What is Latent Semantic Indexing (LSI) and why should I care?
Google and other search engines have been required to constantly improve their algorithms to provide relevant content to searches. No longer is it enough to have a certain percentage of a keyword phrase in your copy, or a certain number of inbound links in anchor text (all still important). Now, the Search Engines are looking to make sure there are plenty of related words in the copy to increase the quality of the search results.
LSI considers words that are not necessarily literally semantically related but that are used in the same context and is more like humans categorize and classify information. LSI knows to connect “Apple” both with “Oranges” and also with “Computer” as it is used in context in common speech.
LSI and The Fractal Connection
The 20th century Mathematician Benoit Mandelbrot used mathematical equations to produce images called fractals which map the probabilities of word occurrences in English. Interestingly, the way we use language and classify documents can be described by mathematical equations similar to those that describe other chaotic systems. Therefore the seemingly mathematical and abstract principles and concepts underlying LSI are surprisingly similar to how humans use words and organize documents.
Because of this, web content that performs well under LSI analysis is likely to be more thoughtfully written, higher quality content and not something a robotic word machine could crank out. What this means to our search engine optimization (SEO) efforts is that every word on the page and not just our chosen keyword is important. In fact, this very document is being written with this in mind, weaving in words we know are seen by Google as related so that this article will stand up well to semantic analysis (one of those related phrases).
Kaizen Marketing’s tagline is “the art and science of continual improvement” and fits in well with this discussion, as using LSI is both art and science working toward the continual improvement of our web page rankings. Used correctly, it can propel a simple web article or blog post into the top 10 Google results in a matter of hours, and I have done that many times recently. This very article is currently ranking at #4 for “latent semantic indexing seo” and #9 for “latent semantic indexing LSI”, within 24 hours using only this method and just a bit of social bookmarking.
How to improve SEO using LSI
By placing additional weight on related words in content, LSI has a net effect of lowering the value of pages which only match the specific term and do not back it up with related terms. In fact, some pages may have their rank lowered by being “over-optimized” for one particular phrase.
So, mix up your anchor text and keywords used. For example, if I were working on SEO terms, I would also use various anchor text combinations so that the linkage appears more natural.
Latent Semantic Indexing (LSI) is good for the web searcher because they get more relevant, compelling content. It’s good for the search engines because it increases the quality of the content in their databases. And its good for your business because it ensures you’ll have content that drives more traffic to your site for more conversions and sales.
LSI Secret Weapon – NEW!
Kaizen Marketing uses a powerful software tool that combs through the top 10 Google results for any keyword phrase and produces a report showing the most common 2, 3 and 4 word phrases that are showing up on these top ranking pages. A new article or post written based on these phases can reach top rankings in a matter of hours. Allow us to help you achieve these results for your targeted phrase. We have made it very affordable for small businesses. And besides, you leaving money on the table by NOT getting your web page onto this valuable piece of commercial real estate. Contact us for a free eval of your site.
Where do I learn more? See technical doc here:
Latent Semantic Indexing (LSI), by Clara Yu, et al., National Institute for Technology and Liberal Education, January 1, 2002.