SEO Expert, Anthony Yap

Google filed a patent which was made publicly available recently called ‘Information retrieval based on historical data’. While not a leisurely read like a novel would be, it certainly gives a great deal of insight into the future of search engine algorithm technology and is certainly worth studying for the SEO enthusiast.

Aging Delay
I mentioned in another post the issue of the Google Sandbox. This is different from the aging delay in that aging delay is a subset of the sandbox. Thus, the aging delay could be another reason why a site would end up in the sandbox.

While in the Sandbox, your site will not rank highly in Google for any meaningful keyword phrases that might bring it any decent web traffic.

Measuring Traffic Data for Search Engine Rankings
Google can use traffic data obtained from sites in establishing how it should rank. It states in part “…information relating to traffic associated with a document over time may be used to generate (or alter) a score associated with the document.” Being in the top 10 is not enough. A site must prove to be useful in order to remain in it’s position. Google wants to know what its users like and dislike and the only way it can know this for sure is to monitor it. User data is now an important factor in long-term search engine placement since clicks are measured as one of the many factors indicating “votes” for a site. Website usability and user behaviors, therefore, becomes an integral factor in effective SEO. In addition to high link popularity, the more useful and relevant the information which translates to higher clickthroughs and conversions, the higher the rankings. So, how does Google obtain information about your site’s traffic even though they don’t have access to your log files or traffic reports? Well, as part of the Terms of Service agreement you signed when you downloaded the Google Toolbar (which I’m sure you read, right?), they have access to anonymous data from user behaviors while surfing and it includes sites visited and duration of visit. If you are a privacy freak, don’t worry because this is not like the malicious spyware that sends personal data without your knowledge. They take the aggregate data received and then use it to perfect their search engine ranking algorithm. Highly traffic sites means these are popular sites, and Google wants to ensure that their searchers find these popular sites easily because they are most relevant to the search query.

Using Click-Through Rates in SERPs to Establish Search Engine Rankings
Google states, “…may monitor the number of times that a document is selected from a set of search results and/or the amount of time one or more users spend accessing the document. [Google] may then score the document based, at least in part, on this information.” Google tracks the URLs in their SERPs and use it to determine which pages are getting clicked on most for specific keyword phrases (much like Google’s Adwords program does). They also determine if the searcher was satisfied with the page they clicked on by keeping track of user beahviors after the site visit – i.e. Did the user hit the ‘Back’ button and click on another result in the SERPs?

Tracking Growth Patterns of Inbound Links to Weed Out Spam Sites
Another interesting factor which might be complex to implement is the tracking of inbound links as stated here: “A sudden growth in the number of apparently independent peers, incoming and/or outgoing, with a large number of links to individual documents may indicate a potentially synthetic web graph, which is an indicator of an attempt to spam. This indication may be strengthened if the growth corresponds to anchor text that is unusually coherent or discordant. This information can be used to demote the impact of such links, when used with a link-based scoring technique, either as a binary decision item (e.g., demote the score by a fixed amount) or a multiplicative factor.” This would mean that as webmasters and SEOs, it would be a good idea to have natural looking growth patterns over time rather than sudden spikes. You also want to be careful of unusually high number of inbound links compared to your competition’s average because this sets off more alarm bells that could flag it for appearing spammy. Natural-looking link text is a must to rank on Google. Therefore, all inbound links pointing to a site saying “mortgage” only is not a good sign and will likely get penalized. You’ll want to refer to my article on natural link building here for tips on link text variations and strategies.

All this leads back to Google’s main purpose which is to provide the *most* relevant results to searchers – something that the Google engineers have been emphasizing more of since last year. With all the spammy Adsense and affiliate sites going around these days, there has to be some way to combat them and rank the good sites. This patent may spell bad news for those well-meaning new site owners as it would take a much longer time (possibly years) to obtain decent rankings on Google especially if they are starting a site in competitive niches. Even with a well-designed and SEO’d site that is high quality and contains lots of useful information, it may not achieve any rankings. If you’re considering setting up a site, getting started now instead of later so that the site will have sufficient time to age in the search engines would be highly recommended.

Any search engine optimization campaign involves link building as a major element in attaining top search rankings.

To rank well long term, a SEO must seek out the highest quality links possible and avoid links from bad neighborhoods, low quality sites, and employing other questionable SEO techniques. In the search for links, how does choosing what sites to obtain links from affect rankings?

All links are not created equal. There are regular links and then there are authority links. Regular links are those from average sites that don’t rank well and have poor search engine visibility. Authority site are established, large, well-known sites. Some authority sites may also be the leading non-profit or industry association. Many of these sites will rank organically for their site name at the very least. Truly powerful authority links will rank for a variety of important keywords as well as their site name. Authority sites linking to a site build up search engine credibility and trust – factors in attaining long-term top rankings.

Here’s a quick test to determine authoritativeness of a site:

1. Well-known industry leader as provider of news/products/services (eg. Expedia, CNN, Priceline)
2. Ranks well for their site name and several important keywords
3. High Google PageRank (usually PR6 or higher)
4. Domain registered on or before 1999
5. Alexa traffic score of less than 100,000

These are just guidelines and do not necessarily exclude sites that are registered after 2000 or has an Alexa traffic score of 101,000.

What about links from .gov and .edu top-level domains (TLDs)? Are these authority as what many SEOs believe?

First realize the .gov domains are extremely difficult to come by. They cannot be bought in the link broker market – at least for now. They can only be earned. Thus, the concept of link reputation.
Let’s take a quick look at some random examples of how these links affect ranking for various keywords searched in Google. Many of these, you’ll find, have .edu backlinks.
Keyword: Football
The site funbrain.com ranks #10 and has staff.edmonds.wednet.edu, sanjuan.edu, sccollege.edu, courses.dsu.edu, www.am.dodea.edu, and teams.lacoe.edu in their top 100 backlinks.

Keyword: Diamond Jewelry
The site jewelrycentral.com ranks #7 and has emporia.edu as a backlink in their top 100.

Keyword: Sales Training
In #3 position, businessballs.com ranks with www.edu.salford.ac.uk and cyber.law.harvard.edu in their top 100 backlinks.

Keyword: Auto Parts
The site car-part.com ranks #8 and in their top 100 backlinks have wright.edu.

Keyword: Phone Cards
At #7, 1stphonecard.com ranks with a number of edu backlinks in their top 100 backlinks namely www.public.iastate.edu, lists.asu.edu, owlnet.rice.edu, www4.ncsu.edu, lehigh.edu, fcim.edu, and www.rso.cornell.edu.

Keyword: Recipes
The site recipesource.com ranks #4 with www.math.northwestern.edu, bama.ua.edu, www.cs.princeton.edu, www.accd.edu, www.oaklandcc.edu, students.washington.edu, and www.nd.edu among the top 100 backlinks.

While authority links are not required for top rankings in all industries at this moment, having authority links will give a site the unfair advantage.

I would love to do a more in depth analysis of top ranking sites in the near future. That would be a great subject for a more in-depth report.

A forum thread at WebmasterWorld discusses the potential for article links expiring after seeing a drop in backlinks from articles.

I noticed the same thing this week – but what I saw was *not* article links being “absent” from the backlink query but rather articles from crap and “out of the box” article directory sites being “absent”. All the articles on quality sites and in a small portion of the article directories (tended to be the more popular ones) were there. Most of the articles that were up on crap scraper sites who grabbed them *via* article directories were there too (i.e. not published on “article directories”).

It’s possible they’ve figured out a way to identify and disclude the crappier, or less useful/linked to/quality article directory sites. It’s also possible they aren’t showing as complete of a list of backlinks that they used to.

I covered article marketing and how anyone can gain quick syndication through good articles that in turn earn the website owner lots of free links from different sites.

Think about it from the search engines perspective: Do they really want to display a lot of links from sites displaying the same *exact* article, page after page? Of course not! Many article links will in fact, be discounted because they come from low quality sites. The search engines only want to count the most authoritative links. Authoritative articles links usually result from other quality sites pointing to the article being syndicated. If an article has a .edu, .gov, or .mil link pointing to it, it makes it even more powerful and worthy of being kept.

So, an article marketer maybe able to submit to hundreds of article sites but will find that only a small percentage of those linkbacks actually count after several months. This is nothing to be really concerned about and will continue to happen as more low quality and content scraper sites pop up.