Are SEs really this dumb?

[aterran]

Over on the forum at www.powerlinks.com the admins insist that incoming links are always good, whether they are relevant or not. They state that relevancy is totally unimportant and that it doesn't matter if all your links come from the same domain (theirs), that somehow the SEs will still think they represent the true popularity of the site.

I have a hard time believing that Google, MSN and the rest
can't distinguish between relevant links and arbitrary ones; between 1000 links from 1 domain and 1000 links from 1000 domains.

Are they right? Is pure quantity of IBLs the only measure of link pop they use?

[Jill Whalen]

Unfortunately, they're pretty much right for now, but it won't always be that way.

[seobook]

I think # of unique domains is more accurate than # of links, but I do not think it is always easy for all search engines to determine the relevancy of a link beyond its anchor text (page title & headers & links to the page linking to you & what other pages that page links to may also count in some algorithms, but descriptive anchor text from a variety of URLs should work well in many search algorithms).

An incoming link is a link. It might help it might not for PR. Incoming links do not have to be relevant. Unless you are paying for incoming links you can't control who links to you. If your paying for incoming links they should be relevant or you are wasting your money.

What you have to worry about is who you link to. That you can control.

[Nacho]

Here are a few threads worth reading about linking:

Link Building 101

How Fair is the Link Popularity Algorithm?

Filthy Linking Rich

Namebombing: Linking To Influence Results For A Name

Rotating links....dontcha love them?

Links can now penalise?

The Right way to Count Links?

Oh, and I'm sure Brian and I will never forget how much fun we had with the 10,000 Links or 100 Links... thread.

Hope these help add to the discussion. Saludos!

[I, Brian]

While search engines may count all links, they won't all be valued equally - certainly not by Google, which has a suite of tools for devaluing links.

[mosaicservices]

This is what Google has to say about link...

" Google goes far beyond the number of times a term appears on a page and examines all aspects of the page's content (and the content of the pages linking to it) to determine if it's a good match for your query."

More on this here....Google Page Rank

Incomming links make sense only when they are related to the site they are linkin' to.

There is always conflict of opinion...

[projectphp]

IMHO, it was hands down a victory for the SEO community, with this comment pehaps the shining beason:

Yes, it is true, Search engines can't speak

Seriously now (I promise)

Are they right? Is pure quantity of IBLs the only measure of link pop they use?

The questions are probably better phrased as "Is pure quantity of IBLs the only linking factor Search Engines do use, or will use in the future?"

My guesstimation is that, as time goes by, linking will be an even more important factor in many SERPs, and that this importance will be derived from new fators being added to the straight Link Pop and PageRank of SERPs from yesteryear.

In a perfect world, all links would have to be relevant to be calculated in the ranking algo.... This is NOT a perfect world!

With links all that matters for ranking currently, is the number of links. (not talking about Anchor text just the number of HTML anchor links)

Like Jill says, it will not always be this way. All the SE's R and D Depts. are working on this problem as I type this and maybe one day in the future the SE's will find a way to determine what is or is not a "relevant" link, but for now there is no such thing as a "relevant" link.

Think about this for a second, if the SEs are having such a hard time determining just what page is "relevant" for your search query, how the hell are they going to determine what is or is not a "relevant" link.

Relevant linking is a complete fantasy created by people that don't understand the limitations of the SE's.

BYW, IMO if you don't understand the SEs limitations or how they actually work you should not be taking money for SEO or SEM.

[David Wallace]

Case Study:

We have a former client who instead of taking our advice to build up their external links by acquiring them for good quality and relevant sites went to a local SEO (in their area) who place run of site links on two forums that are technical in nature. The former client's business has nothing to do with computers or technology but rather is in the home improvement filed.

They had up to 3000 plus backlinks all coming from two different sites. They used two different two keyword phrases as anchor text, one at the first forum and the other at the second.

The Results:

Still cannot be found anywhere on Google for those two phrases but at Yahoo they are on first page. Go figure?

So in this case, Google is a bit smarter than Yahoo because it recognizes the abundance of links from only two domains and completely irrelevant in nature and has not rewarded the former client of ours with good positioning.

Yahoo on the other hand has because before they engaged in this link building tactic, you could not find them on the first three pages for those two phrases.

Care to supply URLs so others can verify and study this "Case Study"?

If you don't care to make the URls public, PM me, I would love to verify this, I'll even post my results...

I don't believe this so called "Case Study" for a second. There MUST have been some other factor(s) that were overlooked by BOTH SEO companies that resulted in the pages not ranking.

BTW, David you seem to be talking about sites here, when you are talking about linking you are talking about PAGES linking to other PAGES, not sites linking to other sites.

Just how does one link a site link to another site? I mean I know how to link pages but linking sites, hmmm interesting concept.


<added>
Here is my case study...

This just brought to my attention:

The folks at the Stanford University Daily Online News are using, off topic/non-relevant run of the site links, to promote online casinos and bingo halls and those off topic links are working...

Check it out for yourselves, just take a look at the links at the bottom of the page.
http://daily.stanford.edu/

Wasn’t Stanford University where Google was born and does not Stanford hold several of the Google Patents???

</added>

[DarkMatter]

Originally Posted by lots0

With links all that matters for ranking currently, is the number of links. (not talking about Anchor text just the number of HTML anchor links)

Think about this for a second, if the SEs are having such a hard time determining just what page is "relevant" for your search query, how the hell are they going to determine what is or is not a "relevant" link.

Relevant linking is a complete fantasy created by people that don't understand the limitations of the SE's.

Google factors authority sites in it's rankings, doesn't it? Expert sites are defined in the Hilltop algorithm. If I get a link from an authority site from my topic, isn't that relevant link going to help my rankings? Isn't a link from a page with my targeted keywords in the title and on-page elements a relevant link? A search engine may not be able to detect true relevancy, but it can use these factors as an estimation. Am I wrong?

Google factors authority sites in it's rankings, doesn't it?

I believe so, but just because google sees a site as an authority in a subject area that does not automatically make any link from any page on that site "Relevant" to any page being linked to.

Expert sites are defined in the Hilltop algorithm.

Yes, but I don't believe that google is currently using the Hilltop algo. Of course, that is only my opinion.

[I, Brian]

Originally Posted by David Wallace

Case Study:

The Results:

Still cannot be found anywhere on Google for those two phrases but at Yahoo they are on first page. Go figure?

Originally Posted by Lots0

I don't believe this so called "Case Study" for a second. There MUST have been some other factor(s) that were overlooked by BOTH SEO companies that resulted in the pages not ranking.

I can quite believe it - there's always the issue of Sandboxing with Google. I've seen that quite a lot.

You would have to run the study for a very long time - maybe as long as 6-8 months - before determining whether the links were discounted for irrelevance, or simply sandboxing.

Newer domains especially have issues with sandboxing, so I would finger that as a contending explanation.

I would suggest that sandboxing is not the best move by Google, either. Perhaps that's another subject though.

[David Wallace]

Originally Posted by lots0

Care to supply URLs so others can verify and study this "Case Study"?

If you don't care to make the URls public, PM me, I would love to verify this, I'll even post my results...

I don't believe this so called "Case Study" for a second. There MUST have been some other factor(s) that were overlooked by BOTH SEO companies that resulted in the pages not ranking.

Sorry - no can do. You'll just have to take my word for it. I don't want to make the company public because I'd like to get them back one day so we can help them build links the "right way".

Doesn't matter if you believe it or not. What I posted is fact. Could there be other factors why this particular site won't rank well on Google? Certainly but the fact remains that while they have all those links from two technical forums, they still do not rank well on Google but do on Yahoo.

Originally Posted by lots0

BTW, David you seem to be talking about sites here, when you are talking about linking you are talking about PAGES linking to other PAGES, not sites linking to other sites.

Just how does one link a site link to another site? I mean I know how to link pages but linking sites, hmmm interesting concept.

Thanks for clarifying my mis-communication. Right - pages linking to one another.

[David Wallace]

Originally Posted by I, Brian

I can quite believe it - there's always the issue of Sandboxing with Google. I've seen that quite a lot.

You would have to run the study for a very long time - maybe as long as 6-8 months - before determining whether the links were discounted for irrelevance, or simply sandboxing

Could be an issue with sandboxing. I'll have to keep an eye on it. It has probably been at least 6 months since they acquired all those links. If I can remember, I'll check it again in another 3 months and if something changes, I'll post it here.

[I, Brian]

I also wouldn't put too much stock in using anchor text on just a couple of sites - Google tends not to reward multiple links from the same site as much as links from multiple sites.

Think of linking as like the US presidential electoral college system: it doesn't matter how many vote you get in any single state, it's getting lots of states to vote that way that counts.

Don't think of states - or even websites - but IPs - - - - even IP ranges.

Putting sitewides out is useful, but I would definitely advise a use of sitewides across multiple sites rather than just a couple, and also ensure the anchor text is nicely mixed up to provide a wider description.

Of course, the sandbox also complicates things. Seems that age of site may well be a very clear factor in this - see:
http://www.platinax.co.uk/news/archi...box/index.html

Just my personal take.

EDIT: Also, apologies if I'm being patronising. It's late so I'm trying to make a general point as much as anything.

[Michael Martinez]

Originally Posted by aterran

I have a hard time believing that Google, MSN and the rest
can't distinguish between relevant links and arbitrary ones; between 1000 links from 1 domain and 1000 links from 1000 domains.

Yahoo! and MSN seem to be influenced by straight link-popularity. Google takes relevance into consideration, but they weight their search results on the basis of many factors.

Unless you can get about 100,000 inbound links, inbound links are not going to make or break your site with Google for most searches. You need a few to validate your site with Google, but after that, you need to focus on content.

MSN and Yahoo! are using less sophisticated ranking algorithms.

[Relevancy]

1: Google can check the whois information of a domain. So if the same name is all over your links it is not good.

2: Google does look at the ip address. All the way up to the B block for link fudging.

3: varied anchor text is evaluated in Google

4: page relevant linking is looked at. Meaning if a domain links to a relevant page and not just having all your inbound links link to your home page.

5: Not as important but still important is the links url. They will see if keywords are in the url linking back to you.

These are all very simple things for Google to analyze and they are all relevant.

[Gerardism]

From when I started following some SEO firms rankings probaby since 2001 or so) that had (and still do have top rankings) for their competitive keywords phrases, aside from the usual yahoo and DMOZ directory links, they have their keywords on all their clients pages linking back to them on the bottom of the pages, and a lot of them are not related at all.

It has been working for a long time, and it seems that it might be a while before it changes. These sites do have a vast number of clients and their sites linking back to them.