Search Engine Optimization

Web design errors and pitfalls


SEO

Search Engine Optimization

 

How to bring visitors to your Web site

An Online Handbook

Previous -   Optimization Metrics   Next - Getting listed in Search Engines

Don't do these things if you want people coming to your Web site.

Non-Html Files - Search engines like "vanilla" hand coded "static" html best. They will also list php and pdf pages, but these often cannot be optimized. Links in PDF pages may or may not be followed by search engines. A presentation file often doesn't give the search engine much information beyond its title. Word Doc files usually do not index well either. Word for Windows can also generate html pages, but these are too full of "junk" code to be useful as optimized Web pages.

Never delete a page or domain name if at all possible - It takes years to build search engine positioning and links. Don't stop your hosting and never change a domain name unless you are really going out of business forever and you know you will never use that Web site. You can add a second domain name, but leave the original site and domain name in place. It took many years to build the popularity of that domain, and the hundreds or thousands of pages that may be associated with it.

No redirection - Avoid redirection schemes as search engines usually view them as spam.

No Javascript - Avoid Javascript menus because search engines cannot read them. Remember that most people arriving at pages within your Web site will do so through search engines, NOT by clicking a menu link.

No Frames - Avoid frames pages or make alternative routes to frame pages.

Try to keep html "clean." There are some sorts of garbage like <STYLE ="FONT SIZE 12 pt"> that may stop a search engine spider while it is trying to crawl your page. Extra garbage code that is made by Front Page or Content Management systems can slow down the page anyhow. There is some indication that a large proportion of "code" – (whether it is total garbage or javascript or style directives) relative to visible text can be very bad for your site. It may be for example, that keyword density is figured for the entire page including the junk code and html directives. If the keyword appears 10 times in 1,000 text words it is "10%." But what if the keyword appears 10 times in 100,000 "words" most of which are html or Javascript commands?

Avoid PDF, Powerpoint, Doc, Flash and Images - Avoid PDF, Powerpoint, doc, flash and images as total page content because it is difficult or impossible to optimize these files for search engines, and because they cannot easily be used to link back to other pages on your Web site in a way that search engines can follow. If you must use them, make a separate html page that links to them and has the keywords and introductory text, plus some appropriately tagged images.

Content Management Systems and Web Logs - As noted, these systems require special treatment and will be discussed elsewhere. There are good content management systems. In theory, the best ones will produce better pages than you can code by hand. Be aware though that search engines may be wary of pages they consider to be "automatically generated.

Relative URL addressing - Never use "relative" URLS like /docs/mydoc.htm. Many search engines cannot or will not follow these. Always use the full path link like "http://mysite.com/docs/mydoc.htm

Don't build a page on images only - Original images can bring a lot of visitors to your site if they are seen by Google photos and other special photo search engines. Use a meaningful name for the image file that contains a keyword the search engines can recognize. However, never build a whole page on an image only. There usually has to be a reasonable amount of text on a page. (exceptions might be large maps - but even they must have a title on the page and should have an <img tag). Do not use Java script for displaying images.

Avoid Bad Links - It is not known whether bad links (links that do not lead to a working Web page) to external Web sites can hurt your site or page ranking, though they probably do. Bad links to your own pages almost certainly hurt your ranking. Bad links to your own pages are usually caused by coding errors. Bad links to other people's Web pages are inevitable, because pages get removed from the Web or moved all the time. That should not happen, but it does. Use a link checker periodically to fix the worst of the bad links at least.

Avoid orphan pages - Every page must have at least one, and preferably two, working full path links that lead to that page. Every major section should have a link from an external Web site.

Avoid Black Hat SEO Practices - Sooner, rather than later, the worst Black Hat practices catch up with their practitioners.

Notice: Copyright

All materials are copyright 2008 by Ami Isseroff. All rights reserved. These pages may not be reproduced in any form in electronic or printed media without express written permission from the author.

SEO

SEO Basics

The SEO Book

SEO Articles

SEO Blog

SEO Glossary

Web Pro World Forum

More Links

Love Poems

MidEastWeb Middle East

Ariga: Middle-East

Zionism

 

SEO - Web Site Search Engine Optimization Contact: Webmaster(at)Yu-hu.com
site map

Web design errors and pitfalls