The structure of a website
When you’re starting a new website, it’s navigation should be designed well.Generally, there are two types of users who open your site; People and crawlers. Unlike crawlers, people are able to click on JavaScript, Flash and other stuff. This is the basic reason why you shouldn’t use such things on your pages. But let’s go behind this problem and explain everything that relates to this topic.
Every serious search engine cares about incoming links. More links = more popularity of a specific page. More popularity = better rankings. It’s simple. There are a few principles which you should bear in mind when creating/redesigning a website:
- 1) Put a navigation menu to each page of your website
- 2) Use a text links rather a graphics
- 3) If using a JavaScript (or Flash) menu, don’t forget to put a sitemap to your site
- 4) Use valid HTML (or XHTML)
- 5) If using images as links, don’t forget to use proper ALT and TITLE attributes
Let’s go through the basic website structures.
1) The Hierarchical model
I strongly recommend to avoid this type of structure. However, it looks pretty simple and it is good when you’re new to WEB pages, WEB development and you’ve never heard of dynamic webpages, including functions and so on.The hierarchical model has the following scheme:
Practically, it means that categories aren’t linked together and there is one-way relation between a category parent and its children. Directories are the best example of this structure. Usually, when the main page get some PageRank, any other page can’t get a better PageRank this way.
If you decide on this model, recognise that your category pages won’t be very popular and their popularity will mostly depend on the main page’s popularity. On the other hand, you can use two-way relations between a subcategory and it’s parent. In this way each category will be linked from it‘s children and it will mean higher popularity for the parent pages.
2) The Network model
This kind of structure is probably the best one and in its essence, it’s about creating links between related pages as well as unrelated pages. If your site is about cars, it is no problem to link two different pages if they’re about two different models designed by one manufacturer.Consider the following scheme:
As you can see, there are parents and children only as objects. Even if some visual relation exists, they are fully independent. Another advantage is that each object can link to another one. This will increase the popularity of the most linked page. In our case, these objects are the most popular ones:
- 1) Parent Object 3
- 2) Child Object 1
- 3) Child Object 4
You can also calculate weight values sent to each object, so you will be able to determine the most popular page too. It is up to everyone to set the highest rank to the Parent Object 1 or another one. Once you assign the first rank value to your starting page (the main page in most of cases), you can determine any page’s popularity.
Well, once you decide the structure of your page, don’t forget about putting together the above models. Probably the most important thing is to link to your main page from each page on your site. What happens if you put a link to your main page on every single page of your site? The popularity of your main page will probably double, maybe it will multiply much more. The following image shows the ideal situation, when a site is divided into a few categories, these categories contain several subcategories and everything is linked together. Subcategories contain single pages with the most important content and every one of these pages links to the main page. Of course, you won’t do badly if you put a link to the root categories to every elementary page.
So if your site has the pagerank of 4 and your site is structured well, containing more than 1000 elementary pages, your main page could get the pagerank of 5 or 6 if everything goes well. Search engines like MSN and Yahoo don’t have any pagerank, but their algorithms calculate a link popularity too, so it will also mean great things.