Sunday
February 25, 2018

Tech@Work: Artificial Smarts of Search Engines

      |
-A A +A

Tech@Work: Artificial Smarts of Search Engines

Search engines have learned to read context.

If you’ve worked hard and finally mastered how to use keywords to improve your site’s standing in search results, I’ve got some bad news: Keyword skills are no longer enough.

The newest way to improve your site’s placement in search results is using latent semantic indexing. When a user conducts a Web search, LSI not only analyzes the keyword phrases that appear on each page of your site but also attempts to connect these words with the theme of the surrounding body copy, which gives the written piece its contextual meaning. Search engines such as AltaVista and Google use the artificial intelligence built into LSI to analyze the links and the words on a page to determine a page’s theme.

Think of it this way: If you type “Chicago Bears” into a search engine that doesn’t use LSI, you might get highly ranked pages of hotels or businesses in Chicago or pages about the decrease in the bear population at Yellowstone, as well as information on the Chicago Bears football team. With LSI, the search engine focuses on the relationship of the two keywords as well as the surrounding body copy that provides context and then serves up specific pages that pinpoint your specific interest.

The ultimate goal of LSI is to reward pages that are well written and that have genuine value to the searcher. With LSI, using a good keyword in the wrong context, will get you a lower relevance ranking. This means that site developers will no longer be able to score high search placements simply by loading sites with desirable keywords or a multitude of links.

When our company recently decided to test the impact of LSI on search results—without any other search engine optimization techniques, such as using a keyword phrase in the title or heading tag—we were pretty amazed. We used a tool called OptiRanker (www.optiranker.com, $49.95 per month), which provides you with a list of supporting, descriptive words that a search engine believes are important to your primary keywords based on what the search engine understands contextually.

After you register online at OptiRanker, type in one of the primary keywords or phrases you’ve chosen for your Web page. We chose a highly competitive phrase, “search engine articles,” for our test. You might want to use “real estate [my town].”

Next, choose the search engine you want to analyze. I used Google, but OptiRanker can also test MSN or Yahoo. Press the Submit button. It can take anywhere from three to five minutes to return the results.

The program returns a color-coded chart showing the highest-ranking word combinations that you need to incorporate into each page of your site. Each word is assigned a number value; the higher the value, the more important that word is to ranking success. The whole point is to use the words that OptiRanker recommends to flesh out your content. The easiest way to do this is to simply print out the list of words and then start crossing off each one that you already have within your copy. The words that remain on the list are the ones you need to re-work into each page or article on your site. You can’t just dump the words on the page; you need to mildly rewrite the content so that you incorporate the words in a readable way.

For a highly competitive term like “search engine articles” with more than 61 million competing pages, our page has continually appeared within the top five results on Google using LSI alone.

The only downside to LSI is that the technique takes time to take effect. Allow at least six to eight weeks for the search engine robot to communicate the changes back to the search engine and for your page to begin gravitating upward. But the wait is well worth it.

LSI is a smart complement to a well-developed search engine optimization strategy. Don’t you want to be as savvy as the technology you use?

1.666665
Average: 1.7 (3 votes)
Your rating: None

Notice: The information on this page may not be current. The REALTOR® Magazine archive is a collection of content previously published on RealtorMag.REALTOR.org. The archive pages are not updated and may no longer be accurate. Users must independently verify the accuracy and currency of the information found here. The National Association disclaims all liability for any loss or injury resulting from the use of the information or data found on this page.