John Linton
I became inerested in 'web marketing' in the late 1990s and have been involved with it to a greater or lesser extent ever since I had my first, very modest, commercial web site put up in early 1998. Since then, and especially from the very first day of Exetel's existence as a service provider I ave spent a lot of time on developing my understanding of how to use web sites in every aspect of a commercial undertaking.
I spent a significant part of each day either thinking of ways of improving the use of the web in our current business or using the data provided from the various web tools we have in place to help make changes or different decisions. I find I use web derived staistics in every, not almost every, but every dscision I'm now involved in. This is partly because we create specific metrics to help in decision making but more importantly its the much more hard to capture and interpret 'trends' that I personally use to guide much of the decision making in which I' involved.
I still remember on my first, ultra modest web site the developer was very proud that she had included a 'hit counter' that dispayed the nuber of times the web site had been accessed. Ove the last ten years I''ve evaluated and used many different web metrics software packages and have become far more adept at using what are now very sophisticated, and very precise, web information 'gatherers'. For many years I lsted after something as sophisticated as 'Netmetrics' (a now defunct but at the time wonderfully powerful web hit analysis suite that went far beyond anything I had seen up to that time). I never had the money avaialble for that level of software so for several years I used a service from Webstats and that was very useful in the past and for the first year or so of Exetel's existence but, as I began to need more detailed information it was not able (at least it wasn't the last time I looked at it) to deliver what I wanted.
In the last days I was building Swiftel and in the early days of Exetel there was a pretty consistent correlation between web site visitors and new ADSL orders - 5% of vistors signed up for a service. That quickly fell after the first few months of Exetel's existence and is now a very tiny percentage of the 750,000 unique 'visitors' that came to the Exetel web site in December. So for the past almost two years I've been using different 'hit collectors' and looking at a much narrower range of 'visitors'.
This has helped change various aspects of the web site to simplify the 'paths' an enquirer takes to reach the information they are looking for (the 'golden paths' in the defunct NetMetrics web analysis software). It's aso allowed me to get a much better idea of what types of people are visiting the web site and by feeding the IPs used by peope signing up on the on line order forms a range of information is provided on the source and demographocs of people who sign up with Exetel and also various other classes of people who visit Exetel's web site on a regalr basis.
One interesting minor report I have been runing for over 18 months now is identifying various company's 'employee' address ranges and producing a series of analyses of how many people from various organisations visit the Exetel web site, how regularly and if there is any correlation between events (date/times I nominate) their visits and those events.
It will need a lot more refining to produce a truly accurate picture but it gives some intersting broad statistics over time.
One thing that has changed so dramatically upwards over the past 6 weeks that it's impossible to mistake is the increase in the number of 'employee IP addresses' that are now visiting the Exetel web site multiple times a day from Optus, Internode and AAPT - now averaging over 400% more than in November and October 2007.
I guess they've run out of ideas of their own?