From: David Ensteness (denstene_at_mac.com)
Date: Mon Apr 19 2004 - 07:14:24 PDT
Joel,
On Apr 18, 2004, at 3:05 PM, Joel M. Sciamma wrote:
> Eric,
>
>> Installed base is what matters, pure and simple.
>> You don't sell software to a market share percentage, you sell to an
>> installed base.
>
> When 98 of every 100 computers sold are not new potential customers,
> for
> your software or your virus, it makes a difference.
Depends on your market, I talk about this more below, but it comes down
to this: Apple's overall market share might be smaller but most
developers develop software for specific markets so the question for
them has little to do with Apple's total market share. It has to do
with Macintosh market share in their market.
> The installed base is also gradually shrinking due to the age of the
> equipment and more fragmented, IMO, due to OS X.
Mac OS X has been a huge success. More Windows users than before, more
first time buyers than before, more developers than before, more
applications than before,
> Most new releases for Mac OS are OS X only - the economics of
> developing and
> testing for 9 and X simultaneously don't look too good - most shareware
> authors have made the switch, quite rightly. FileMaker (Apple) is now
> OS X
> only. Most other core packages are too or shortly will be. Does
> developing
> for 9.x make any commercial sense despite the much larger installed
> base?
No, however that is a different argument than the one you make for the
Mac OS X market being irrelevant. Developing for Classic Mac OS does
not make sense because it is no longer a developing platform. The
Classic Mac OS market can't increase because you can not buy new
hardware, you can't buy new OSes, you can't increase or maintain the
market size. None of those things are true of Mac OS X. The Mac OS X
installed base is continually increasing in size while the Classic Mac
OS market is shrinking.
> So, the installed base we are addressing for Mac OS is effectively 12M
> users, a handy potential market where you can be a bigger fish in a
> smaller
> pond, but growth will be harder to sustain.
Where do you get this number you keep citing?
> If your software is cross-platform, what would you say is the point
> where
> maintaining a team to code/test/support for the minor platform becomes
> uneconomic? 10% of total revenue? 5%?
I think you are jumping from general to specific here, its a bit of a
fallacy. The video editing market is dominated by Macintosh, the
publishing market is also dominated by Macintosh, therefore lets look
at say, Adobe in publishing. Adobe gets sales from: bundled software
that comes with scanners and such, business licenses, consumer sales,
and education sales.
Now, overall the Macintosh market share is 2.5-3%. Installed base is
closer to 10% based on 30 million Mac users. What is Adobe's biggest
cash market? I assume its the business market, aka publishing firms.
Those are dominated by Macintosh, so that would mean that while their
total sales might be higher to Windows users, the largest single market
may very well be to Mac users.
Dig deeper.
The same may be true of many other areas. Until just last year Apple
was the largest single brand installed in the education market. There
were more Windows PCs there, but no particular brand had more market
share than Dell. Apple is now fighting with Dell for education market
share, but in terms of installed base they are holding on. Market share
does relate to installed base over long timelines, but one is not the
same as the other.
Also, even in your example, do you think you could convince a company
to throw away 10% of their earnings? A buddy of mine coaches diving
part time. He is pretty burned out and only does it a couple times a
week. Says its really a waste of his time, but thing is he does it at
this point because he does need the extra income. Is it a waste of his
time? No, you know why? Because the money he gets from doing it is more
than he would make if he didn't do it.
> I have a new product that is wholly developed on the Mac (9.x) but
> hosted by
> a cross-platform (9.x/OSX/Win) CAD package. The Mac used to be a real
> force
> in CAD but is being gradually rendered irrelevant - all my customers
> and
> prospective customers use Windows - how much of my limited resources
> do I
> commit to Mac-only features?
I disagree with this. I have connections to a large number of landscape
design architects who do upper residential, large scale developments,
and commercial design work. They all use CAD software, brands vary as
do platforms. Macintosh is not the dominant platform, but it is a very
large contingent making up somewhere around a third of the designers I
am familiar with. Some run private shops, some run mid-sized firms.
PowerCAD is extremely popular and just went Mac OS X native, I used to
do consulting for several of these firms and I received word from a lot
of clients that they were very happy to jump to Mac OS X now that their
main app was native. A few had been hold backs but most were running
Mac OS X except when drawing.
Also I have to note that before 1999 I new very, very, very few Mac OS
CAD users, CAD was generally used on an NT platform across the board.
> If it were 2, or even 5 in every 100 customers, would it be worth it?
>
> A lot of good people committed themselves to coding for the Newt and
> were
> badly hurt by its cancellation. Was their motivation a tiny installed
> base
> or the promise of a serious and growing market share?
According to this no one should code for anyone but the biggest market
share holder and the largest potential market. This simply does not
make sense though in practice. Car makers fight to be in the sub $20K
and sub $15K car market. Not one of them makes money in those markets,
Saturn has never turned a profit on its own, its subsidized. Is this a
bad strategy? It sure seems dumb on the surface ...
But ... thing is that they want the return sale. Maybe you buy a Saturn
when you are just out of college for $14,000. GM loses money ... but,
when you get married and buy a new car maybe you spend $22,000 and they
maintain you as a customer. Then once you have some kids, need a truck
to haul the boat, an SUV, a wagon, whatever ... personal experience of
not, most car buyers in the course of their life stick with the same
car maker through most of their life. The lesson? Go get em early.
How does this pertain? The sub $15K market is big, but it sucks, you
can't make money there. Its like the sub $600 computer market. You
can't make money there. Dell makes their money on volume sales and up
sells. Dell is also the only PC maker other than Apple that has
consistently made money over the past five years. So lets consider this
strategy. Apple could drop its smaller markets [education, consumer,
bio-tech] and go after the largest computing cash cow there is
[corporate sales]. Would this be good for Apple? No ... why? Because it
doesn't work for anyone, you need your products selling in a range of
markets in order to do well. But to a developer does Apple make sense?
Depends on your markets and what you develop. QuickTime has more
exclusive content releases than any other media player ... Lord of the
Rings used a ton of Apple and Adobe software products on Mac OS X
systems. Pixar just ported RenderMan. Apple reports that their
developer base is increasing in size as major UNIX developers are
moving to it.
Let me clear up one more fallacy: Apple's installed base can rise as
their market share decreases. WHAT? HOW? Well, its not as strange as it
sounds. Market share is based on a percentage of the whole. If Apple's
market share each quarter does not grow at a rate faster than the total
market grows their market share will decrease, however, this continues
to add new customers to their installed base ...
Is this a good scenario? Not really, but its not a deal knell either.
Couple other notes:
Apple is 100% debt free. Are any other major hardware or software
developers?
Apple has $4.6 billion dollars, not the same as MS having $45 billion
but its still a huge amount of money. Apple is still a very big
company.
David
-- This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries Official Newton FAQ: http://www.chuma.org/newton/faq/ WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/
This archive was generated by hypermail 2.1.5 : Mon Apr 19 2004 - 08:00:02 PDT