John Linton ....seems to be different to previous years.
One thing that seemed anomalous in the latest ABS figures was the huge jump reported in average usage per customer over the past 12 months.
http://www.abs.gov.au/AUSSTATS/abs@.nsf/allprimarymainfeatures/6445F12663006B83CA256A150079564D?opendocument
While I very much doubt that the figures provided by any of the ISPs (except Exetel) are particularly accurate I did wonder what the reported huge, getting on for 100%, increase over the past 12 months actually meant - assuming it was in any way accurate. Did it simply mean that the average internet user was just ramping up their daily usage? That is a possibility but what changed over the past twelve months to make such a huge difference? Is it just that the majority of ISPs (including Telstra Retail) have inflated their download allowance so much over the last twelve months that a relatively small proportion of end users are encouraged to download so much more that it distorts the 'average' figure?
It would be very useful to be able to understand the download bases of any customer base and track it more forensically than at least we do. We have always used relatively crude analyses of usage by customer and don't understand any of the 'whys and wherefores' of usage at all beyond percentage bands of usage across the whole customer base. What we do know, but unlike what the ABS statistics seem to show, is that all of the download growth is within the top 10 - 15% of the customer base with the bottom 85% still downloading at the rates they did three years ago. It is quite hard to understand why this would be the case and it should be changing in line with the facilities that are available. Why this isn't yet the case can be surmised but not proven.
In stark contrast to residential services, business services downloads do not seem to grow at all. Most business services provided by Exetel have unlimited or very high included download allowances and the trend in our business sales is to sell higher and higher speed connections - 100mbps is becoming quite 'common' as a business service - but the 'average usage' per customer per month remains where it was more than two years ago. Perhaps this will begin to change over the coming year or so but, right now, there is absolutely no indication that will happen.
Although the cost of pure IP has fallen significantly over the past two years and, as far as I can see from early indications in the 'negotiating season, is likely to continue to fall, the major 'saviour' for the increased residential usage is the caching provided by Google, Akamai, Pipe and other peering points which is now approaching almost 50% of total IP bandwidth presented to customers. The cost of this bandwidth is sub $10.00 per mbps compared to 2 to 3 times that for pure IP. As the pure IP price continues to fall so do the peering costs which has lead to an overall drop of almost stair step proportions over the past 18 - 24 months. Interestingly enough (despite MS and other major software caching) business users don't seem to use the cache as much as residential users).
The question that we are attempting to deal with is what happens if the 'average' residential user (at least as described in the ABS statistics) doubles their usage over the coming year. An indication of why that might be an issue was Internode's statement that they were having to raise some plan prices because their customers had begun to use "all of their allocated allowances" It sounded wrong but you understand what was meant. Is that happening to 'more aggressive' offerings from other providers? The next two ABS reports might be quite interesting.
Copyright © Exetel Pty Ltd 2011
ABN 350 979 865 46