Subject: Re: Larry Ellison on FSBs
From: "Ben Tilly" <>
Date: Tue, 25 Apr 2006 16:15:52 -0700

 Tue, 25 Apr 2006 16:15:52 -0700
On 4/25/06, Thomas Lord <> wrote:
> Ben Tilly wrote:
> >> That's a pretty funny use of the word "good".   We're saying the same
> >> thing in different ways.
> >
> > Here's a suggestion.  If it seems that I'm saying something really
> > stupid, then either I'm being a moron, or I'm not saying what you
> > think I am.  Rather than continually responding on the assumption that
> > the first is true, perhaps you might want to consider the second
> > possibility.
> >
> >
> What should I do if I think you and I are expressing similar ideas but in
> different ways, possibly not really understanding one another?

Well I think that we're expressing very different ideas.  The fact
that my use of the word "good" was so funny is part of that.

> In general, I think that it's part of the function of a list like this to
> rehearse fundamental ideas in lots of different terms.   Repetition and
> re-expression evolves into a good cannon.   Tragic if misunderstandings
> leave bad feelings.
> Please, I didn't mean in the slightest to suggest you were saying anything
> stupid.  Quite the opposite.   We might disagree over the trivial issue
> of what should be considered obvious vs. what is not but.... geeze, that's
> a minor thing.

If anyone considers Christensen's thesis obvious, then I disagree with
them on the meaning of the word "obvious".  In retrospect?  Sure it
makes sense after it is explained.  But it isn't obvious by a long
> > Which of the following do you consider to be BAD management practices?
> >
> > - Listening to customers.
> > - Pursuing lines of business that make money.
> > - Avoiding lines of business where you're losing money.
> > - Giving money to suborganizations that you're trying to grow.
> Those aren't management practices.   Those are abstract ideas.  Those
> are things to think about in management -- they don't determine any
> particular mgt. decision.   Those are design patterns.

And they are design patterns that work really well until they don't.

> For example, does "listening to customers" mean doing what they literally
> say?  Or figuring out what they need before they themselves know?
> Bad management (easily recognized in retrospect) is that which results
> in losing.

This definition does nobody good until it is too late to be useful. 
Anyways my point is that the exact same management strategy that
normally leads to increased revenue and increased profits will also
lead companies over evolutionary dead ends.  You can call that "good"
in one circumstance and "bad" in another.  But it isn't that the
management is good or bad, it is whether it is well-adapted for the
circumstance that you find yourself in.

> > The problem is that you'll create an organization that is sized wrong
> > for the target opportunity.  You'll also likely create a dependency on
> > your regular cash infusions.
> Right!  That's why I want you in office/gov't!   Not enough people there
> get that!

Well you may want me there, but nobody else would.  Including me.

> > (Also without a business plan, you have no idea whether there is a
> > good way to spend $5 million in a year.)
> Ah.  Well, with an extra zero you've got more money than needs to be
> spent for
> the scale of the problem.  With a smaller zero you've got less.    Those
> are pretty
> lean and mean figures.  Some recombination between competing mgrs may very
> well be the best strategy in this game.  It's kind of a genetic algorithm.

The problem is that with too much or too little money for the
opportunity, you're guaranteed to lose.  With too much to spend you
can't find the real opportunities interesting enough.  With too little
to spend, you won't be able to match your more effective competitors.

However it is generally harder to invest too little than too much. 
Because successful businesses frequently start with a supply of equity
that looks suspiciously like a personal line of credit.

If your pockets are deep enough, you could just wait until someone
succeeds and then buy them.  But when you do, be very careful to not
accidentally destroy that which makes them succeed.  (Unfortunately,
any cashout event tends to do that with many companies.  Because key
people prefer being in startups, so they walk out the door to start
another job, and you lose their knowledge and ability.)