Search Engine Tips
How Search Engines Work
If you want to design websites
that rank well on such search engines as Google, Yahoo, or Ultraseek, you'll
need to understand some basic principles about search engines. All search
engines use robots — also known as bots
or spiders — to scan websites and catalog information. These bots
index pages by reading text and following HTML tags.
What makes search engines capable of providing relevant results are the powerful algorithms they use to analyze the information gathered by the bots. Search engines rank webpages based on their analysis of a site. By understanding how search engines work, web designers can improve how well their websites rank in search engine results.
Search Engine Tips
It is very important to create sites that bots can read and understand. Here are some helpful tips for making your site search-engine friendly:
- All search engines use the "Location Frequency Method." They look at the text closest to the top (location) of the page and how often text is repeated (frequency). Your title and header tags are very important tags and should always contain your most popular key phrases. Avoid using large images and scripting at the top of your pages as this pushes the most important elements of your page (the text) to the bottom.
- Know what key phrases you should use on each page. Key phrases are terms your target market will use to search for your website. Sometimes the terms you think people would use are not always the best ones; you should test this to find the best key phrases. Once you know the terms, use them in the title and header tags, place them in named links, and repeat them within the text of the page. Optimize each page on your site for a specific key phrase.
- Bots cannot read images or image maps. If you've created a site that is primarily images with very little text, you will not do well in search queries compared to sites that have relevant text. Remember, bots are "text hungry" and need to determine the overall theme of your site. You can use ALT tags on your images, and they are important for many reasons, but they have very little weight on the overall ranking process of the search engine.
- Avoid the use of frames. Search-engine bots will index each of your pages individually, not within the frameset. This means that all your internal pages become orphaned and people cannot navigate your website unless each page has a link back to home.
- Avoid the use of "Flash Only" or "Image Only" websites. Most bots choose not to read Flash, so they see little text or relevant information to determine your website's theme. Meta tags in Flash can help, but are usually not enough to help you rank well in the commercial search engines.
- Use meta tags within the head tag of your document. Although these tags are not as useful as they once were because of overuse by web designers, a few search engines still use them. Verity Ultraseek — the search engine used on the UNLV web server — still indexes this information and, in fact, weighs meta tag information quite heavily. For this reason, you should use meta tags for your UNLV webpages to ensure that your pages rank well when visitors search for you from within the UNLV website.
- Do not lock your links in images and JavaScript. If you use rollover images for the main navigation of your site, the bots may not be able to understand how to travel through your website. All bots follow HTML links, but lack the ability to understand links in Java, JavaScript, Flash, etc. If your links are trapped in images or JavaScript, Flash, or Java, you will need to provide text links at the bottom of your pages.
- Dynamic websites (ASP, PHP, ColdFusion, ASP.NET and JAVA) have difficulty being indexed by bots. Any urls with a "?" in them are a potential problem to search engine bots, which may not follow such. Google is the only search engine that can effectively index dynamic websites.
- Bots return to websites about once a month, so it may take a while for your changes to show up in a search. To see what search engines have visited your site lately, refer to your web server log files. For fun, the bots have been given names by their engineers. Verity Ultraseek's bot is called Slurp, Google's bot is called Googlebot, Lycos' bot is called T-Rex, and Altavista's bot is called Scooter, to name just a few.
Enjoy the traffic your website receives with your new search-engine friendly pages.