(well pounds as I am writing from the uk, but you get the idea.) If you had the right data you could log into the CIA network. If you had the right data you could bring down ebay. If you had the right data you could email everyone in the world? But where’s the line - what data constitutes a waste of hard disk space - a pointless entry?

You could quite easily watch every aspect of a users session on your website, in fact I am fairly sure using AJAX you could record every mouse movement when on the site, but to draw the line isn’t easy. I personally stick with google analytics, hit tail and my personal coding. Knowing the percentage of users that have javascript installed is a good example of good data, knowing how many of them are using internet explorer over firefox - also good data, but thier ip irrelevant of use - not so useful.

The conversation about what constitutes worthwhile data is an extremely relevant point, as technically it can make money or loose money. Popular blogs with good content updated often (new, steady datastream) are usually the more successful.

However at this point I look at data specific to the creation of Microsites.

Perhaps a bit early in the runnings I would like to explain a project I instigated last week. The concept of microsites is not a new one, the idea being that you create a site irrelevant to your base site, aimed at a niche or long tail phrase that either acts to increase your linking in the sector or work or to work as a secondary to your web presence.

Anyway If you have the right data you can feed thousands of pages with unique content and hopefully establish a microsite that both provides a modest income in the form of adsense ( or similar ) revenue and provide good linkage too.

So heres the process:

1. Register a hosting account elsewhere to your normal provider - (different class ip is the start of your google hiding tricks)
2. Register a good domain with your keywords in it - e.g. cheap-wheel-nuts.com if you are a wheel nut seller. Obviously they might have all been taken but just try and get your key words in there - use typo’s if very popular topic.
3. Find data - There are lots of ways to get data - depending on your market it might be a bit difficult, but as I have found out alot of industries have industry bodies and other third parties that have legal requirement to provide database data for public use. Another way is potentially to use data protection act to aquire data. - The key is the data hasnt been mass published before. - The more entries the better. The two examples I found with relative ease had 3150 entries and 43000 entries respectively, on each entry there was 5 - 25 fields, plenty to feed pages.
4. Build a site using dreamweaver or something simple. Keep it all simple. Basically make a 1 page template that provides a slot for adsense and linkage to your main site - then maybe 1 header image and a main space for content.
5. Build your content into the pages dynamically, combining clever php and .htaccess mod re-write.
6. Install pagex (a module i am developing - is in alpha), hit tail and google analytics (if you want), google adsense for ad revenue.
7. By using a max of 10 dynamic pages you should have a small battle ready site. - Next use a crawler to make a google sitemap, and using a new google account tell google about the sitemap.
8. Make sure the pages are optimised, use completly long tail titles that are absolutly unique between pages.

I am keeping a bit hush hush about this because I am still testing it. Whether or not the data is suffiently different enough to everything else that google wont disallow it, or what we will see. The data which I have used for my example has 43000 entries, is uk based and is distributed as a legal requirement of the industry. Each line of the 43000 has 20+ fields of data which effectively act as unique information.

Grand total of pages when put together 70000+ pages. Spam you thinking? I don’t agree with Spam in any form, but with this example it isn’t. The information is actually really useful and although its a long way of doing it, it does act as a service. Although a simple search could have been done using only 2 pages, this way does the same but gives me 69998 more unique pages.

In total thats 4 unique pages, 3.5 hours work, roughly £45 in registration/hosting fees but potentially 70000 niche pages, lets just hope google doesnt black list it as spam!

Too be continued….