Home arrow Website Advice 1
Website Advice 1



Website Advice 1


You cannot compete at the top level unless you put search engine optimising before any other consideration.

This is something you will learn to appreciate eventually - but to save you many, many years of hardship, we're telling you now.

So, whether or not you become a client, here are some things you ought to know as a website owner that will save you time and trouble. In fact it helps if you know one or two of these things before you become an owner, as you will see.

1. Never buy website technology without speaking to a consultant first

This is the #1 piece of advice for potential online businesses. The pitfalls here are massive, and impossible to escape once you’re in. Your web application, whether it be for a smaller dynamic site, a big CMS site, or an online store, must be built from the ground up to be modern-SEO compatible, or you will never be able to compete online at the top level.

Developers, that is to say the software authors who create website systems, tend to consider this aspect last. They are too busy building in features, and forget the basics. Instead, they need to start with search engine optimising - they need to design good SEO into their work as a core component, though few do. We have to sort out the end problems, which in some cases is impossible, in others just very expensive indeed.

Any dynamic site, either brochure-based or for ecommerce, should fulfil the following conditions before you even think about trying it:

  • All pages must validate 100% with the W3C right out of the box, with no dev work. This just means the pages must be coded correctly, from the beginning, with no obvious faults, without additional developer input. It is a very simple and basic requirement - and obvious, one might assume; but surprisingly hard for some to comply with.
  • URLs should be SEF-compliant. To translate: the page addresses should be short, human- comprehensible and therefore 'search engine friendly'. It is better if the CMS writes native short URLs, since this causes less server overhead, although server cacheing reduces the load. In practice many content management systems have the URLs rewritten by the server, and the end result is more or less the same.

The URLs are the toughest test for any app, as up to now only the cleverest and most forward-looking of developers have realised that SEO is the first job and not the last. To be fair to the majority, mod_rewrite + htaccess can well be a component of perfect URLs. Very few apps produce perfect URLs without server assistance, Plone for instance being one of the few.

However, there is a clear difference between producing an absolutely perfect URL via Apache, and simply removing a couple of queries and levels from the address. A perfect URL has no ?, or =, or &, or anything else like that in the address. Some software authors know how to accomplish this, some don't. Easy to check: look at one of their sites.

Search optimised page adresses

Here is a correctly formatted address:
http:// www.yoursite. com/camping/tents/blue-tent.html

(We put a couple of breaks in to stop it coming across as a link.) You can clearly see that it’s an URL (page address) for a certain type of tent. Here, though, is a poor dynamic site’s equivalent. This is a genuine example, and believe it or not is one single line:
http:// www.yoursite.com/ Merchant2/ merchant.xyz? Screen= PROD& Product_Code =CMPG 1368& Category_Code= CamTen3758& Product_Count=8

We had to break it up as it crashed the page formatting as one line. No one could possibly type this without error, as it is twice the length of the address bar; people hate these URLs; and so do search engines. You cannot compete at the top level with page addresses like this; and you cannot compete on a level playing field even if they are rewritten and half-corrected by the server via mod-rewrite and htaccess. That is not to say you cannot get good search positions; only that you cannot compete in tough markets against tough opponents with URLs like this.

There are plenty of people who will argue on this score, but they are usually trying to sell you something…

Page code validation

Go to a web page on a site built with the application you have in mind; perhaps one from a ‘gallery’ of customer’s sites that the software house show as examples.

Copy the URL to your clipboard; then go to this page and enter the URL in the test box there:
http://validator.w3.org

The results should show up as green; there should be no red; there should be no ‘fail’ notice. To be fair, it can be tricky to get some production pages to validate fully, so you have to be able to evaluate how serious the errors are; one or two might be allowed if they refer to charset problems for instance.

These are the basic requirements for any website application, i.e. software that serves your web pages to the Internet; whether for a CMS, an ecommerce site, a semi-dynamic site, or just a normal 'flat' site. After these have been complied with you can go on to the next stage.

If your proposed supplier states that these requirements are difficult to comply with (or even impossible, or pointless, as we have heard occasionally), then recommend them to examine the major open-source CMS, Plone. It’s free; it can probably handle up to a million pages; the pages validate 100% with all authorities out of the box; it does not need mod-rewrite or an htaccess file to fix the URLs; its URLs look like the first example quoted, not the second; and it is used by household-name enterprises. Naturally, it isn’t perfect, as nothing in this world is; and though it may be free, there would be considerable development costs for customisation work.

However, as an example of how to do it properly, it is worth comparing with any application that you intend to pay for. Paying for an application that is demonstrably second-rate in the first place doesn’t seem like a good commercial decision.

It is up to you whether or not you want to succeed online. If you do, then the first thing that you need to consider - not the last - is web quality optimising. Once again, here is the mantra you need to learn:

2. You cannot win at the top level unless you put SEO first

Our definition of SEO differs from some others because we believe it refers to a quality improvement process. Quality does not necessarily equal expensive - some of the highest-quality web applications are free.

 
Web Business Managers