Every now and then an event occurs that changes how the SEO community views the way websites are optimized and structures promotions. The purchase of the rights to the Orion Algorithm by Google and equally important, the interest that both Yahoo! and MSN took in the algorithm as they vied for ownership themselves, marks just such an event.
Bill Gates said to Forbes magazine about Orion:
“That we need to take the search way beyond how people think of it today. We believe that Orion will do that.”
What Is The Orion Algorithm?
There is much confusion about the Orion algorithm and much secrecy around the specifics. Here’s is the “What’s Been Said” and “What It Means” breakdown:
What’s Been Said: Ori Allon, the developer of this technology described Orion in this way:
“The results to the query are displayed immediately in the form of expanded text extracts, giving you the relevant information without having to go to the Web site–although you still have that option if you wish.”
He cited an example of the keyword phrase “American Revolution.” The search would not only provide extracts with the phrase, but also additional information on topics such as American history, George Washington and the Declaration of Independence.*
* CNET News, April 10, 2006
What It Means: Most on the web take this to mean the results from Google will be displayed similar to those at Ask.com where you will be able to get a sample of the site and some of it’s quality content without having to visit the actual site. The part that most caught my attention however is where he cited the example and noted the additional phrases that would be considered and the impact having this technology will have on the way queries are dealt with.
From this standpoint, the Orion Algorithm, in its essence, is a whole new way to score the value of websites that appear on the Internet. Rather than determining the value of a website based on the specific query being entered into the search box, Orion may dig deeper and query related phrases as well. Now, this may not be an entirely new concept, directories have been providing a “Related Categories” option for ages however the addition of this function to standard search engines and what this may well mean for the methods required to rank sites on them is extremely significant.
What Is Relevant?
One of the main hurdles that SEO’s will face in reaction to this new function is determining exactly how the additional relevant phrases are determined. There are a few possible sources the come to mind:
Directories (least likely) – The directories are already using “Related Categories”. It is possible that the engines will choose the simplest possible means of determining relevancy and opt to use sub-categories of a directory listing and to use the “Related Categories” as the supplemental keyword sources.
Alternatively they could simply run the search itself on their directories and reference the categories that come up and run supplemental searches for those categories.
The main drawback to this approach is that many popular keywords would not be cross-reference accurately. For example, a search for “seo” would result in a supplemental search set of “promotion”, “web design and development”, Internet marketing” along with a variety of other phrases. While these phrases are related by industry a visitor searching for “seo” may well not be interested in “web design and development”.
Thesaurus (unlikely) – It may be that the engines choose to reference a thesaurus for related phrases however this doesn’t work for many keyword phrases. Single word phrases would be doable however multiple keyword phrases would be far more difficult and acronyms (such as “seo”) would find no related words in the more common thesauruses.
Search Behavior (highly likely) – The most likely source of the relevancy data is also the most difficult to predict and this is search behavior patterns. While I have had some disagreements with members on a couple SEO forums over whether the search engines can in fact know your search patterns, the conclusion is that they indeed can under many circumstances. Search engines will be able to compile enough data based on the users they are documenting to assess overall search behavior (and here you thought all those great tools the engines come out with were just them spending their money altruistically).
If Google “knows” that after someone enters “seo” as a query they follow that up with “seo service”, this is likely to then be used as a supplemental search. Similarly, if they also know that these same searchers tend to also search shortly before or after for another common phrase, say “w3c compliance” then this too is likely to be used as a supplemental search.
Agree To Disagree: Implementation
Now that we have a better idea of what the Orion Algorithm is and how it works the big question is, what will it’s implementation mean to search engine users and to how websites get ranked on those engines. At this time there appears to be two main schools of thought:
What I believe, and
Everything else that’s been published
I’ll be the first to admit that my interpretation of how the Orion algorithm will affect search engine results is either not shared by other SEO’s (at least those who have a published opinion on the topic) or has not been thought up by them. That said, my take on the Orion Algorithm did not initially include their predicted affect whereas I now believe that it is likely both implementations will be tested if not brought into full effect within the next 12-18 months (this may seem like a long time but if you want to develop a strategy to react to it this is about the lead-time you may well need). So â€¦ what are these two possible outcomes?
Where we all agree: the addition of key parts of web content in the results. This is how the algorithm is explained to function by its developer and is thus the obvious conclusion to most in regards to how it will be implemented.
Everyone else: related information displayed separately. From what I have read, the majority of people believe that the related phrases will be displayed separate from the original query (though rightfully no one seems to be predicting exactly where or how). Essentially this will give searchers the ability to view information on related topics quickly and easily.
This is certain to be included in some capacity and we have already seen similar functions added to the Google results for specific queries though not to any capacity reliable enough to be launched across all Google search results.
And then there’s my opinion: integration in standard search results. To me it seems short-sighted to believe that Google will leave a technology that allows them to drawn information and relevancy on multiple related phrases to just displaying multiple options on a results page. With the processing power they have at their disposal why would they not reference a site against its ability to rank for these other phrases and base the final results on that? Let’s a take a quick peek at the pros and cons of such a move:
Cons first: Processing power. That about covers the downside and I’m sure we’re all aware of the fact that if this ever becomes an issue they have more than enough capital and technical know-how to get around it.
Pros: Imagine a world where running a search for a query took into consideration whether a site ranked for multiple related phrases. What do you suppose the impact on the results would be if only those sites that had content related to a number of areas of a topic ranked highly? The answer: a much more relevant set of results.
Fortunately, while there may be some disagreement in regards to how this new algorithm will be integrated into the search engine results pages the resulting actions required are the same. Whether the new functions will be added in the form on additional links and information on the results pages or whether they will be taken into consideration when ranking the site for the initial query, sites that rank well for a multitude of related phrases will fare better than those that rank for just one of the phrases.
The action required then on the part of SEO’s and website owners is to provide quality unique content on all the possible areas that may be considered relevant to the main keyword target. Once this is accomplished then these areas need to be promoted in order to insure that they rank well.
The resulting web will be one that rewards websites with a large amount of quality content on the highest number of topics related to a specific issue. If one considers the end goal of any of the major search engines, to provide the most relevant results possible, this new technology is sure help promote these types of results and insure that the searcher is receiving results that are likely to provide the information they’re looking for.
And let’s also consider this: should you choose to be an “early adopter” and begin making changes to your site, adding new content, optimizing it and getting it ranking well, what will the results be? Even if Orion isn’t implemented for another decade your website will gain stickiness and rank for more related keywords bringing you more targeted traffic and keeping it on your site. Could this possibly be a bad thing?
This atricle was taken from: Beanstalk