On Fri, 2008-03-28 at 18:32 -0700, C. Savinovich wrote:
> Let's put it in even simpler terms. True story by the way: Spain cellular
> sells for 19cents per minute the cheapest anyone can find (this is only an
> example)... So I calculate I can resell for 23cents per minute. But then
> everybody tells me that they prefer to buy from X, who was selling at 18
> cents. It took me time to figure it out, but now I know: I billed at 6
> cents increments and they billed at 60 seconds increments, and probably
> their minutes lasted 55 seconds. It worked for them beautifully since
> average length of a call on that route was 6 minutes.
>
> Conclusion: the cheapest quality routes charge 30/30 or 60/60 to cover for
> the cheaper price you think you are getting.
I would put that in the same category as all tier 1 dont do 6/6 or 1/1.
It may be true for some, but I dont think that its true for all (nor do
I believe the reverse of that). It depends on various factors, how you
interconnect, where in some situations, what plan you can convince the
sales rep to give you, etc. The same carrier may do 1/1 and 6/6 but
charge different rates for each and the rates may be so different that
1/1 isnt worth it.
You also said that 19 cents is the cheapest anyone can find, but that
someone is offering it for 18 cents, so by finding the provider that
offers it for 18 cents wouldnt that mean that you found something
cheaper than 19? Ok tangent safe to ignore this paragraph :)
If you look at it this way, statistically speaking 50% of the calls will
be in the lower 50% of the bill increment and 50% of the calls will be
in the higher 50% of the bill increment. Yes statistics dont always
match the real world, but it works as a base guage, and generally they
should match up with a large enough sample.
Assumptions made: the carrier is billing in 1/1 increments. The itsp
is billing 60/60. Cost is 19 cents, its sold at 18 cents.
50% average slack time per final minute means that 9 cents of slack
money is generated.
With a 6 minute average the charge to the customer would be 0.18*6=1.08.
The charge to the ITSP would be (5.5 minutes given the 50% statistical
average listed above) 5.5*0.19=1.045, for a grand profit off that one
call of 3.5 cents.
If you had 20 minute averages you would see a loss of 10.5 cents. For
each 20 minute call you need three 6 minute ones to make up for it. The
break even point with these numbers would be about 9.5 minutes, above
that you lose money on the call, below that you get a minor profit.
Yeah do a few thousand calls per day at the 6 minute average and you
might be able to pay your bills, 5000 calls is only 175
[dollars/euro/etc] per day, which really isnt enough for rack space,
bandwidth, office space, support staff, etc. You would be lucky to
break 5000/mo off that.
Now if you are big enough and doing enough total volume to enough
different places and the math is similar you may be able to make a
reasonable profit, making up in volume what you dont get per call.
Smaller operators would have a problem just paying their bills let alone
salaries however. Larger operators can generally negotiate better rates
as well, which in turn makes them more profitable however they generally
have more overhead which can reverse this trend.
You also have a huge exposure, if the 6 minute trend goes up, even by a
minute, you will see a huge downturn in your income (about 16%), and you
are really close to the border where you lose money on each call.
In short you dont make enough to sell below cost by doing that unless
you only get short calls or your initial increment is much longer than
the average call (like telecom usa with their 20 minute first increment,
and the average voice call when they advertised heavily was 4.3
minutes).
Where slack billing should come in is for gravy. If they charged 19
cents to beat your 23, they would never lose money (on the call itself),
and would make a small amount per call, in this example it would be
about an average of 9.5 cents per call, far better than the 3.5 cents
above. Do a few thousand calls per day and the gravy pot starts to fill
up. It is likely that they have a deal somehow somewhere for a cost
closer to 18 cents, perhaps with a direct connect to the carriers,
perhaps something else.
Now if you offer a "regular" and a "premium" service level where the
only real difference is the bill increment and priority (ie regular can
be dumped in lieu of a premium call, or just wont go through if there
arent enough channels available, etc) you can bill regular at carrier
cost using slack to make some extra, where premium is billed at a higher
rate, so the profit comes from the additional amount charged.
In this way you get customers who are willing to pay more for better
quality, as well as the ones that arent willing to pay more, and instead
of having gobs of idle time on your system you turn a slight profit.
Now if you also had a lower level than regular you could sell idle
capacity potentially below cost (total cost, bandwidth, servers, support
staffers, as well as termination charges) to lose $500 is better than to
lose $1000 so getting something in *some* situations can be better than
getting nothing. When it comes to phone minutes (and network bandwidth,
and ...) if you dont use it at this moment in time its gone forever, you
can never recapture minutes that have already passed.
The rapidly falling costs of bandwidth, rack space, etc is making the
last option less desirable because the costs involved to actually
operate have dropped. They will never reach zero, but in time they will
approach zero.
It used to be that it took millions to get even a small infrastructure,
and operating costs per day were quite high with all the skilled labour
that was required. Those carriers are more likely to give "best effort"
quality just to slow the attrition of funds from their bank accounts.
New providers, especially ITSPs, are far less likely to see any value in
that.
--
Trixter http://www.0xdecafbad.com
Bret McDanel
Belfast +44 28 9099 6461 US +1 516 687 5200
http://www.trxtel.com the phone company that pays you!
_______________________________________________
--Bandwidth and Colocation Provided by http://www.api-digital.com--
asterisk-biz mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-biz