[1264] in Commercialization & Privatization of the Internet
Usage based billing
daemon@ATHENA.MIT.EDU (Kent W. England)
Thu Aug 29 09:36:56 1991
From: "Kent W. England" <kwe2@BBN.COM>
To: stan@karazm.math.uh.edu
Cc: com-priv@uu.psi.com
In-Reply-To: <199108290239.AA01439@Menudo.UH.EDU>
Date: Thu, 29 Aug 91 09:31:14 EDT
>From: "Stan Hanks (bcm)" <stan@karazm.math.uh.edu>
>
>... it's ... hard to justify a "large" fixed monthly expense
>*if you have no prior history of usage*. .... Having
>relatively inexpensive usage-based billing will make it easier ...
>to sell a connection internally. Once you *HAVE* the connection, and
>actually start using it, it will become quite clear in relatively
>short order that a fixed-price connection is the right thing to do.
>It's just that initial jump-start where you really need usage-based
>billing.
>
I agree pacifically, but I would note that another alternative would be
to have a speed dial on your connection that is easily changed and a
billing based on that speed. The speed would not necessarily be a
hard kind of limit (like a fractional T1 connection) but might be soft
like the SMDS and FR leaky bucket algorithms or Rick Adams' "soft"
fractional T1 service, that accommodates some burstiness. But easily
changed and easily understood in terms of service offered and received.
Start slow and crank up the speed as need and perception of value
increase.
The main disadvantage of usage-based billing is the uncertainty of cost.
And the client's cost is the service provider's revenue, so the
uncertainty is on both parties in a usage-based or estimated pricing
plan. Avoiding risk due to uncertain expense/revenue leads to
mechanisms to offer flexible service without uncertainty. Ie, speed
dials. This feature (and ESF) is why I love fractional T1.
But I would reiterate that usage sensitive pricing may have its place.
It all depends on what the parties in the bilateral negotiation want.
--Kent
