K.I.S.S. - Web Design for SEO (and Happy Users)

K.I.S.S. - Web Design for SEO

by Bill Marshall
of http://www.oyster-web.co.uk

Last updated: 12 Sep 2007

Think of your favourite Beatles song, or the idea that someone made a million out of that made you think "why didn't I think of that?", or even your favourite software application. Chances are they'll have one thing in common - they're all simple and elegant, seemingly so obvious that you think they've always been around. That's how most Web Site designs should be if you want people to use them, come back to them, spend money on them.

So many websites, particularly business sites, seem to have been built on the basis that they want to cram as much in as possible. Your eye wanders about, trying to work out what the site is trying to tell you and how it's organised until it eventually gets fed up and loses interest. Your message is lost, the information you were trying to convey is never found, the sale is never made. If that sounds like your site then it's time to rethink it.

A Quick Historical Digression

HTML was derived from SGML (Standard Generalised Markup Language) which was invented primarily for the various diverse NATO countries to send each documents and specifications that would always be compatible. SGML is concerned with document structure - showing you what is most important, what is next in importance and so on. In many ways it was like the structure that is supposed to lie at the heart of a word processor document - Heading 1, heading 2, Heading 3, body text 1, appendix etc. etc. The idea was that the structure should always be standardised so however the document was interpreted by the local user's computer the structure was consistent.

The first web pages using HTML were very similar to this idea. Everything had a grey background (easy to read against without eyestrain) and all the text was in Times Roman. Headings of various levels were always the same sizes.
However once HTML got sophisticated enough to include images and basic style commands things went a bit awry. That was at the height of the DTP boom when everyone reckoned they were publishers and could break long-established style rules with impunity. "Web sites should be like DTP documents" they cried and set about using every trick they could think of to make page layouts that were a visual leap forward on the existing ones. Tables were drawn into service as layout devices and since everyone was using pretty basic editors they would spend hours slaving over nested tables to pull their layouts together.

Even then there were dissenters who said that trying to force the end-user to view things the way the designer wanted was against the basis of HTML/GML, but everyone was far too captivated by the new possibilities and such appeals were lost in the rush. Netscape and Microsoft responded to the demand by pushing the boundaries of display ever further (and ever further apart) in their new browsers, and it got so complicated that anyone who understood HTML could charge £80 an hour. Needless to say the code produced then was awful - hacks and get-rounds abounded, JavaScript reared it's head and was pressed into service to show different pages to different browsers (which of course meant you had to keep multiple copies of sites updated). You started to see the notorious "best viewed in ...." notices as designer who couldn't manage to code for more than one browser tried to cover their lack of knowledge by claiming their chosen browser was better than the others.

Then came the WYSIWYG editors - excellent tools in the right hands but just like the DTP programs of a few years earlier they made everyone think they were web designers. They created layouts using even more complex tables. Some of them created other layouts using "layers", and if they ever mixed with tables in the wrong manner then all hell would break loose in one or other of the main browsers.

Sadly you can still see plenty of code like this around today - nested tables five levels deep, spacer gifs holding cells open or moving text around, font tags which repeat themselves on every line or in every table cell. And in amongst it all you find absolutely no structure at all. There are no H1, H2, etc, no paragraph tags because each paragraph is in a table cell. no list tags for the same reason. In some cases a bit of basic CSS has crept in but it only seems to make things worse - instead of font tags (or sometimes as well as them) we have span classes on every cell. The end result is code that is impossible to read or debug which completely dwarfs the actual Content of the page.

Now stop and think for a minute. If we can't read the code or easily find the content then how easy will the search engine Spiders find it. When you thought about that one how about this - if the page has no structure then all text is as important as all other text. So if someone with visual impairment comes along with a screen reader or other simplified browser and tries to make sense of your page, what are they going to find? And what happens if someone has their own built in style sheet or views the page with large text?

There's got to be a better way!

And there is.

Article Continued

If you are interested in having optimized web site design methods applied to your website simply contact us.

Search Engine Friendly design and Redesign Consultancy

Related Content
Free Search Engine Positioning Report