Cop out: failing to perform app portfolio management

I pointed out in my prior post that there is some debate in the IT community about whether the functions of IT really shouldn't be performed 'in house' but should be outsourced.  I think that the answer depends on the questioner.  There are cases where it clearly does no good for the functions to be insourced.  There are other cases where insourced app development, maintenance, and support operations provide a clear competitive advantage.

Regardless, the reason for outsourcing should not be laziness. 

If a company finds it 'too hard' to figure out if the value of IT exceeds the cost of IT, they need help, expertise, and guidance.  In some cases, it is tough to ask for help.  But just because you may have a hard time reading a map... should you sell your car and take the bus?  Taking the bus is a good idea, but not out of fear or laziness.  It is a good idea for other reasons.  Same thing with outsourcing IT.

Basically, every business that spends over 20% of their total IT budget on application maintenance must take the time and energy to create an application portfolio management.  Otherwise, that cost is unaccounted for.  Project management tells you what things cost but it gives you no way to dashboard that cost against value or measure amortization over time. 

I had a PM recently tell me that the company should spend the money to 'get it right' on his application because 'the business will benefit for ten years.'  It fell to me to tell him that, strategically, the application was providing redundant functionality and that the corporation was making plans to phase it out in three years, so his project had to yeild value in one year, or be cancelled.

That kind of information changes the nature of a project.  What was originally thought of as a 'long term solution' became 'quick fix engineering.'   The amount of refactoring dropped substantially.  The replatforming of components disappeared.  The team focused on producing better reporting and cleaning up an integration issue that improved productivity, rather than making the under-workings tidy.

That kind of information is not available without portfolio management.  Sure, you can ad-hoc it.  But if your company spends money on IT, and it is managing this money with an ad-hoc approach, your investors should be banging on the door, demanding the head of the CIO.

More importantly, if you decide to outsource, the information on your portfolio costs will be immediately valuable both for deciding if there is value in outsourcing and for measuring the performance of your outsource partner.  Otherwise, they can make short-term cost cuts that you would never have approved, report a nice bottom line, and decieve you into believing that they have saved you money. 

The mechanism for measuring the portfolio costs and value should be kept in-house.  (Otherwise the fox guards the henhouse).  If your outsource partner currently provides this service, bring the service in, even if you leave your IT outsourced.  Then, and only then, can you independently measure the performance of your choices.  That goes for all companies, whether they outsource or not.

So go ahead and outsource, but go with your eyes open.