Subject: Re: HBS WK: Who will win Microsoft or Linux?
From: "Stephen J. Turnbull" <>
Date: Tue, 14 Jun 2005 12:37:59 +0900

>>>>> "Kelly" == Kelly Anderson <> writes:

    Kelly> I term it Goldilocks software... not too horizontal, and
    Kelly> not too vertical.

This doesn't quite ring true, because Goldilocks applied her rule
along a single dimension, but "vertical" and "horizontal" aren't
opposites here.  Also, the Goldilocks metaphor is already in
widespread use as a rule of thumb for price discrimination.  It turns
out that for a wide range of consumer[1] products the following holds

    Given prices LOW for a visibly lower quality, but reasonably
    usable, version of a product, and HIGH for a higher quality
    version, a large fraction of customers choose LOW.

    If to this mix you add a medium quality version at an intermediate
    price, as long as the medium quality version is perceptibly
    different from the other two, and the MEDIUM price is sufficiently
    high (but considered lower than HIGH by the customers), the MEDIUM
    price will draw a lot of customers who would otherwise choose LOW,
    but very few from HIGH.

Exploiting this by using three versions instead of two is often called
"Goldilocks pricing" (see Shapiro and Varian, _Information Rules_, for
example).  For the kind of customer affected by this, there's weaker
evidence that "too much" choice can result in confusion, so many
customers may return to the minimum satisfactory quality.  (Both
studies of actual market data and "marketing focus group"-style
experiments show these effects AFAIK.)

This is of course especially effective in IT, where "value-subtracted"
quality differentiation is so easy to accomplish.

    Kelly> If a piece of software is too widely appealing, like word
    Kelly> processors, Operating systems, web browsers, etc., then it
    Kelly> is a prime candidate for a widely supported open source
    Kelly> effort. If it is too vertical, (as in CRM, payroll, the
    Kelly> sorts of things that Perot Systems, PeopleSoft and the like
    Kelly> go after) then large companies will pay people to work on
    Kelly> open source efforts to increase the bottom line elsewhere.

I think there's a lot to what you say, but especially for the "very
horizontal" class the size of demand is a countervailing influence.
Although many software firms, especially Microsoft, are known for
trying to crush their competition, that doesn't mean that there isn't
substantial profit to be made by getting to the market earlier with a
better product, then getting out later, if and when the market gets
commoditized.  (Shapiro and Varian call this the "3M strategy".)

[1]  Here, consumer products are those which have a simple interface;
the user does not need to be technically savvy to be able to use it,
and therefore is also likely to rely on rules of thumb to respond to
price/quality comparisons.

School of Systems and Information Engineering
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.