On Fri, 24 Jan 2003, Gordon Lennox wrote:
> But the customer is a slightly more complex beast than that. They may want
> to pay as little as possible. They may want to think they are getting good
> value (and not being ripped off). But they may also be happy to pay for
> something they don't intend to fully use. They may also be happy to pay a
> premium to get certainty in billing.
All of the above I agree with.
However I don't think your examples support your point... I think you're
saying usage/consumption, when you mean value-extraction.
> posting a letter (don't tell me that it costs more to post a letter across
> the country than it costs to send one across town<g>)
It does, in the United States, now, for first-class postage. That change
occurred less than a year ago, _instead of_ a general increase in rates.
Also, you only pay when you post a letter (consume resources) and not if
you don't post one.
> a magazine subscription (when you don't always read the magazine)
You pay for exactly the number of issues and amount of postage and amount
of sales/advertising effort (thus the volume discounts for multi-year
subscriptions) that you receive. Thus, you pay in proportion with the
resources consumed. Regardless of the amount of value you extract from
> a TV license or cable subscription (during those magic weeks when
> you are on holiday away from the box)
You pay a flat rate which amortizes the costs of the service-package
you've bought, whether or not you're able to extract value from it. If
you choose a basic package which doesn't require the provider to return a
lot of royalties to the studios, you pay less. If you want movies, you
pay more. If you want pay-per-view, you pay a lot more. If you want
voice or digital radio services, you pay incrementally for those as well.
> local phone calls (at least in one country we know<g>).
Yes, that one supports your point, but even that isn't universal... You
can pay a preminum to get flat-rate calling, but the cheapest plans don't
> I do not have the figures anymore but the cost breakdown of a mobile call
> still probably shows a surprisingly large chunk devoted to tracking who did
> what with whom and where and for how long.
Typically about 60% of the cost of the call.
> If a business has the choice then I do not see the choice being
> to invest heavily in detailed billing of something whose price is dropping
> so fast. You could be quite vulnerable in the market to somebody who decides
> instead to invest in providing much more bandwidth.
Yes, in some ways I agree with this as well... In that if one has
credit to expend, overbuilding with a faster, dumber network will always
give a large competitive advantage in the short term while your own
network is empty and other people are already trying to figure out how to
pay for upgrades. However, this is admittedly a short-term strategy, and
if everyone followed it, we'd have ruinous turnover in providers, since it
depends upon being able to bankrupt one's competitors to succeed.
And I agree that expending a lot of money on detailed accounting systems,
rather than investing it in ways of providing increased _value_ is an
unfortunate waste. However I don't think Patrik is advocating anything
complicated, just a simple two-or-three-bin system. L3 Asia did a
nine-bin system with reasonable success for a while. I've been advocating
a two-bin system for many satellite-dependent large ISPs in small
countries (flat-rate unlimited usage at gigabit speed for local traffic,
but per-bit metered for use of the satellite terminal).