I've been thinking about business ideas lately and this topic comes to mind.  Many of the large and well publicized acquisitions by Yahoo and others are for general purpose large user base applications.  These include picture sharing and social networking in one form or another. 

Today in a meeting one of my colleagues made the comment that "custom is always better" in reference to COTS versus home grown software.  This started a thought process on the value of software and a continuum for measuring the value. 

On the low end you have a product like Writely or Basecamp.  These applications are simple and general purpose.  On the other side you have software for a custom application and in the extreme case it is for one customer, one site or perhaps one special purpose like running one of a kind machinery.  Now I believe that we can place value of the software along this spectrum.

A piece of blogging software has almost zero value because it is on the commodity end of the spectrum.  Anyone can write blogging software and everyone apparently has.  There are so many open source blogging engines that it's almost comical.  Let’s say that the developer/publisher makes one cent off each user so volume is important.

On the other hand specialized software is expensive.  Software for complex processes like specialized work flow, monitoring and industrial processes is very expensive.  COTS packages for this type of application are not cheap.  In this monetary range many organizations hire programmers either full time or as consultants to build custom applications.  These often are the most expensive of the software options, yet like a tailored suit often provide the best fit.

The problem I’ve seen is that companies that have a full time development staff often opt for COTS packages.  While many times this is a great fit for example buying the MS Office suite is a good idea, trying to wedge a work flow system into an organization is not such a good idea.  The more custom your application, meaning the more a COTS solution needs tailoring for the organization, the more a custom developed application makes sense. 

These types of application can pay very well but you need a lot of pay from these jobs because you very likely can only sell one of them.  You can’t mass produce these applications on a CD and put them on the shelf in Wal-mart. 

What does this mean?

I think that the companies purchased by Yahoo, Google et al. are lottery winners.  The development effort was equivalent to buying the $1 Lotto ticket at the local gas station.  When you win you win big but most the time you lose but it’s not a big deal because you haven’t invested much.

The other side is more like a Texas Hold ‘Em game.  You’ve got fewer competitors but you know who you’re playing against.  This is where the big boys play; the COTS players with the deep pockets, if you will. 

I think my spot is somewhere in the middle.  I don’t want to write yet another blog engine but I don’t want to make shrink-wrapped software.