This article occupies that grey area between sensible use of SEO techniques and bending your web site out of shape, to take advantage of every possible edge you can get with the Search Engines.

The theory goes and it makes sense, that Search Engine Spiders read your page in the order it is presented to them in the source code of the page. Thus to increase your chances of a good ranking you should have your keyword dense introductory text near the top of the page.

But when using a typical Mamo template this does not happen. In particular the left hand column of your template which is full of largely useless textual content, will come prior to your page content.

The solution to this is to alter the template:

As an overview you could consider a section of a template as following this format:

<table>
   <tr>
       <td>Left content</td>
       <td>middle conent e.g. the page content you want to see first</td>
       <td>Rigth content</td>
   </tr>
</table>

You want to create a page with the middle content first, so you alter the template thus:

<table>
   <tr>
       <td valign="top"></td>
       <td valign="top" rowspan="2">Middle Content</td>
       <td valign="top"></td>
   </tr>
   <tr>
       <td>Left Content</td>
       <td>Right Content</td>
   </tr>
</table>

The exact tweaking will vary from template to template.

Is this sort of tweaking really worth it? We think if you are seriously trying to get the best out of your pages, the answer has to be yes.

You can use the robot meta tags to stop web spiders used by search engines from visiting and indexing parts of your web site.

I think the first reaction of many webmasters to this topic, is to think

"what",

"why should I care if the search engines index, the lot, it can only mean more hits for me from somewhere right?"

This view is understandable but may be mistaken on at least a couple of grounds.

• Some web sites, are a lot bigger than they look and spiders have to stop somewhere. If they didn't they could literally go on for ever as page content is dynamic and with parameters so are page urls. A site can actually be impossible to spider fully.
• Search Engines may actually down rate your site on the basis of pages devoid of sensibly searchable criteria.

To take marsjupiter.com as a case study, we run some statistics for the Distributed Computing Team, Team Ninja the statistics cover hundred of our own members and hundreds of competing teams, acres of measurements on how our members and teams are doing across a number of DC projects.

The searchable interest of a lot of these pages is negligible, but the number of linked page runs into thousands.

Thus we think it a wise idea to tell the search engines not to index certain pages and not to follow links on those pages.

You do this with a meta tag of the form:

<META name="Robots" content="noindex,nofollow">

valid values for content are:

Along side the robot meta tags, search engines can also be instructed via the robot.txt file.

This may seem an obvious point, but it is a trap that is easy to fall into.

It is all very well being thorough when developing your pages, about checking links. But the web is a far from static place, and it is easy for errors to creep through.  This applies both to external and internal links.

For external links, a program such as Xenu's link sleuth is recommended.

For internal links the control panel on your site, should have an error log facility. Check this log regularly, you may be shocked by what you find.

With all the hype around SEO Search Engine Optimisation it is perhaps a good idea to stand back and consider some of the "facts".

The first fact that needs considering is the much vaunted statistic that something like 85% of traffic to web sites is delivered via Search Engines.

This is a quite astonishing figure and in common with all hyped statistics it is worth considering a few questions:

• How was this figure collected?
• Who paid for the survey?
• How current is the data?

However despite the possible exaggeration of the numbers, it should be clear from simple observation of those browsing the web, that Search Engines are responsible for one hell of a lot of web site visits!

So, clearly you need to put a lot of energy into SEO on your site! Well maybe, but maybe not. There are actually a few factors to be considered:

Will people find your site via the Search Engines anyway?

It is going to depend a lot on the nature of your site. If like us, you are stuck with a site that wants to see a lot of people finding us, by typing the terms "Meta Search Engine" into a Search Engine, then you have your work cut out for you. There are hundreds of sites using such terms competing to be seen.

Conversely if you run the "Sellefield Sea Swimming Society" then so long as you bother to give your site the slightest relevance to the subject and submit your site to Search Engines, then you will hit the top of the listing for people searching with these terms without having to bother with SEO at all.

Should I concentrate on paid advertising?

Paying for advertising especially where you can ensure that your site appears as some form of "sponsered" link above or along side search results, is a way of side stepping SEO.

This may well be the pragmatic/only route if yoo want to be visible in highly competative areas.

We really laugh at those SEO adverts that promise top ten listings in Search Engines. There are after all only ten top places for any keyword combination and the competition can be fierce. You can only promise a top ten place if in the small print you insist the customer settles for a relatively obscure set of keyword on which the placement will be judged.

How will traffic to your site build?

The 85% figure for traffic directed to sites by search engines even if taken at face value, is only an average. Sites will receive some of their traffic through:

• Link Exchanges
• Revisits
• Being the homepage
• Paid Advertising

The key here is revisits, some sites, a case in point being Ninjamicros are largely based around a core of people who regard the site as their "home on the net" in the case of Ninjamicos this is built arouind distributed computing teams. Other sites such as BBC News will be regulariy revisited as a reference point. Sites like this can at least to some degree assume that traffic will build slowly of its own accord based on the excellence of the site and the degree of resource it offers.

Conversely there are some reference sites such as the Code Project that whilst they offer truly great resources, are liable to be very Search Engine dependent. It is likely most people reach the Code Project by searching for an answer to a particular programming question. 

Conclusion

We would conclude by returning to the beginning:

"the much vaunted statistic that something like 85% of traffic to web sites is delivered via Search Engines. "

85% that is 6 visits out of every 7 made to a web site!

Given statistics like that, can you really afford to ignore the issue of making sure that you are getting your fair share of those visits?