John Linton's Personal Musings

I read this article with interest earlier this morning:

http://business.theage.com.au/broadband-price-war-looms/20071129-1dsa.html

As you may have noticed from previous comments I've made here Exetel has been attempting to find a wireless broadband service that we can sensibly add value to for some considerable time. We have had several discussions with Vodafone over the past 12 months whose network we use to provide mobile services to Exetel customers. We've also had very frustrating and eventually pointless discussions with two other mobile carriers and are currently in longer term discussion with mobile service aggregators in the USA and Hong Kong.

None of the discussions we've held to date have produced any tangible results and while it seems possible that we may be able to do something with at least one of the 'foreign' aggregators it will take quite a bit of time. So I was interested to see the first realistic 3G service hit the market as from next Monday and will be even more interested to see the response of Telstra and Optus to that 'break through' price/download combo from Vodafone.

Personally, I think that $A39.00 a month for a 3G broadband service with 5 gb of downloads and a no charge modem is easily the most attractive broadband offering made available this year - certainly gets my nomination for best broadband offering of 2007.

I'd like to know what allowed a 3G provider to so suddenly change its pricing for data over 3G from ridiculously unaffordable to more affordable than the best wire line services available so suddenly. Was it a new technology breakthrough or was it just a 'stroke of the pen' decision by Vodafone sales and marketing? I'll never know of course.

So Vodafone has established that $A8.00 per gb downloaded (in the eyes of the customer) with no set up fee and no modem cost is the way to go in Australia in 2008. I think that would set the benchmark for 70% of all broadband decisions made in areas where there is VodaFone coverage. If I was concerned before about how tough a year 2008 was going to be this Vodafone announcement consigns all my previous concerns to 'background noise'. Not just the Vodafone announcement in isolation but the inevitable responses by the other 3G carriers and (assuming they actually do have the network bandwidth to connect huge numbers of customers) the 'sea change' that would bring in terms of the preferred technology to deliver data services.

It doesn't take much imagination to see that Vodafone et alia will be serious contenders for any new Federal funding for broadband to the bush and increased coverage of broadband at faster speeds generally. So there isn't just the spectre of a Labour/Telstra FTTH impact but now a very real technology shift that is logically what is really needed in the next decade.

I guess it's time to give our Vodafone contacts another call and see what is possible for wholesale customers in 2008.

....Send In The Clowns.

I've had to deal with some less than enjoyable situations and issues over the last few days but I cheered up immensely and actually laughed out loud twice when I read the Eftel CEO's address to the Eftel shareholders which is published here:

http://www.asx.com.au/asxpdf/20071127/pdf/3162mw1nqmtk3f.pdf

I'm not sure what intellectual abilities the Eftel shareholder's who attended the meeting actually possess but whoever wrote the CEO's address obviously didn't think it was very high in making some of the most extraordinary claims I've ever seen relating to a public company's performance.

Then again - if you have a great deal of negative issues in your annual figures I suppose you still have to try and say something positive even if you are going to have to see what you said in print for the rest of your company's existence.

I thought on reflection after I re-read the address that "clutching at straws" was way above what was said to the Eftel shareholders - though perhaps you might read the address differently and maybe I just didn't grasp the subtlety of what was said. I thought the following statements were absolute gems of the most incredible business statements of 2007:

1) "The 31% upsurge in sales has been a standout performance, which compares especially well with other listed companies"

Does the Eftel CEO think so lowly of the shareholders that he believes they are functionally innumerate?

According to the annual report sales revenues have increased over the previous year by around $A8.5 million - from $A25.972 to $A34.4. Mathematically that is, indeed a 31% increase in booked revenue. However it isn't an "upsurge in sales"; it simply reflects that Eftel bought the failed aaNet and added the aaNet revenues from the second week of July 2006 (ie. for effectively the full year).

Now I realise that aaNet's public statements before they went broke were highly suspect about their number of customers and their monthly revenues but Eftel did pony up $A1.6 million to buy a business that was bankrupt and, if there were "around 16,000 ADSL customers" as stated then the annual revenue from that number of customers would be around a little less than $8 million.

So, as I read the figures, Eftel's revenues may have increased by around $500,000 on around $A34 million - "not exactly an upsurge, Bob"

2) "This year, we became the first Internet provider to announce the planned rollout of, nationwide, VDSL2".

Now really this is a major insult to anyone's intelligence!

Nationwide - Telstra has around 6,000 exchanges. Eftel "plans" to roll out (perhaps tip the egg cup would be a more appropriate phrase) a whole 80 DSLAMs - how is 80 out of 6,000 "Nationwide?"

But that pales in to insignificance compared to the claim of being able to deliver VDSL to ANYONE by any nominable date as the VDSL standard will not be defined until February/March next year and the most likely time that any end user could be connected to such a service would be September/October. No-one else is offering VDSL because it's incapable of being delivered - and may never be delivered.

To trumpet that "this is the most exciting event in the history of Eftel" (announcing a service that doesn't exist and that when it does it won't be available from Eftel to very many people anyway) is incomprehensible.

And my final laugh, which made me spill my coffee, was a CEO of an Australian public company attempting to give credibility to his verbal hyperboli at a serious meeting by quoting a post on Whirlpool!

I think that said it all.

As is the case for most Australian businesses, Exetel's planning is done on the basis of a financial tax year of July 1st to June 30th. Partly because we started the current form of the Exetel business in January 2004 many of our contracts run from January/February which means this is a particularly busy time of year in terms of contract renewal discussions/negotiations although many of our contracts have end dates spread over each month of the year due to 'interim' changes.

Exetel is, by no measurement, anything but a small/medium company in overall Australian terms based on its current revenues, numbers of customers and number of employees; and in terms of the main telecommunications suppliers in Australia - Exetel is very small indeed. However the annual spend we have to manage and contract is over $A30 million a year and the contracts are very serious indeed as they, effectively if not contractualy, are guaranteed by the personal assets of the owners of the business. While that has never been a problem in the past and isn't a problem at the current time it is something we are beginning to address in discussing new contract terms with each of the providers to the current business.

Another thing that we have made a point of in this 'round' of contract discussions is that we are no longer prepared to enter in to anything but short term contracts for the supply of services. If a supplier wants a contract of longer than 12 months on any monthly spend greater than $A20,000 then we are taking the very firm position that we aren't going to use those services beyond the end of the current contract. This has caused a problem for at least two providers who want to insist on a longer term. Personally I'm not interested in some account manager/sales manager's desire to increase the "bookable" value of Exetel's business to them and, as a reasonably responsible decision maker in charge of the disposition of our family's life time savings, my view is that the changes that will become ever more difficult in 2008 make any contract longer than a few months a serious liability.

Because I view 2008 with a measure of misgiving and because some of the current Exetel suppliers view it as their 'right' to tie Exetel up in long term fixed price contracts (something that when we started the company we had to accede to as we had no track record and absolutely no 'bargaining power') I think we are going to have to make some quite hard decisions regarding what services we continue to offer beyond the end dates of the contracts due to expire in 2008. As we have never leased any of the equipment we use to operate the business we have no exposure, in terms of ongoing lease/HP payments on the hardware used to provide the services, if we decide to stop offering a service. In fact, as by far the largest 'investment' in hardware is in Cisco boxes, we would make a capital return if we stopped offering some services at some future time.

I think some of our current suppliers have been more than a little surprised at our position of not considering re-signing with them on anything but a month by month basis when their current contracts with us expire; I think at least one has expressed their surprise in what I would consider to be somewhat inappropriate terms. I have attempted to explain to each of the suppliers we are talking with that our requirement not to enter into further specified term arrangements other than month to month is due to the likely continued volatility in pricing of services to end users by both Telstra and Optus as well as the uncertainty of future decisions by the Federal Government and the new Federal Government's relationships with Telstra in terms of funding and availability of new delivery infrastructures.

As we firm up the business plan for the second half of 2008 we are finding it prudent to base those plans on a completely different service offering mix to that which we currently now base our business on. If all my concerns turn out to be just 'jumping at shadows' then, perhaps, nothing will change and Exetel's service offering mix in the last half of 2008 will be pretty much the same as they are now. If things do take a negative turn, in terms of Exetel's aspirations, then I want the flexibility to be able to react appropriately and not to be 'locked in' to contracts that give Exetel no hope of making decisions that it may need to make.

I seem unable to make these points with a number of suppliers and possible suppliers to Exetel in a way/ways they either want to understand or are even prepared to accept are sensible for both parties. I think part of the issue is the constant change in account representation that has been the case with almost all of our suppliers over the time we've been in business - we lost the services of the two remaining 'good' account managers (one to maternity leave and one to 'following the boss to his new company') in the last month. So we find ourselves dealing with yet another new 'account manager' with whom we have no rapport or empathy and therefore the consequences of that.

There must be an easier and more enjoyable way of spending the time between now and Christmas.

I met yesterday with one of the two suppliers that we have been considering for MSANs in some capital city metro areas to advise them that we would be deferring proceeding with their proposal until at least February of next year, and depending on what transpires between the new Federal Government and Telstra we may not proceed at all. I decided to do this in person as I'd known the CEO of the company for over 30 years and felt bad about wasting so much of their time.

They were disappointed but more understanding than I had expected (they had put a great deal of work in to the proposal and discussions over several months) but said that they had expected the deferral and two other proposals they were working on had been similarly been put on hold. We had a 'Christmas' drink of a very nice 20 year old single malt I'd never heard of (in their beautifully appointed board room) and wished each other well for the coming year.

Before I'd driven back to the office I had a call from the sales person who had done most of the contact work saying that would I reconsider the decision to defer the project if they could find a way of reducing the purchase price "substantially" and deferring all payments for six months and if there were any adverse actions by Telstra/the new government in that time they would accept return of the equipment with no charges made and they would also pay out any Telstra rental early termination charges.

To say I was surprised would be a massive understatement, I was totally overwhelmed, so I thanked the sales person, explained that I was still driving, and said that I'd get back in touch with them later in the week.

I'm still bewidered at both the speed of the changed offer and the magnitude of discount to the pricing and the unbelievably generous payment and cancellation terms. Having thought about the implications over night I'm still completely baffled as to why such a rapid change in a pretty serious, at least for Exetel, business proposal could/would be made and in such an informal manner. Presumably its a very minor deal for the supplier concerned in terms of money and they must be very confident that there is no real chance of the equipment being returned. Even so the deferred payment and the almost 25% price discount seem quite extraordinary.

Perhaps its yet another sign of increasing age bringing less confidence because, 'in the old days', I think I wouldn't have hesitated to accept such a favourable 'negotiation win' and would have turned the car round and gone back and signed up without a second thought. Now - I'm not sure that I will do that and, in fact, it's made me so uncertain that I can't even think what the right decision making process is to make a sensible decision.

Looking at the new numbers it is a 'no brainer' and it removes all of the costs of a six month ramp up period to transfer and gain enough customers to reach a comfortable ROI; but.......

Maybe it's time for Exetel to have a decision maker that doesn't look gift horses in the mouth and has the confidence to make major decisions based on thoughtful analysis and sensible assessments?

I think I'm beginning to lose those abilities.

I briefly mentioned a few days ago in these musings that I had decided that the time had come to take the next step in the management of Exetel not just from my point of view and needs but to better meet the requirements of every person who works in Exetel in any position (plus the needs of our customers and suppliers in some circumstances). We began this project last Tuesday and Wednesday by having two brief meetings at which we discussed the concepts and outlined the five major aspects of the end system.

Since more people have begun to give some thought to the basic design of this system the concepts that I loosely described have begun to firm up in to a true GURUS:

"Grand Universal Report on Unwelcome Scenarios"

I re-listed the number of reports/graphs/numbers that I look at each day in fulfilling my roles at Exetel and came up with a little over 300 separate data sets (now there's a word I haven't used for a very long time) and I went on to add another 100 or so that I was pretty sure other people within Exetel would also look at on a multi-daily basis - there may well be more than that.

So the current aim for the GURUS is to display a single number on two or three screens around the North Sydney office and from a screen able to be displayed on any PC in the world at the top and then a single number against each of the ten different "departments" that comprise Exetel.

If the number is 'zero' then there is nothing happening within Exetel at that moment in time that requires 'attention'. Clearly the company number couldn't be 'zero' if one, or more, of the ten departments registered a number other than zero. The 'departments' displayed on the screen would be 'ranked' based on the number being dispayed at that point in time.

The number itself is derived from measuring over 400 seperate 'data sets' in the various devices that Exetel uses to run its business. These devices and the data base include:

Over 50 Servers
Over 16 LNS, Core and Border Routers
Over 100 GigE accesses and cross connects
Cerberus help desk servers
IronPort spam filtering servers
Allot P2P filtering
PeerApp P2P Caching
Mitel VoIP PABX
Asterisk messaging servers
Activities of over 65,000 customer services and over 5,000 'in progress' customer transactions
etc, etc

The whole 400, or in a department case some lesser number relevant to their direct responsibilities, are aggregated against a set of user selected times and values (the values reflecting the relative importance of the individual 'data set'). The main 'number' will be updated each second (or some other user defined period) based on the data collectors retrieving the current values from each of the nominated 'data sets'.

If any number is 'out of the acceptable range' then the indivdual responsible for that part of Exetel's operations will be 'messaged' telling them which report to look at (via a 'click here for details').

Exetel has recently initiated detailed job goals for each of its personnel and zero defect programs for each of its departments. It's the intention to use the GURUS data collectors to also report (not publicly) against the agreed job goal performance for each person within Exetel and the progress of each department's Zero Defect program.

While, poorly described in the way I've done above, this may seem like some Orwellian 1984, cloud cuckoo land inflexible, and therefore, useless attempt at substituting computer reporting for 'real' management - it isn't. I, and many other people within Exetel, constantly refer to these hundreds of reports and graphs every moment of the day and it's becoming very time consuming.

This GURUS attempt is just a logical step along the way to give each Exetel person additional assistance in exercising their decision making abilities and providing extra capabilities and the original concept was based on me thinking about how I could reduce my information reviewing workload.

I'm quite excited about seeing just how much of this concept we can bring to reality.

....."Cheer up - it could get worse". So I cheered up and sure enough........

I guess if you allow stupid people to make decisions then you get to live with the consequences of stupid decisions. While I realise that most of the people who voted for Labour couldn't be expected to have any idea of what it would mean economically for themselves and everyone else how could they bring themselves to vote to have that inanely grinning, vacuous, moonfaced image made part of their daily lives? As my gran used to mutter in such circumstances 'there's nowt so queer as folk". (of course in her usage, "queer" had no connotations other than what the OED gave as its meaning in the 1950s).

Being a sensible person I was taught to 'play the hand you are dealt' a very long time ago so business life goes on pretty much as usual though I'm now faced with the inevitable uncertainties that mean that each decision will take longer and, almost certainly, will result in either no decision or at best much less 'adventurous' decisions for some time. Unless you're a Union 'official' I guess everyone's in the same position so it's a little like still playing poker but for lower stakes.

I'll give it a month or so before trying to make a sensible assesment of what influence Telstra has bought with their 'campaign donations' to the Labour party which will be something of a disappointment for two companies we are in the final stages of signing contracts with but, then, that's their problem to deal with and one that they'll share with a whole lot of other people. So I can say first hand that the change of government has taken around $A4 million out of the economy immediately (and maybe forever) and three immediate prospective employment positions are on indefinite hold. I doubt these minor decisions by Exetel will mean much in themselves but I also imagine that they'll be replicated to a greater or lesser extent in many businesses of our size around Australia this week and going forward for some time - perhaps for some considerable time.

Just keep in mind the statement by the failed pop singer who's now our new Minister For The Environment that defines the ethics and competence of government in Australia: "we'll promise whatever's necessary - and then change it once we get elected".

OK - I'm done with the pointless and pathetic whining.

The really good news over night was, of course, that Manchester United lost to Bolton and Arsenal won.

I spent some time yesterday trying to finalise the ADSL2 pricing structures for December and came up the best that Exetel can do given the current issues and, now, the degree of uncertainty in any major investment decision. I've finished the work more or less but one thing I noticed, and I'm undoubtedly biased, is just how difficult it is to vist an Australian ISP's web site and try and work out what you have to pay and what you actually get. With perhaps one exception it took me a long while to understand whether or not I needed a telephone line and if I did who supplied it and at what cost and then what charges would be made for the calls.

Maybe its the introduction of "naked' ADSL that's introduced yet another variable not addressed clearly enough on so many web sites. I've looked at combining all of the ADSL1 and ADSL2 and the new 'naked' ADSL2 pricing but it seems to be far too much information on the one page. Maybe there are too many plan options which I'll look at reducing. My major concern is that I'd always intended to base an ADSL/VoIP service offering on the premise that Exetel could always control the costs from wholesale carriers by taking out the insurance of a flexible infrastructure deployment. With the result of the election yesterday that may well no longer be an option and I haven't had time to consider whether the whole 'naked' ADSL project will have to be put on hold.

As I have other, personal, things to do today it will have to wait till Monday I think.

Then of course:

http://www.middlemiss.org/lit/authors/obrienj/boreelog.html

We had our monthly board meeting yesterday which we usually hold over a lunch or dinner at one of the many nice places to eat in Sydney - the only benefit there is in being a director of Exetel. These meetings are usually very pleasant with even difficult issues and situations being discussed and dealt with without casting too much gloom or for very long. Over the years we've only had three situations that could be described as serious which is not too bad over a four year period.

Steve made a comment during the engineering report/discussion that didn't have the impact on me at the time as it did when I was dealing with some other issuers earlier this morning. The comment that now 'gives me pause' was "there will be more changes made to the Exetel core network over the next six weeks than has occurred in total over the past 18 months".

Exetel has continued to develop/change since we first 'went to market' almost four years ago as any growing company will obviously do. However, Steve's comment made me consider what the past few board meetings had authorised in terms of additional capital expenditure and new infrastructures and it all adds up to quite massive changes in the key elements of the ways we deliver services.

We activated new PoPs in Queensland and Victoria in the middle of this year relatively painlessly and without any significant changes to the way we delivered services in Victoria and Queensland other than obtaining some financial efficiencies for Exetel and shortening some transit paths for users in those States. Although there was significant cost involved it wasn't anything like what will now begin to happen and continue for the next 6 weeks or so.

We had already begun to build out a second Sydney CBD PoP as it seemed to make sense to have a PoP in the Verizon data centre as much of our current wireline and VoIP connectivity is via Verizon. The new changes and developments planned by Steve mean that the second PoP will become also used for data termination and switching as we will begin to terminate carrier data carrier interfaces in Verizon to duplicate the ones we currently have in the Powertel datacentre.

The first of the major changes will be the duplication of the Telstra Wholesale GigE circuit that's used to terminate all Exetel's NSW ADSL1 customers. This is currenty done via ne GigE circuit to Powertel's data centre with 900 mbps of VLANs. Some time 'soon' a second GigE from Telstra will be terminated in the Verizon data centre and we will split the NSW ADSL1 traffic using 500 mbps VLANs on each bearer. This in turn means adding and redploying half of the LNS routers used for these services as well as adding multile GigE cross connects between the two data centres.

We also will now deploy the P2P caching engine and storage, probably some time early next week, which is another major change to how the core network is configured with also a need to add and re-cable key routers and switches.

Almost simultaneously we need to upgrade our VoIP switches and double the number of PRIs used for VoIP traffic in December and then double the PRIs again in two steps in January and early February and at the same time ad new routers and switches to the new Melbourne and Brisbane PoPs as well as activating a PoP in Perth at abut the same time.

While we're doing this we also have to provision and deploy new ADSL2 services using "naked" ULL and we are launching a calling card product sold via retail outlets in three capital cities in December.

A frightening amount of change for a small company but almost certainly only the start of what wil be required in 2008 if we go ahead with a dslam or wireless 'roll out.

Then there's today's election and the uncertainties that a change of governent will introduce in to the communications service business (and all of the other aspects of business of course).

If you read this and vote in Australia please don't vote against the current government - I don't think I can handle the concerns that would bring.

I continue to be confused at just how to make this service available to Australian users in the most practical way and at a sensible monthly price for both Exetel and a customer. I would like to make the offering more attractive in terms of price and features than the current iiNet offering (not particularly hard to do) and the recently promoted Telstra $59.95 for 12 gb broadband and telephone line (and now it turns out it's really only an offering on cable and if ADSL is used then the promised 8 mbps speed becomes 1500 - or so it seems to be).

Then there's the major considerations of what the more popular 'independent' ISPS such as TPG, AAPT, iPrimus and Internode will offer in the next few weeks or so.

I think I have all of the information to make a 'final' decision over the weekend and will be able to start offering the service in the not too distant future based on the following, rounded calculations:

ULL monthly rental - $11.00
DSLAM cost per port - $6.00
Exchange space rental per user $2.00
Back haul per user - $4.00
International IP per user - $12.00
Admin/Siupport/Depreciation - $3.00

giving a total monthly cost of providing a 12 gb/48 gb service of around $38.00 per month ($41.80 inc GST).

The cost calculations above are the costs once the planned number of likely connections are made which will take 9 - 12 months. In the early months the cost per service will be considerably higher at around $53.20 per 12 gb customer.

So the question to be answered is do we include low cost/free telephone calls in the offering from the beginning or do we take the view that for those people able/willing to use VoIP already have a VoIP service (and therefore the VoIP equipment of their choice) and they won't really want to give up their current DID(s) and other VoIP numbers to get some free local and national calls?

Tough call.

I think offering very low cost calls to mobiles and international calls (instead of the very high priced VoIP calls on the iiNet service for instance) would allow end users a better choice as to whether they wanted to use an Exetel VoIP service versus someone else's VoIP that they may already have.

Anyway - I'm pretty sure that the price points are right in terms of meeting what I'm guessing at being the likely prices from future 'naked' ADSL2 service offerings. While I obviously can't know what pricing other ISPs will have available to them I'm pretty sure that the base costs of terminating a ULL, renting exchange space, leasing DSLAMs and back hauling the traffic are unlikely to be more than a few dollars different and I'm VERY sure that they will have no better economies in all other aspects of offering such a service.

So I think the numbers currently on the web site, with some minor changes to some of the download allowances, will be the final pricing whatever we decide in terms of included or not included calls turns out to be.

It will need a 're-arrangement of all current ADSL2 (and perhaps ADSL1) plan prices to allow for a 'logical price structure of:

1) ADSL2 including line rental = 'Ceiling Price'

2) ADSL2 no telephone line with $10.00 of voip local and national calls = 'Ceiling Price' - $5.00

3) ADSL2 no telephone line with no VoIP calls = 'Ceiling Price' - $15.00

I was reviewing the proposed "Naked ADSL2" costings last night in light of a 'letter' we'd received from Telstra at our home address. As with everything from any marketing department it was interesting in what a strict legally invigilated marketing department like Telstra's is allowed to say compared to the restrictions that small companies like Exetel are allowed to say. That aside I was interested in the accommodations that Telstra is now making to the less intellectually gifted of their residential customers (presumably like Annette and me).

I thought the most amusing statement was the opening "major benefit" which read:

"Get speeds up to 8Mbps with BigPond Broadband, which won 'Internet Technology Of The Year in 2007'.

The qualifier, in small print on the back page didn't actually say who awarded this offer "Technology of the year" - it started with "based on Telstra tests.... which, while implying that Telstra made the award to itself, presumably really referred to the fact that you can't actually get 8mbps.....as I said.....only Telstra could get this sort of thing legally approved.

Anyway - I digress - what interested me was the fact that both iiNet and Telstra BigPond have obviously decided that $59.95 is the 'sweet spot' for residential internet users. This is good news in the event it proves to be true because it is the figure that Exetel and several other ISPs believe is the most likely acceptable cost for a residential user of telephone/internet services.

The key differentiaor in the Telstra offer is their delivery of 10 cent local calls via PSTN for the first time plus their provision of 12 gb of downloads as a base offer - finally acknowledging that 200 mb (so adamantly stuck to for so many years) isn't of much use 'even' to the very lowest of internet residential users.

What's more interesting is that Telstra's offer is vastly superior to iiNet's newly released "Naked ADSL2" offer at $59.95 and has caused me to, yet again, re-think the pricing and construction of a new Exetel offer.

Essentially Telstra provides up to 8Mbps/384kbps internet with 12 gb of peak downloads/uploads for $59.95 over a standard telephone line/number AND with 10 cent local calls and a free modem and install on a 24 month contract.

Compared to:

iiNet who provides ADSL2 speeds with 3gb of peak downloads/uploads at $59.95 on a 24 month contract over a non usable telephone line with 'unlimited' local and national calls if you care to buy a VoIP enabled device.

Telstra also give a $110.00 credit. Telstra's offer is also available to virtually every Australian user while the iiNet offer is available on their own DSLAMs.

I think the convenience of the churn option with the Telstra offer (its ADSL1) is probably also an advantage to the users/new users to whom this offer appears to be pitched.

Both offers have the major turn off of a 24 month contract and iiNet's offer of 3gb peak (which is more like 2 gb of downloads plus 1 gb of uploads) is not terribly usable to the average ADSL user who probably uses an average of more than 2 gb of downloads these days.

Based on past history TPG will offer the same as iiNet at lower prices but with the Telstra 12 gb of downloads and Internode will almost certainly offer something slightly better than the iiNet offer when they get around to doing it.

I think the Telstra offer kills all other ADSL1 offers with their retail price to a customer around $35.00 a month cheaper than the equivalent service to a wholesale customer. What Telstra Retail are now offering re-defines ADSL1 as a minimum of an 8mbps/384 service and has reduced the price of local calls to what the lowest cost VoIP providers offer.

In terms of the iiNet "Naked ADSL2" offer - that's a lot easier to deal with. I think the Telstra offer kills it stone dead and therefore any other ISP will be able to meet it/beat it with some sort of ADSL2 offer.

However life seems to just get tougher in competing with Telstra.

Since we started offering communications services to end users in January 2004 one of the most impressive, to me, things we have done is to build an internal reporting system that enables our very scarce 'management' resources to track every aspect of
the technical and commercial aspects of the business on a second by second basis. The current system was, obviously, built step by step as we began to install infrastructure and to sign up customers.


As we added different services that required different types of infrastructure we obviously had to expand the engineering controls for
those services as well as building the commercial controls needed to manage a continually growing amount of payments and receipts. We also had to comply with more and more, and more and more intrusive, legislation.


I start my day, every day of the week reviewing most/all aspects of how our company is performing by looking at over 60 different aspects of our business displayed to me via graphs or other reports. What used to take me a few minutes in the 'old days' now takes me over 90 minutes and, as constantly amuses Annette, involves me writing many columns of numbers throughout the day to keep track of how the day develops - because while we've done a truly great job of implementing reports and controls from our routers, servers and financial data base, the company's growing complexity has defied the rationalisation of the simple reporting structures we started with.


All the information is available, and I and other people, are still able to 'micro-manage' every aspect of the business with the
availability of real time data but it now takes some time because much of the data now needs to be concatenated differently than previous iterations needed to be done.


So, yesterday morning we held the first brief meeting to start the process of completely re-designing our internal management systems
along completely different lines to the one we've built piece by piece over the past 4 years (and which had their genesis in the consulting work done for successive companies for many years before that).


Our aim is to produce a completely new way of looking at the 'internals' of Exetel from multiple viewpoints:

- Senior management
- Operational management
- Each different individual staff member
- Our customers (for 10 different services)
- Our major suppliers


I looked at the data base management systems from the big three software companies where the concept of relational database and user based query languages has been 'sold' to major corporations for at least three decades and decided that they didn’t suit the purposes we had in mind and certainly not the budgets we have in mind. We've survived/grown using mysql and very bright and amazingly competent programming and systems administration personnel for so long and
they've produced so many exceptionally good tools that my belief is that we can do more, and do it more quickly, ourselves than investing half a million dollars in new software and tripling our programming staff - maybe I'll be proved wrong.


So I have high hopes that we can develop the software that will link our many different services provided via an ever growing number of
different hardware and infrastructure implementations (this year alone we've added Quintum Voip switches, Iron Port spam filtering, Allot P2P management and, shortly, PeerApp P2P caching - as well as increasing our PoPs from one to four) but do it in a totally innovative way that will reduce the current 'report reading/decision making time' back tothe few minutes at the start of each day it used to be and remove the necessity for manually linking those reports.


The final goal is to give everyone who's associated with our business much better information (that's relevant to them) than we do
at the moment.

I read this article with some interest yesterday:

http://www.itwire.com/content/view/15399/127/

As an avid 'devourer' of telecommunications articles from around the world I get a lot of information and ideas from what's published on various industry web sites and news letters; not just the facts that are always useful in some way but often the marketing "facts" are equally useful. I've been around long enough not to think too unkindly of what is ascribed to various people as the 'journalists' who do the reporting do make mistakes or mis-hear what is said to them.

The only 'true' sources of statistics regarding the number of ISPs operating in Australia and the number of broadband and other types of customers come from Telstra (who know down to the last user how many ADSL users there are in Australia and to whom they are connected and they don't tell anyone except themselves and the BigPond sales operations) and from the ABS who report on ADSL numbers but as they use questionnaires filled in by individual ISPs those figures can't be called reliable within probably around 20%.

The only other source is the individual annual reports of the various publicly listed ISPs and, sometimes, the press releases from such companies (though the press releases seem sometimes to be at variance with previous press releases and as they aren't subject to the same scrutiny the annual reports are they aren't really reliable).

The IT Wire article cites iiNet (a large WA company) and a small WA company, Eftel, as being active in pursuing the purchase of smaller ISPs to increase their customer bases. Fair enough - it's a 'strategy' that both make a point of in their annual reports and other published information. Not a very successful strategy for either company though based on their ongoing losses made since they embarked on this 'strategy'.

iiNet declare the sale to Vodafone of iHug in their current annual report (the proceeds from which sale seemed to comprise the majority of their annual 'profit') and if memory serves me correctly they seemed to sell iHug for less than they paid for it some three years previously.

Eftel make a note of two acquisitions in FY2007 which turned out even worse than the iHug acquisition for iiNet. In FY2006 Eftel reported a $A799,000 profit and, after acquiring aanet on 10/7/06, they reported a $799,000 loss in FY2007 (allowing for the tax adjustment) - not a great acquisition - paying $A1.6 million for a user base that lost a further $A1.5 million in the first year you owned it. It seems the aaNet customers were losing an average of $8.00 a month each at the time of acquisition (at least) so why pay anything at all for such customers?

Other information in these annual reports also indicates that further acquisitions might not be the way to go. Maybe iiNet could borrow more money than it already has after borrowing big time to pay for its acquisition of OzEmail and other 'investments' but a quick 'banker's look' at the iinet balance sheet as at 30/6/07 shows that there is only a tissue paper thin positive difference between current assets and current liabilities. The same situation is shown on the Eftel balance sheet with a major deficiency in the current asset/current liability ratio. However I may be misunderstanding the summary figures shown there - although they seem pretty hard to misunderstand.

The other curious figures in the Eftel annual report are what seems to contradict the statement in the IT Wire article - "Eftel have around 40,000 ADSL customers after the acquisition of aaNet doubled their customer base". Eftel reported revenue had increased in FY07 over FY06 by around $A8 million ($A26 million to $A34 million) which included revenue from aaNet from 10th July 2007. While I have no knowledge of the exact situation of the defunct aaNet's ARPU it would be reasonable to assume that it was around $50.00 per month per customer (and I only base that ARPU on what Exetel's ADSL1 ARPU is). If this figure is anything like correct then it would suggest that the number of customers that aanet had on acquisition was something less than 15,000 which the IT Wire article quotes as "doubling the Eftel customer base" and the annual report justifies the acquisition by describing aanet as "fast growing". I think aanet always wildly exaggerated the number of customers they had before they went broke but I hadn't realised by quite how much.

So iiNet and Eftel, both from WA, are going to further consolidate the ISP population of Australia - I just wonder where they intend to find the money to do that and, based on previous experience, why they would think it's a good idea.

I suppose this concept of growing a customer base by buying customers from another company makes very little sense to me - it's clearly my problem in grasping the financial benefits - pretty much like my inability to grasp the value of advertising. If I wanted to grow the Exetel customer base by, say 40,000 ADSL users, I think it could be done for less than $50.00 a customer over a period of 6 - 9 months which is probably half the cost and pretty much the same timeframe as trying to buy 40,000 customers from other ISPs - and it's so much safer.

I noticed in one of the weekend paper inserts that Optus retail appears to be 'ending' its promotions of some of its recent bundling offers. There are so many of those offers that it's hard to keep track of them and this current "offer ends soon" could simply be the pre-cursor to even more attractive offers in the standard "run up to Christmas" advertising splurges so beloved of marketing departments.

I also had a look at the recently announced iiNet "naked" ADSL plans yesterday afternoon when I was in that very pleasant frame of mind that comes after an excellent leisurely lunch overlooking splendid views and in good company. As with everything in advertising it seemed to me that all any marketing department does is simply copy what some other marketing department has done (exactly) and to cover the aspects they have financial difficult copying they just obfuscate and , at least it seems to me,they obfuscate to the point of lying.

I don't see how some of the statements made about iiNet's "naked ADSL plans on the iiNet web site have got past their legal department's scrutiny - but presumably they have so I didn't worry about it other than to be amused at how stupid marketing departments think the human race is - perhaps they base that on what they see in their make up mirrors?

My view of 'naked' ADSL is that it's going to be an advantage to a customer who makes very few PSTN/ISDN telephone calls and either uses their mobile for that purpose or uses VoIP - it seems a good opportunity for someone who wants ADSL but doesn't want telephone call services on the line that is necessary for ADSL to also have to pay a rental for a telephone line they won't use - I guess even a marketing person can grasp that simple scenario?

There are three problems that have to be overcome to make a 'naked' adsl offering attractive.

1) Convince the carrier that 'owns' the telephone line from the end user to the exchange where your DSLAM/MSAN is located to sell it to you for a realistically low price - in which negotiation you will get significant help from the Telecommunications Act and the ACCC.

2) Ensure that the end user can make telephone calls cost effectively (either by using their mobile or by you providing them with an unbelievably simple to use VoIP service).

3) You also then have to price the 'naked' ADSL service so that it's more attractive than a 'normal' ADSL service that is offered by other people (including yourself).

1) is pretty much a no brainer thanks to the recent decisions of the ACCC and you can get a 'naked' ADSL line for around $A10.00 a month in some realistic volume spend.

2) is solved by companies such as Optus by using their low cost buy price of ULL to continue to offer 'unlimited' local and national calls (and calls to Optus mobiles) for no charge having shifted the cost of the telephone line to the ADSL service charge but by giving the free calls the service looks like "naked" adsl but more attractive. For iiNet its 'solved' by the marketing department obfuscation (....just plug your current handset in to your modem.....). I guess the iiNet support desk will get to explain to the customers who try that they should have bought a VoIP compatible modem and made sure their handset is compatible with that modem - maybe then the iiNet legal department will look up who approved the advertising for the service?

3) Now this is a little tricky as neither Optus nor iiNet's offerings actually seem to be at a lower cost than 'normal' ADSL (though it's true that the Optus bundled free telephone calls makes this a cost judgement based on the call spend of each individual).

As Exetel continues to slowly increase the number of ADSL customers who also use our VoIP services as a percentage of total customers we are interested in dropping the requirement for the customer to pay either us or another provider for a telephone line that is basically not going to be used. Our current barrier to doing this is that Powertel don't offer such a service and Optus charge, effectively, the same price for an ADSL service with a 'live' telephone line as they do for just an ADSL line - yes - not a typo or mis-print - that is what is charged (at least to Exetel).

I've never been able to work out why Optus do this but, once again, some marketing department obviously came up with the concept which, I assume, made some sort of sense 2 years ago but has been overtaken by events and will continue to be increasingly uncompetitive - at least for companies like Exetel.

Fortunately the combination of the pleasant lunch and a perfect Sydney summer late afternoon allowed the positive aspects of this strange situation to overcome the irritation factors and I think we will be able to come up with a solution that allows Exetel users to gain the key benefit of VoIP which is much reduced call costs and the saving from not paying for a telephone line they don't really need any more.

So after a bit of 'number crunching' I came up with these plans which I think will address the current issues while allowing the customer to actually use their current telephone exactly the way its currently set up with no change of number or line or anything else:

plans data line speed peak + offpeak total cost

Naked 1 ADSL2+ speeds 2GB + 48GB $45.00
Naked 2 ADSL2+ speeds 4GB + 48GB $55.00
Naked 3 ADSL2+ speeds 12GB + 48GB $65.00
Naked 4 ADSL2+ speeds 16GB + 48GB $75.00
Naked 5 ADSL2+ speeds 32GB + 64GB $85.00
Naked 6 ADSL2+ speeds 64GB + 64GB $111.00

All plans would include unlimited local and national calls with low cost international calls and calls to mobiles....and of course "not playing with the big boys" nor having a 'marketing department' it won't be necessary to weasel out of delivering the apparent downloads by including uploads in the data usage calculations!!

I must get the costings verified so we can 'launch' the new plans by the end of the week.

(I must get 'legal' to check I haven't somehow mis-lead any potential customer with the claims made or 'inadvertently' breached anyone's copyright!!!)

Looking at the available figures about ADSL and broadband take up in Australia, and they seem to be at best incomplete and some just look plain rubbery, there is a 'feel' to those reports that the take up of broadband in Australia has slowed over the past year or so.

I base this view on the available figures plus two other Australian factors and some recent data from the USA.

The first is what I see as almost 'desperation' in the ongoing advertising activities of Telstra and Optus and, to a lesser extent, other companies who spend money on advertising. I don't think I could count the number of times "free" was used in the weekend papers either in ad space or on brightly coloured inserts.

Whatever you may personally feel about the use of advertising I don't think that so many companies spending so much money on so many give aways indicates a booming broadband market. I understand that some people say when things are booming "make sure you get your share by increasing advertising spending" but I don't think that 'they' also say "and by the way sell everything at a loss" to make sure you benefit from these boom times.

The second indication comes from what my acquaintances in the modem business tell me about their unit sales which are lower than at any time they can remember which is another indication that (despite so many ISPs giving away modems) the number of new DSL take ups is lower than its been for a long while. I give credibilty to these statements as they are backed up with ever lower prices for modems which seem strange to me as Exetel isn't a big buyer of modems. Even at our low volumes we are being offered prices that are 20 and 30% below what was being offered less than 2 months ago.

If this in fact the case then I'm encouraged. A strange thing to say? Perhaps - but the reason I'm encouraged is that it means that more peope will already have ADSL and therefore will have real experience to judge whether their initial ISP actually delivers what they want. It also means that Telstra and Optus will lose the advantage of being abe to give away modems and the cost of new activations (around $140.00 per new customer) which a small company like Exetel can neither reaaly justify investing in the finance nor would ever make enough money on the monthly plan charges to ever recover the cost even if we invested in the finance.

A third 'straw in the wind' is in reading that the US broadband market has slowed down quite appreciably and there are, according to some commentators, real reasons to think that the broadband market in the USA has reached saturation point. As the US situation generally is a little in advance of the situation in Australia these figures, assuming they're accurate are very revealing.

You can see these articles and the related links here:

http://gigaom.com/2007/11/13/3q-2007-broadband-report-card-b-minus/#more-10672

which basically say:

"For the third quarter in a row, U.S. broadband growth lost steam - UBS Research’s John Hodulik in a note to his
clients pointed out:

Roughly 53% of the nation’s 111M occupied housing units, or
58M homes, had broadband at the end of 3Q. However, only 77% of
households have a PC (June 2007, CEA), and we estimate 90% of homes
currently have access to broadband service. Assuming the 77% of homes
are evenly distributed among that 90%, then 69% of the 111M homes, or
76M homes, have a PC and can get broadband. This puts penetration of
eligible homes at closer to 76% and suggests the days of wireline
broadband connectivity as a major growth driver of U.S. telecom are
largely over.

I attended the Australian Deloittes "Fast Technology 50" awards lunch yesterday. I seldom go to these 'networking' functions but I figured this one would be useful if only to hear the key note address. This proved to be a correct assessment with the key note speaker (a person who'd built a company from ground zero and sold it within ten years for $A300 million) giving a good address and making some telling points.

A lot of the information provided by the Deloitte speakers and the actual "Technology Fast 50" publication was also of value and interest as it was always likely to be. It was also good to spend a little time in an 'atmosphere' and with people completely different to those you spend every second of the day with - with the best will in the world you have to lose objectivity in doing that.

As at all such functions I got to talk mainly to a Deloitte staff member but, for once, I got some realy good ideas and was impressed enough to request a follow up meeting to see if it would be practical for Exetel to put in a novel version of an employee share scheme program. We have been trying to find a way for our key personnel to share in the profits made by Exetel, now there is some real prospects of there being some real profits to share, and I'd never heard of 'shadow share' schemes and other versions of profit sharing programs.

The other thing that impressed me, and you may well say "no sh** Sherlock", was the fact that the real money to be made by small companies, such as Exetel, is by not doing what we are doing - re-selling someone else's services and giving almost all the money from our activities to those few suppliers while carrying all the risk and working so hard for a bit better than a break even resut. There were so many examples of that in the one room it was a little bit overwhelming.

So many of the companies there were started up with a good idea and if they sold something then all of the money was gross profit as the 'product' cost nothing other than the time and effort of turning the idea into something someone else would pay money for. One of the original intentions of building Exetel was to develop the management software for both the technical network functions and the actual major business 'indicators' that would, of themselves, become a saleable 'product'

By and large, over the past four years, we've done that with not really that much to be done to 'prettify' our main control and management tools in to a suite of highly advanced but very easy to use functions that would assist any other company better manage its network and server functions and, as far as I can see, the major 'health' issues of any commercial enerprise.

So, and ths may appear very odd, I almost certainly scrapped all of the in progress planning on new projects for 2008 (which have been part of the development of the Exetel business over the past 4 years, in a time of less than 15 minutes and reminded myself that the original objective of building a 'perfect company' wasn't to become and remain a reseller of someone else's services but to sell our own unique servces using 'Exetel' as the 'demonstration product'.

Doubtless there will be some 'middle path' to be found but, at least for the moment, I am again more closely focussed on what attracted me to going through, one more time, the hellish events, back and mind breaking working hours and the nightmare problems that are invariably involved in creating a new company from a 'greenfields' starting point.

I came away from that lunch more focussed, encouraged and happier than I've been for some time....and I haven't said that on many Fridays over the past four years........

As we get closer to making decisions on what, if anything, Exetel will do in 2008 to continue to meet the needs of our customers and as our VoIP services continue to operate at levels equivalent to the quality of PSTn and ISDN services we also get closer to making a decision on a "naked" ADSL services (or a telephone line that doesn't have a dial tone and is used only for data services (which these days mean also for VoIP telephony.

I read the recent product releases by Amcom and iiNet with interest and believe they will be relatively successful in the limited exchanges in which they offer services. However they didn't appear to me to be particularly aggressive - and maybe they don't have to be at this stage.

Now that I know a great deal more about the costs of providing dsl from an 'owned' dslam in a Telstra exchange I was surprised at the iiNet pricing which seemed very high to me given the costs available to a much smaller company like Exetel.

As far as I can see the costs of providing a 'naked' dsl service from your own DSLAM would be around $26.50 (inc gst):

$9.50 for the Telstra ULL
$8.00 for the backhaul per customer with unlimited data (dropping to $4.00 per customer over time)
$6.00 for the lease of the DSLAM port (dropping to $3.00 over time)

at the low numbers of dslams that Exetel could finance and would be far less than that for the numbers that iiNet or Optus or TPG would have already deployed. To that you would need to add approximately $1.20 for each gb of data provided which at iiNet's low end offer of 2 + 2 would cost around $3.00 per customer at most.

So it's hard to see how such a service could cost more than $30.00 a month (inc GST) for a small company like Exetel on 'day one' and this cost would be expected to steadily fall to around $23.00 per customer over 12 - 18 months.

It would seem to me that an 8 gb plan on such a service with 'unlimited national and local calls' could be offered for around $45.00 and would make a higher profit than any current ADSL2 or ADSL1 plan.

Of course such plans could only be offered where the ISP had its own DSLAMs and my estimated back haul costs are major city metropolitan.

Exetel will see what our current carriers can offer in terms of a 'naked dsl' service over the next 2 - 3 weeks now we know what the actual costs to a provider are for delivering such a service but based on the current prices from Optus (around $50.00 per month) we doubt we'll have too much success in that direction.

Who knows? Maybe Telstra will wholesale a service that can be provided at a realistic price as their costs would be a fraction of the ones I've used in the above calculations. However it's difficult to see Telstra offering a SSS ADSL2 wholesale service 9as they don't offer it via BigPond. By mid 2009 there will be more pressure on Telstra to wholesale ADSL2 which, possibly they might consider doing if for no other reason than to make life more difficult for those ISPs who have invested in their own DSLAMs.

Then again, it is also going to be interesting to see what sort of wholesale market there is going to be for broadband services by the end of 2008 - my prediction is that almost all the 'krill' will have gone broke or been absorbed in twelve months time and the market that Telstra and Optus will have for 'wholesale' of broadband services by mid to late 2008 will be a fraction of what there is even today.

Those 'minnows' that have invested in their own DSLAMs will be caught between the pincers of a Telstra/Optus wholesaled ADSL2 and an as yet unknown new fibre broad band with 3G/4G wirless tearing part the highly profitable lower end of the broadband market by that time.

Maybe a viable 'naked' broad band remains the only 'space' for the minnows?