine what pages might be more valuable in results.

2. Many sites have poor internal linking. Certainly better internal linking improves rankings, but given that lots of pages of content, especially on very large websites are quite isolated in terms of internal linking (and often external links too). Such pages can still be very useful, and if a certain site has a lot of other reputable content (so higher authority) then it can be deduced that other pages are also more likely to be quality. Therefore utilizing domain authority can help determine what pages might be more valuable in results.

3. Spam sites and other low quality sites that have low domain authority will be kept out of results if domain authority is used to help determine what pages might be more valuable in results.

4. Google's continued preference towards the websites of brands has potential to link in with domain authority.

It would also be good to see something more solid. You mentioned some tests you have done, could you describe them and share the results? Also you mentioned some patents and papers form Google which support your conclusion - can you reference them?

I do like how you approach solving SEO problems and I did send you a PM so if you want to chat further I'd be keen too. If you have some interesting sites there's a good chance I can plug them on my own sites too.