Marketing & SEO Discussion List - LED Digest

Home arrow Full Issues arrow 2005 archives arrow LED Digest 2003: Special Issue - Linking
LED Digest 2003: Special Issue - Linking Print E-mail
==================================================
                 The LED Digest
             Moderated Discussion List
     "Effective Online Advertising, Since 1997"

         pair Networks: The LED's Web Host
   Hosting and Domain Reg. from a Trusted Leader
  pair.com for Hosting  |  pairNIC.com for Domains

==================================================
List Moderator:                     Published by:
Adam Audette                          LED Digest
post, led-digest.com     http://www.led-digest.com
..............................................
July 29, 2005                          Issue #2003
..............................................


            .....IN THIS DIGEST.....


==== SPECIAL ISSUE =================

        --== The Linking Debate ==--

                ~ Michael Martinez
"...where is the proof that reciprocal linking IS
necessary to get to the top."

                ~ Tom Aman
"...the sophistication of the searcher will also
have a big effect..."

                ~ Rich Dudley
"...I'd like to offer a third opinion - that of a
business owner."

                ~ Steve Pronger
"...let's hear a Google engineer say..."


===== CONTINUING =================================

[Please note that past quotes by the author (or "self-quotes") are
cited with double tick marks, like this:

>> I was talking to a blue
>> dog and then it bit me

All other quoting is standard. Thanks- adam]

-----------------------

From: Michael Martinez
Subject: Linking

> I have never stated that reciprocal linking
> is necessary to "get to the top".
        - Dirk Johnson, LED 2000

I never said you did.  However, many people do, whether here or in
the Web-based forums, and that's not the sort of thing that requires
substantiation through citation because it's just too common a
position.  If people cannot find examples of this point of view,
it's because they aren't looking at the various discussions.

> ... we stress that reciprocal linking is an opportunity
> to establish the brand, and should be approached as
> a branding function, and not an SEO function. The fact
> that linking is used in search results is certainly important,
> but search results are actually derived from the branding effort...

Most search results are derived from on-page content and nothing
more.  80% of all searches are non-commercial:
http://snipurl.com/gkt8  [searchguild.com]. Branding is not relevant
to the majority of Web content.  It can HELP, and can give you an
advantage over non-branded content, but it usually doesn't come into
play at all.

> It hasn't been for a very long time because Google has long
> since adjusted its algorithms to compensate for the practice.

> ... where is the proof of this statement [that Google
> has long since adjusted its algorithms to compensate
> for the practice of linking]...?

A more appropriate question is, where is the proof that reciprocal
linking IS necessary to get to the top.  Google says nothing like
that:

http://www.google.com/webmasters/guidelines.html
http://www.google.com/webmasters/facts.html
http://www.google.com/webmasters/seo.html
http://www.google.com/webmasters/4.html .

Look at the very first entry on that last page.  "How Google ranks
pages". They start out by saying, "Google's order of results is
automatically determined by more than 100 factors, including our
PageRank algorithm".  They conclude that paragraph by saying, "Due
to the nature of our business and our interest in protecting the
integrity of our search results, we limit the information we make
available to the public about our ranking system."

So, the best answer comes from Google: they use more than 100
factors to determine ranking, and link popularity is only one of
them.

I don't sell reciprocal linking services, so I am under no
obligation to make them sound necessary or to justify the practice.
It's outdated because there are many, many sites which come up first
in search results WITHOUT resorting to reciprocal linkage.

Anyone who wants proof need only pick out random searches and look
at the results.  People search for information on health matters,
entertainment matters, news, technology, family history, general
history, celebrities, and other popular topics all day long.  The
top results usually do NOT have reciprocal links.  Those sites don't
pursue reciprocal link programs.

> Links matter, and it is the unique domain
> count that seems to matter most.

Links matter, but reciprocal links don't matter that much, and most
inbound links are actually INTERNAL (from the same domain).  Natural
linkage patterns emphasize the importance of popular or key pages on
large content Web sites.

> Michael, there are number of very logical and legitimate arguments
> as to why Google would not want to nullify reciprocal linking...

Dirk, you've injected this "nullify reciprocal linking" phrase into
the discussion.  I never said anything like that, so I'm under no
obligation to say anything further on the matter.  I find that
people resort to straw man arguments like this when they cannot
refute the main point.

Google advises people to look at more than links.  They are the best
authority on what works for Google regardless of how vague they
choose to be.

>> So, we'll continue to disagree because
>> Google has left you guys in the dust.

> Michael, many of my long term, most established clients
> would smile all the way to the bank at that statement.

I am sure they would. After all, they are happy customers, and they
don't engage in active research in this area.  I do.

>> It becomes more and more difficult to achieve top rankings
>> through link campaigns for two reasons: as specific categories
>> become more hyperoptimzed, they require more linkage...

> This statement needs some careful analysis.

No, it's a straight-forward representation of the facts.  "Careful
analysis", in Web-speak, means "Let's reword it into a straw man
argument and knock that down."  That dog won't hunt.

I explained "hyperoptimization" in detail in my "Changes in Google
Ranking Strategies" paper earlier this http://snipurl.com/gkte
[xenite.org]. It's too lengthy to cite here.

> Secondly, I am not sure what "over optimizing" is.

Then why are you disputing what I say with respect to
hyperoptimization?

> Would you please expand upon your definition of "good"?

In my opinion, a "good" link is natural, not achieved through link
exchanges or spam pages, and comes from a site that links to many
different sites without regard for whether those sites link back.
The most commonly cited "good" link providers are those from popular
directories like Yahoo!, DMOZ, Zeal, JoeAnt, et. al.

News sites, popular entertainment sites, online magazines, and
similar large content providers also provide "good" links.

The common factor to these kinds of links is that they are not
purchased or exchanged.  They are natural.

Most commercial site can create their own "good" links simply by
linking to other sites without regard for whether those sites link
back.  In fact, I find many commercial sites do just that.

I understand that many commercial site masters want the inbound
links.  But I research between 80 and 100 companies online every
week.  The vast majority of their Web sites have very few inbound
links.  Most of the sites are, in my opinion, poorly designed.  But
they are rarely optimized for search and usually don't need to be.

"Professional" link builders always make extravagant cases for why
their services are necessary.  The bottom line is that such services
are only necessary when all else fails.

Most commercial Web site builders really only need to focus on one
thing: Making your site findable by typing in the company name.  The
number 1 obstacle to my daily research is that I often have to
figure out how to find an official company Web site because they
didn't bother to their company name in the title tag or in an H1
header tag.  Most of them don't even use the H1 header tags.

I often find the official sites through links in specialized
business directories whose names are so obscure most of you have
probably never heard of them.  I seriously doubt all those links
were inserted by hand.  These directories are clearly automated
"yellow pages" sites.  They are not the "spamad" sites I refer to
occasionally.  They are laden with ads, and I have no doubt the
business listings are provided as a means of getting people to look
at the ads, but they are nonetheless useful resources for people
like me who have to find out information about companies on the
Internet.

In my perfect world, every business Web site would provide a few
simple pages that are linked to by all other pages on their sites:

1) Main page with company name, address, and telephone number.

2) "About Us" page with a concise, 1-2 paragraph explaining what the
company actually does (I can live without the family history, the
statements of pride, the generalizations about how important in the
industry a company thinks it is, etc.).

3) A Contact Us page with names, titles, and brief descriptions of
job functions (the contact forms are fine, but generic,
one-serves-all forms don't help).

4) A locations page listing the corporate headquarters first (even
if you have thousands of locations that are listed in a database).

5) A clear statement about whether the company is privately owned,
publicly traded, a partnership, a subsidiary of another company
(with a link to a mama-company site), etc.

Some sites have all this information. I love them.  Most sites
don't.  Oh well.  I don't live in a perfect world.

And that's a fact, too.

Michael Martinez
http://www.michael-martinez.com/


-------- next post --------

From: Tom Aman
Subject: Linking and other SEO methods

With all the discussions on SEO methods, seems to me that the best
method is the one that seems to work for you.  I believe that,
unless somebody like Google actually publish details of their
methods, any particular method, be it link exchanges, keyword
optimizations, page content optimizations, etc., etc. are really
only guesses.  From the discussions here it sounds like different
methods have worked well for different people.

I suspect that Google, as well as the other major engines, are
forever tweaking, testing and refining their spidering and retrieval
methods, always trying to find the methods that return the best and
most relative results for most searches.  What this means in SEO
terms is that what worked well today may not work well tomorrow and
what works well for one engine may not work well with another.  The
only constant is that the "best" method will probably change over
time.  What it all comes down to is that if your method is working
well for you, stick with it.  If it does not work well, try another
approach.

One thing that is suggested from time to time is that one should
examine log files to see what keywords have been used to lead a
searcher to your page, then optimize for those keywords.  This
always puzzles me a bit.  Using those keywords, the searcher has
found your page and presumably others will, too.  But this tells you
nothing about the keywords, potentially important ones, that a
searcher used that DID NOT find your page.  And then there are the
keyword variations, and I suspect this is an area where a really
good SEO will do wonders.

For example, a search on "bird houses" returns 3,500,000 results,
"birdhouse" returns 730,000 results, "birdhouses" returns 1,220,000
results, but "bird house" returns 12,200,000 results (seems to
include pages with "bird houses", "birdhouse" and "birdhouses" as
well as pages with "bird", "birds", "house" and "houses").  Would
you have expected one of the variations to pick up all of these
others in the results?

One thing that should always be remembered is that, aside from the
search engine and its indexing and its methods of determining
relevance, results are also very much affected by the searcher and
how that person approaches searching.

For example, in doing genealogy research, I did searches for an
ancestor named "Adam Aman".  If this term is entered exactly like
that in Google, the results will include such things as "A man
called Adam...". Google returns "a man" as a match for "Aman" for a
total of about 2,680,000 items.  But by entering the search as
"+Adam +Aman", the results now include only those pages that contain
"Adam" and "Aman", for a result of about 173,000 items.

By doing a little more refining (meaning by adding some more search
terms using + or - to include or exclude pages containing
specifics), this search can be reduced to the REALLY RELEVANT items
that are of real interest.

The point of all this: while rankings for specific keywords may be a
critical SEO objective, the sophistication of the searcher will also
have a big effect so you need to understand your potential audience.

To understand how complex this whole subject has become, it is worth
taking a look at "Publications related to Google technology" at
http://labs.google.com/papers/index.html.  To really start to get
some idea of the complexity of this whole area, take a look at the
item "Who Links to Whom: Mining Linkage between Web Sites".

Also, check out the link at the bottom of the page "More papers by
Googlers" that leads to http://labs.google.com/papers.html.  The
link under the section "information retrieval" titled "The Anatomy
of a large-scale hypertextual Web search engine" is one of many
papers linked here that give some idea of just how involved the
entire search engine area has become.

Tom Aman

Aman Software
http://www.cyberspyder.com


-------- next post --------

From: Richard Dudley
Subject: Reciprocal Linking

I've been following Dirk and Michael's exchanges on reciproal
linking, and I'd like to offer a third opinion -- that of a business
owner.  I'm not a professional SEO. From this point of view, let me
chime in with my feelings:

1) In real life, there are people who recommend a good amount of
business to us, but we refer little business back to them.  It's not
a mutually reciprocal agreement.  We're a florist, and we refer
brides to DJs, photographers, etc. based on what we think of that
person's work.  If we refer a bride to a bad photographer or a DJ
devoid of personality, that's a poor reflection on us.  We can't
have that, even if that unpersonalbe DJ sends brides our way.  As
such, our referrals carry a significant amount of weight (high page
rank you want to think of it that way).

2) We feel strongly that whom we link to is as much a personal
referral as if given by voice, face-to-face.  We don't participate
in any reciprocal linking agreements.  In fact, we don't even have a
links page on any of our sites.  I do, from time to time, mention
and link to other vendors in our blog (www.bloomeryweddings.com/blog),
but these are unsolicited.  The mention is made because it fits
the context of the blog post.

3) I would laugh at you if you suggested me setting up a links page
in order to get on someone else's links page.  I don't care what
minute help that might be in the SERPs, we strive to distinguish
ourselves from the rest of the pack in the real world, and we don't
want to be part of the pack in the online world.  A links page is
merely the pack.

4) What I want is unsolicited links to our sites from blogs, forum
posts, informational web pages, etc. because we have products,
designs or content that someone has purchased, found useful or
interesting.  It's these links -- which are essentially personal
referrals -- that we feel actually generate business.  Being #1 in
the SERPs isn't as important as selling a lot of products.  Being
high up in the SERPs is helpful, no doubt, but personal referrals
are better.  I did a small analysis of our visitor and purchase
stats last year, and while search engines accounted for the majority
of visitors, customers who followed an unsolicied link from a
forum/blog/etc purchased with a much greater frequency.  While I may
not be the most visited site by focusing solely on organic links, I
don't care.  It's about sales and the bottom line, and frankly,
chasing SERPs doesn't have the payoff for us.

5) I wouldn't pay one nickel to anyone whose service consists of "I
found these sites--if you link to them, they'll link to you" (OK,
yeah, I know that's what this list's original mission was, but...).
I would, however, consider paying for a content advisor, who
searches forums, message boards, WordTracker, etc. to find the hot
topics, then advises us on what content we could add to the site to
generate these unsolicited links.

Rich Dudley
www.bloomeryweddings.com


-------- next post --------

From: Steve Pronger
Subject: Linking

> Just like Mr. Martinez, we too have been able to help clients
> get qualified search engine traffic with 20-40 high quality,
> non-reciprocal links... if someone won't link to my or a client's
> site without reciprocation, then that person must not have thought
> too highly of the site's or page's content. If reciprocation is
> required, the link is not an objective "vote."
        - Shari Thurow, LED 2001

Fair enough Shari, but I'm wondering how you obtain those 20-40 high
quality, non-recip links. Directory listings? Most of them charge
these days (unless you want to wait forever for a free listing) so
is it really a "vote" if you paid to be there? Paid text links? Same
story. Blog comments? Nope. Article submissions? Better, but really
the main reason webmasters use your article is to add content to
their site. The link to you is payment for use of your article, not
to vote for your site.

These are all one-way links and quite legitimate in my view if done
responsibly, but don't really constitute a vote. How about links
from sites you own or are within your circle of influence? No, the
only way a link could be considered a vote is when a site links to
you for no other reason than they wanted to.

But how do you "obtain" such links for a client? Contact 20-40
webmasters and say "Here is a great website. Please link to it.
Please!". Sure, if the site is an "authority" in a particular niche
(as is Michael's) you may have some success, and natural,
unrequested, one way links will occur. But if the site is about car
hire is anyone really going to link to it because it's an authority?
OK, a travel site might link to it, but 9 times out of 10 they will
want a return link. And why not? The exchange adds value to both
sites. But, that makes it a "reciprocal" link. And according to
Michael, Google have long discounted the value of that link even
though it benefits human visitors to both sites.

> I have heard multiple Google software engineers
> publicly state not to link to "bad neighborhoods".

OK, let's take that as a given. I have no argument that many people
go about recip linking the wrong way, and that linking to a "bad
neighborhood" can hurt you. I'm sure Dirk Johnson would agree. But
let's hear a Google engineer say "a link from a site that you link
to is nullified for no other reason than reciprocation". I'll
concede.

Steve Pronger
http://www.stevepronger.com


-------------------------------------------------------
The LED Digest is sponsored by pair Networks:
pair.com for Hosting | pairNIC.com for Domains

Copyright 1995-2005 Orange Wheel, LLC. All Rights Reserved.
-----------------------------------------------------------------

"Mathematics takes us into the region of absolute necessity, to
which not only the actual word, but every possible word, must
conform." - Bertrand Russell