John Linton
Exetel constantly reviews it's service plans and prices as part of it's core business raison d'etre of providing the lowest cost (to the end user) communications services available from any supplier. I spent much of yesterday analysing our current costs of delivering ADSL1 and ADSL2 in terms of the changes in usage that have occurred since I did the last major review in early December 2007. I do this each year at these times as this particular time frame presents the largest set of differentials of any period of the calendar year. Broadly speaking there is a noticeable drop in overall usage (both customers connected and amounts downloaded and uploaded) in late November/early December followed by a slight rise in mid December followed by a steep drop in the last two weeks of December followed by a steady increase throughout January that lasts until schools return in early February where the increase in usage 'platforms' before beginning a much slower increase that continues until the next late November.
Over the 2007 year this was pretty much the case. However, as those of you reading this who are Exetel customers can see from the total IP usage bandwidth monthly report here:
https://www.exetel.com.au/members/a_mrtg_trafic_total.php
the pattern in January 2008 differed from the three previous years in not showing steady growth throughout the month but remaining steady for the first three weeks and actually declining in the fourth week of January. What you can also see is that the 'average' IP bandwidth usage increased from around 900 mbps in January 2007 to around 1,650 mbps in December 2007. This is a key figure as it indicates, very dramatically, that the average Exetel broadband user is using 50%+ more bandwidth that an average user of 12 months ago.
I've commented earlier in these jottings that a surprising (to me) trend has been that TPG has become the most popular ISP to whom Exetel users go when they churn away (matching and sometimes exceeding BigPond as a destination more often than not most days). I took the trouble to analyse what sort of customers were leaving Exetel and found pretty much what I expected - 95% were on 1500/256 high end plans with an average monthly download in excess of 80 gb. So it's understandable that such heavy down loaders would react to Exetel's reining in of their 100 gb+ downloads in off peak time to go to an ISP who offers (apparently) 150 gb for $80.00 a month - though the TPG fine print ought to cause some hesitation. This situation would contribute to less bandwidth being used - over time and would partially account for the unusual January usage.
If the calendar 2007 continues throughout calendar 2008 then it's pretty clear that the 'average' Exetel broadband customers will use more of whatever downloads are included in their plan than they did in the previous year and therefore the bandwidth required is unlikely to grow any differently than it has done over the previous 12 months - effectively growing by almost double (taking in to consideration that any benefits of the NetEnforcer's ability to smooth bandwidth usage in the future will be significantly less than it was in 2007).
So we will need to progressively add bandwidth in 2008 from the current levels (from all sources - P2P cache, Akamai cache, P2P filtering, Pipe peering and direct IP) of just under 2 gbps to something approaching 3 gbps based on current estimates of likely broadband customer growth.
The good news is that the cost of the direct IP bandwidth will fall by around 40% which will ameliorate the cost of buying an additional 50% over the coming twelve months but the net cost of IP bandwidth will not allow any reduction in end user pricing as the actual 'average' cost of IP bandwidth per customer will rise slightly.
The really bad news is that much of the growth Exetel is experiencing is coming from ADSL2 services which have a far more expensive cost of connectivity bandwidth (the link between the end user and Exetel's PoP). This, more than the cost of IP bandwidth, will almost certainly not allow any reduction in ADSL2 pricing and may well result in the price of ADSL2 services increasing.
So I looked at the various cost components of the various ADSL1 and ADSL2 services and, unless there is a significant monthly port rental reduction (which personally I see as not very likely) then the only likely movement in Exetel's broadband prices in 2008 would be upwards - not downwards.
Exetel's 'over heads' (personnel, office accommodation, utilities etc) are already at the lowest levels that are possible (as far as we can see) and our forward expense line planning shows that there may be a slight increase on a per customer basis as we continue to modify how we operate the company towards the end of 2008. There is no realistic chance that even this minor component of the cost of delivering broadband services will fall in 2008.
The mooted reduction in direct IP bandwidth costs will in fact be the only factor that gives some chance of Exetel's broadband pricing remaining the same - at least thats how it appeared to me yesterday after spending much of the day looking at figures in spreadsheets until my vision told me it was time to quit.
So I've decided to do nothing in terms of pricing for the time being but will begin to address the issues of monthly port cost rental and customer connectivity bandwidth in the coming months. I don't view this as having a particularly good chance of success in itself but, perhaps, in combination with a move away from wire line broadband to GSM broadband there may be a new 'dawning' of reality in some of the current carrier's understanding of what future services and pricing will need to be to retain their current business levels let alone grow the current levels.
Don't hold your breath though.