Understanding Heuristics


In order to maintain a web site with the best search engine optimization possible, you have to be familiar with heuristics. This is simply a term for recognizing a pattern and being able to solve a problem or come to a conclusion quickly and efficiently by consulting what you already know about that particular pattern. In other words, using heuristics is a way to solve a problem, although it’s not always the most accurate way. 

Patterns, proximity, and stemming

Heuristics are important in search engine optimization because they allow for variations in the way that users search for a particular keyword or phrase. Because a combination of factors must come together to create a ranking for your web site, heuristics make it possible for some, but not all, of those factors to be present.

For example: Let’s say you run a travel-planning web site. If a web user is searching for ‘‘springtime vacations,’’ a search engine crawler will visit many sites, with varying keywords, keyword placement, and keyword density. In effect, it will give each a score, calculated on a baseline for relevance. It may find one site with the phrase, ‘‘some writers get their best ideas in springtime or while on vacation.’’ But it won’t score that site high, because it doesn’t meet baseline criteria very well. The keywords are separated and the context is wrong. In addition, links from that site are unlikely to support the idea of planning a springtime vacation. The search engine likes your travel-planning web site better, because it has a lot to say about ‘‘springtime vacations.’’ The crawler doesn’t stop with your site, however, and it doesn’t look just at the words in your links, although it helps if those say ‘‘springtime’’ and ‘‘vacation,’’ not something vague like ‘‘trips.’’ 

The crawler will actually go to your links to see if they’re truly helpful for the user who wants something about ‘‘springtime vacations.’’ If your links are irrelevant to that, the crawler may decide you’re running a link farm, designed to catch its attention without really delivering. Conversely, if a high percentage of your links are actually related to springtime vacationing — travel information, garden shows, trips to tulip festivals — then the crawler may score you high and put your site high on the list it compiles for the user. That user, after all, is the crawler’s customer — and you hope also yours.

The crawler has operated heuristically, making its best judgments at each stage of the process by examining the patterns and proximity of words used on a web page. Keywords apply to heuristics because they provide the pattern by which a problem (that is, the search) is solved. Why do you need to know all of this? Because understanding the pattern by which your site is ranked will help you understand just how important it is to properly choose and place keywords that will improve your search engine ranking. Think of it as a rule of thumb. 

Heuristics provides a working guideline by which a search term is ranked. Certain words or phrases used in certain patterns can be indicative of a specific topic. To take the concept of patterns and proximity one step further, there is another element in the way words can be used together called stemming. Essentially, stemming is the growth of one related word from another, using prefixes and suffixes.

In other words, all the forms of a root word can be considered stemmed words. For example, if you’ve chosen ‘‘book flight’’ for your travel web site, then the stemmed keywords for that phrase might be as follows:

■ Booking flights
■ Books flights
■ Pre-book flights
■ Re-book flights
■ Re-booking flights
■ Pre-booking flights
■ Re-booked flights

Suddenly, one word becomes many, which means you have many more opportunities to use that base word. A lot of discussion about SEO has gone into the debate about how effective it is to use stemming in your web site content. The general consensus seems to be that stemming is a good method for boosting the number of times you can use a keyword on any given page.

Here’s the rub: You still should be cautious of overusing any keyword or set of keywords. Some search engines have started to recognize only the root word of a keyword. For example, if you use the terms ‘‘pre-book flights,’’ ‘‘book flights,’’ and ‘‘re-book flights’’ in the same article on a page, the search crawler might recognize all forms of ‘‘book’’ as the same word. In other words, it’s possible the crawler could view your combined use of these terms as more than is considered ethical by the standards of keyword density. So what’s a person to do? Use the words but be mindful of the 7 to 10 percent keyword density already discussed. The idea of stemming becomes very useful when you find you’re having trouble using your desired keyword or phrase often enough on a web page without making the tone of the page completely foreign. If you’re using several forms of a word, however, you can do that to meet your goals for keyword use and still create content that flows naturally.

It all goes back to creating pages for your user. Don’t use a keyword on your page just because a search engine might like it. Landing in a search engine’s favor should only be a goal as long as it leads you toward creating a useful and informative site for visitors. If those visitors find your content hard to read because you’re showing the maximum allowable number of keywords even though they don’t make any sense on your web page, you’ll turn your visitors away, and then what’s the point of your efforts?

Remember too that rankings are achieved through a complex combination of factors, not all of which are completely predictable. These guidelines are just that — but they can help you set a standard for how you plan to use keywords.

Thanks for reading.

Image source: Google

Post a Comment