Knowledge Sharing - About SEO

Wednesday, August 8, 2012

What is LSI?

LSI is short for latent semantic indexing. This refers to different words that have the same or similar meanings (or words that are otherwise related). For example, “housebreaking a dog” and “housetraining a puppy” are two entirely different phrases, but they mean about the same thing.

The reason this is important is because Google analyzes webpages using LSI to help it return the most relevant results to the user.

For example, a page that has the keyword “housebreaking a dog” but NO other similar words (like housetraining, paper training, potty training, puppy, dogs, puppies, etc) probably really isn’t about housebreaking. End result: Google won’t rank it as high as a web page that does include a lot of relevant, related terms.

What does this mean to you? When you create a web page around a keyword, be sure to also include the keyword’s synonyms and other related words.

Pure LSI analysis isn't scalable enough to handle the volumes of data that Google processes. Instead they use more streamlined and scalable content analysis algorithms that have some basis in LSI and other related technologies. It also appears that this analysis is ongoing and not just a one time run through the system.

Cliff Notes: Don’t write content that a drunk 4th grader would be ashamed of. Spend the extra couple of minutes to write decent stuff and you’ll be fine.

What is a C-class IP and why should I care?

A computer’s IP address is it’s address on the Internet. A C-Class block of IPs are ones which are next to each other. Links from the same IP have very limited value. Links from the same C-Class IP block have a little more value but still not much. Links from different C-Class IPs are worth the most.

Not as important as it once was, especially when it comes to sites hosted on huge shared server clusters like those at HostGator/ThePlanet, BlueHost and others. The shortage of available IP addresses is driving this.

Most importantly tons of domains all on the same IP or C-Class that all interlink are the fastest way to announce to Google that you’re trying to cheat the system. This may have worked a couple of years ago, now it’s just a flashing neon sign telling Google to deindex you.

What is the "freshness" factor?

Search engines such as Google prefer “fresh” (newly updated) web pages and content over stale content. That’s why when you first add content to your site – such as a new blog post – this page may sit high in the rankings for a while. Eventually it may sink to a more realistic ranking.

It’s this “freshness factor” that allows your pages to get those higher rankings, even if the ranking is temporary. Thus updating your pages frequently can help push them to the top of the rankings.

This is one of the primary reasons why you hear people talking about how “Google loves blogs”. Google doesn’t love blogs, Google loves regularly updated sites.

What are meta tags?

Meta tags are information that you put between the <head> tag of your web page’s source code. These meta tags primarily tell search engines and other user agents about your site’s content (description), keywords, formatting, title and whether you want the search engines to crawl (and index) the page.

There are also some tags that are shown to the user, such as the title tag (which is the title that appears at the top of your browser).

Note that the big search engines no longer take these tags into consideration when ranking your web pages (with the exception of the title tags). Some smaller and more specialized search engines still utilize the keywords and description tags when ranking and displaying your site.

What is a doorway page/cloaking?

Cloaking refers to showing one page to a search engine and a different page to your human visitors. Doorway pages are optimized pages that pull in SE traffic, but this traffic is immediately redirected (either manually or automatically) to a different page.

Google and other search engines do NOT like these practices.

What's the truth about duplicate content?

There is no duplicate content penalty when it comes to multiple sites. Otherwise, your shady competitors could just create near-clones of your site to make your site disappear. But that doesn’t happen. Indeed, run a search for a PLR article and you’ll likely see many SE results for that same article.

TIP: Nonetheless, it’s better if you have unique content, rather than competing with others for the same keywords using similar content.

What about duplicate content on your OWN site? In other words, what happens if you have two web pages with the same content but different file names? In that case, refer to the question on rel-canonical for instructions on how to deal with this.

Tuesday, August 7, 2012

What is rel="canonical"?

If you have two or more pages with similar content, you can tell Google which is your preferred page to show in the search engine results. This is referred to as your “canonical” page. If Google agrees this designated page is the best version, it will show this preferred page in its index.

To tell Google which page you want listed as the canonical page, add the following bit of code into the head section of the similar (non-canonical) pages:

<link rel="canonical" href="http://www.example.com/filename.html"/>

Naturally, you should replace the example.com/filename.html with your actual domain name and file name.

For example…

Example.com/file1.html is your preferred canonical page, the one you want displayed in the search engine results. You don’t have to add any tags to this site.

Example.com/file2.html and Example.com/file3.html have similar content to example.com/file1.html. As such, you’d place the canonical code within the <head> tag of these two sites to tell Google that example.com/file1.html is the most important page.

The most common reason to do this is to tell Google that these pages are all the same –

Example.com
www.example.com
www.example.com/index.html
Example.com/index.html

Don’t go overboard with this and certainly don’t use it on stuff like paginated comment pages because they are “similar” but contain the same post. They contain enough unique content to be treated as unique and Google will start to ignore your legitimate canonicals if it finds too many instances of you misusing it.

Yes, Google thinks it’s smarter than you, deal with it and move on.