Urls are the entire text that appears in the address bar (e.g. http://www.thesite.com/folder/a-page.htm). Generally they are made up of two parts:

• The Domain Name - e.g. http://www.thesite.com/folder/a-page.htm
• The location (not always there) - e.g. http://www.thesite.com/folder/a-page.htm
• The page name - e.g. http://www.thesite.com/folder/a-page.htm

The Domain Name

There are several important factors here

• ideally it should include your most important keyword in it so if you own a flower shop in your town the domain www.flowers-mytown.something would be ideal
• The ending of the domain name should reflect the users you want to attract - .com for the world; .co.uk for users from the United Kingdom and so on
• Domain names that have been around for a long time carry more weight for search engines so, depending on price, it may be worth buying one that someone has been sitting on for resale purposes

Note once again this is 'all things being equal'. If we do a search on Google for "Land for sale Poland" we can see that there is a website called "http://landforsalepoland.com" but it is not at the top of the search results. (Position one is actually held by a client of ours!) There could be a number of reasons for this including:

• The other sites are older
• It's smaller than the other sites (which have more pages)
• There are errors in it's HTML
• Visitors don't stay on it for long
• ... and so on

So domain name is important, very important, but not vital.

Canonicalization issues

Often overlooked but important is a way that search engines can suspect you of showing duplicate pages with the same content even though you don't mean to.

Many websites have this issue on their home page which can be accessed as:

• http://www.mysite.com
• http://mysite.com
• http://www.mysite.com/index.htm
• http://mysite.com/index.htm

So there are four pages (and that's often a minimum) that all have the same content.

You can solve this in three ways:

• For Google and some other search engines they will take your sitemap.xml as definitive so long as you have not accidently made links on your site which point to the other versions.
• Use the rewrite rule in the .htaccess file to forward visitors and search engines to the right page (so, for example, anyone who type 'http://www.mysite.com' will be instantly forwarded to 'http://mysite.com')
• Use the canonical meta tag to tell search engines the definitive page (this says "I know you came here via 'http://www.mysite.com' but please remember this page as 'http://mysite.com')

The URLs

The remainder of what appears after http://www.mysite.com/ is what we will look at here.

Originally websites had pages like http://www.mysite.com/products.htm and http://www.mysite.com/services.htm but these don't really help search engines and they don't truly reflect the contents of your page.

When users search the internet they also see the url before they visit and many will use it as an indicator of whether or not your website will be useful so bad URLs can affect your Click Through Rate (which we will cover later).

A better scenario would be for a website to have pages with urls like this:

• www.bobs-plumbing.com - so the home page does have a keyword in it - "plumbing"
• www.bobs-plumbing.com/central-heating-service.htm
• www.bobs-plumbing.com/24-hour-emergency-contact.htm
• www.bobs-plumbing.com/leak-fixing.htm

Now so long as these pages do contain the content they claim the search engines will give them a strong weighting and users will also find the link more tempting.

Dynamic URLs

These help programmers by reducing the number of pages they need to create for content which is often pulled from a database. So a news site may have one actual page for displaying the category headlines but the content will vary depending on what is in the urls.

As an example:

• www.thenews.com/headlines.php?category=business
• www.thenews.com/headlines.php?category=international
• www.thenews.com/headlines.php?category=sports

These are all the same page from a programmers perspective but the content shown in the page will vary depending on the category.

The problem is that the urls to the pages look ugly on search results and are difficult for search engines to read. Luckily with the help of an .htaccess file and a piece of coding called 'mod rewrite' we can change these urls into something more friendly like:

• www.thenews.com/business-headlines.php
• www.thenews.com/international-headlines.php
• www.thenews.com/sports-headlines.php

These urls are much more attractive in the search results, much easier for the search engine to read, and nothing under hand about it. This is exactly the way big sites like Amazon work and it is a vital part of SEO.

Server Location

Now although you may have the right domain name, and the right urls there is one final part to the jigsaw puzzle that plays a role and that is the physical location of your server.

In other words where your domain and urls are called from is important, sometimes vitally so.

If your website is all about dating in the United Kingdom but you have tried to save a few pennies by registering your domain and buying your server space in Pakistan this might be a false saving.

The search engines know exactly where your site is hosted and a UK dating service hosted in Pakistan - well that's just odd, perhaps even suspicious. All things being equal the sites hosted within the UK will take priority.

Now if the country you are hosting in is fairly unregulated you can also find yourself going through bouts of unfair penalties. If a large number of servers are suddenly used in a country for spamming or hacking, perhaps even servers from the same company you bought yours from, perhaps even on part of the server where your website is - then you are going down...

So it is safer, even though it may be marginally more expensive, to choose a respectable webhosting company in the country or continent where you are planning to find your clients or much of your hard work may go to waste.