Tool Box
Issue No. 18 - August/September 2004
Search Engine Optimisation
things to consider before you design a new website
by Frank Grasso
A search engine uses a spider (an automated piece of software – also known as crawlers and robots) to find content to include in its searchable database. Spiders are getting more intelligent with every search engine update, but they still rely on the source code of a web page for content.
Four source code rules
1. Images and movie files are invisible to search engines.
Avoid putting keyword-rich content in images or rich media files. Search engines read source code and can only read the filenames, not the words that are inside an image file. An example of keywords in an image file is when a page heading is written as an image. The heading could read "blue widgets made to order" but the file name is likely to be "image.jpg". The search engine would read the name of the file (image.jpg) and not the keyword "widgets". If this same heading was written in text the search engine would read the keyword "widgets" and know that your website has something to do with widgets.
2. Eliminate any unnecessary code or comments, text or images
Search engines calculate the keyword density of each webpage according to a search query. Consider a customer looking for a "blue widget". A search engine has the task of finding the best match for that search query. Leaving link analysis aside for now, (which is a key determinant of where you finally rank) search engines calculate how many times the keyword appears in the webpage as a proportion of the total source code. Unnecessary code text or images can increase the amount of content in a webpage and therefore decrease the keyword density.
- Put JavaScript in external files
- Do not clutter the code with unnecessary tables or comment tags
- Apply source formatting to the webpage
- Use title, keywords and description tags but avoid any other tags
3. Use keywords to name images, files and subfolders
By using meaningful and descriptive ter...
