Electric Currents

Pundit Gets It Almost Right!
JR posted in Windows on December 31st, 2005

In John Dvorak’s December 21 article, he examines Microsoft’s onrushing venture into software subscriptions as a revenue source.  What he is missing is the reason that Microsoft is doing so.

Dvorak points out that Microsoft built a large part of its fortune on selling software directly to consumers and that this flew directly in the face of the previous paradigm which was licensing usage on a “per seat” basis for a fixed period of time.  He notes that he doesn’t understand why they would now decide that the previous model was so good that Microsoft needs to get into that business.

Dvorak then goes on to say that Microsoft should instead invest effort into improving the Office suite and points to PowerPoint as needing particular improvement.

Dvorak is so close to answering his own question, but doesn’t see it staring him in the face.  The problem that Microsoft wants to fix with a return to a software subscription model is that they are having trouble improving the Office suite at all.  It is widely acknowledged that users don’t use even 10% of the capabilities built into Office (and, if my experience is any guide, that may be as little as 2% (still hitting two returns to space paragraphs?)).  Why try to improve the software if no one is using all that is already there?

In conjunction with this is the Microsoft monopoly problem.  They already so dominate the Office space that their real competition is from previous versions of their own suite.  Why upgrade from the nicely working version of Word 97 when you are already not making use of any of its outdated ‘advanced’ functions?  Microsoft has successfully created a near monopoly in the Office suite.  How can they make more sales when the existing stuff is still doing everything and more than the users need to do?

If you want to get a revenue stream for something that isn’t changing and isn’t likely to change in general, a subscription model is a great idea.  No problem with doing any more work on the software, just keep it working properly in the core functions.  Any bells and whistles added are “giving increased value” to subscribers and can be done without risking a revenue stream.

There is plenty of reason to test a subscription model for software.  But I have to agree with Dvorak, it will ultimately be a waste of money.