Insider Information On Search Engines And Reciprocal Linking
July 12th, 2007Happy May Day, everyone.
Of course, it’s not really May Day, is it? It’s really some other day. July 12, actually. Or at least that’s the day this post was written. You’re obviously reading it on some other day altogether.
For the moment, however, let’s pretend it is May Day.
May 1, 2007.
During April 2007, internet users in the United States initiated — according to the statisticians at Search Engine Watch — the staggering total of 7.3 billion search-engine queries.
From here, the numbers get a bit more fuzzy. Depending on who’s crunching them and how the metrics are massaged, Google “owned” somewhere between 49 and 64 percent of all those 7.3 billion searches. Which leaves, figuring conservatively, about 3 billion search queries that were divided up among the other engines.
(Note: Google’s market share also varies depending on whether you include Google-client AOL’s 5 percent and Netscape’s 1 percent or so. Some web watchers include client searches in Google’s raw total; others tally client’s numbers separately as a way of determining advertising market share.)
Put another way, all the “also ran” search engines get roughly six times as many customers each month as Walmart does in all its thousands of brick-and-mortar stores combined.
Put yet a third way, the SEO industry’s obsession with search-engine optimizing sites solely for Google ignores the reality that sites which return in position three on Yahoo, four on MSN, and five on Google have the potential to be visited by literally billions more people than a site which returns one or two on Google and 11 or 12 on Yahoo and MSN. (A phenomenon unofficially know as “the Zen of Appearing On Return Page One.)
Since this blog is about relevant link exchange for the end user, and not search-engine optimization, let’s focus on what these numbers mean for your reciprocal linking strategy.
Fact: There are hundreds of search engines.
Importance of Fact: Zero.
When it comes to natural results, there are only five services that matter: Google (itself, AOL and Netscape), Yahoo (itself and its wholly owned subsidiary Alta Vista), Microsoft (itself), and Ask (which provides its own returns and those of Lycos through its Teoma subsidiary.)
The rest of those hundreds are specialized research sites that require paid membership (the minority) or scam sites that exist to further various nefarious schemes aimed at people who really believe those spam mails offering to put their sites at the “top of major search engine” rankings in 29 days for only 29 dollars.
Today’s question is: What do we know about the effect of editor-based, ethical reciprocal linking on search returns offered to the three billion searchers a month who don’t query via Google?
With roughly 27 percent of the market (roughly 810 million queries) in April, Yahoo is by far the most important search engine in America after Google. Here’s what Yahoo’s webmaster guidelines have to say about linking (all emphasis in this post are ours):
Pages Yahoo! Wants Included In Its Index
– Original and unique content of genuine value
– Pages designed primarily for humans, with search engine considerations secondary
– Hyperlinks intended to help people find interesting, related content, when applicable
– Metadata (including title and description) that accurately describes the contents of a web page
– Good web design in general
Can’t get much clearer than that, can you? Yahoo WANTS (their word, not ours) sites with good, viable, links. And, though they don’t explicitly say so, it’s reasonable to assume that if they want sites with links, they are not as favorably inclined toward sites without them.
This is not to say that Yahoo automatically ranks sites with robust, relevant link pages above sites without them. Links, good or bad, are just one of a myriad of factors involved in all the engines’ ranking algorithms and there are almost as many exceptions that prove the rule as there non-exceptions. It is, however, fair to say that based on their guidelines Yahoo would rank a site with good, positive links ahead of one without links if everything else about those particular sites was equal.
Moving on down the (pardon the pun) “hit” parade, we come to MSN with a tad over 10 percent, roughly 300 million queries, in the April search engine market share sweepstakes.
If you check out the Microsoft FAQ entitled Microsoft FAQ About Website Ranking you’ll find this: Live Search website ranking is completely automated. The Live Search ranking algorithm analyzes factors such as web page content, the number and quality of websites that link to your pages, and the relevance of your website’s content to keywords.
As does Yahoo, Microsoft makes it clear that quality linking is an important factor in attaining a good ranking. So important, that they list it as one of only three factors singled out for inclusion on their “About Website Ranking” page.
Nipping along to Ask (5 percent, 150 million monthly queries), we find that “Teoma was the first, and is still the only, major search technology based upon the clustering concept of subject-specific popularity: ExpertRank. In fact, Teoma means ‘expert’ in Gaelic.”
If all that is Greek (or Gaelic) to you, you’re not alone, we don’t understand it either.
Fortunately, Ask’s Web Search page is a bit less murky when it comes to the importance of linking to its search algorithm.
It tells us that the Ask crawler “follows hyperlinks from the page” and applies its “ExpertRank algorithm (which) goes beyond mere link popularity (which ranks pages based on the sheer volume of links pointing to a particular page) to determine popularity among pages considered to be experts on the topic of your search.”
Again, it’s a bit hard to translate that from the Gaelic with absolute clarity, but hidden behind all the chest thumping about ExpertRank being a bigger, better stud than Brand G, Y and M’s “mere” algorithms, there is a definite implication that quality, relevant links may be even more important to attaining a good ranking on Ask, than it is on the big three engines.
There are, after all, a myriad of definitions for the word “expert,” but you can look in every dictionary ever published and you won’t find “irrelevant” listed as one of them.
So, that’s that.
Not quite.
There’s still the G-word.
The three-billion-a-month googrilla.
Google has recently been revising its webmaster guidelines and, as of this writing (July 12, remember?) the latest version says:
When your site is ready:
• Have other relevant sites link to yours.
• Submit it to Google at http://www.google.com/addurl.html.
Do not, repeat DO NOT, disregard that word RELEVANT or there’s a good chance a few billion people won’t be finding your site.
There are numerous reasons why positing irrelevant single or bulk-harvested links on a website is a form of Google Roulette.
Further down the Webmaster Guidelines page, for example, it says “In particular, avoid links to web spammers or ‘bad neighborhoods’ on the web, as your own ranking may be affected adversely by those links.”
How does that relate to relevancy? Simple. Unless you’re a web spammer living in a bad neighborhood yourself, how could a link to or from a web spammer possibly be relevant to what you’re offering?
Going to Google’s How can I create a Google-friendly site? page for a final example, we read “Make sure that other sites link to yours (but) … keep in mind that our algorithms can distinguish natural links from unnatural links. Natural links to your site develop as part of the dynamic nature of the web when other sites find your content valuable … ”
As with expert, there are many definitions of “valuable,” but irrelevant isn’t one of them.
So there you really do have it. The truth about search engines and linking as told by the four major engines themselves.
Bottomline is, they all want to see solid, relevant, informative natural links that were acquired one at a time and added to a web page with at least a minimal amount of human interaction and intelligence.
And none of them look kindly on bogus links automatically harvested at random or purchased in bulk from a link farm.
It can also be fairly said that none of the major search engines particularly appreciate crawling over sites with no links at all.
Despite what many outside “experts” may claim, the engineers, mathematicians and philosophers inside the search-engine world are as well aware as we are that links are what make the “dynamic web” dynamic and that a web without links would dry up and die as surely as a world without water.
