A Call for Application Portfolio Management

So I have neglected my blog as project life has taken over, and all I have been doing as of late is living and breathing CPIC, PMO, and on odd days, legacy systems and a COBOL mainframe evaluation. Time for a vacation really!

But all of the stuff is interesting, and certainly going on because of people doing good things (BTW, the equation of good in my book goes something like this:  right thing + smart thing = good thing). And while that does not surprise me, it does make me happy and excited to be working on these projects. The mainframe one is unique, mostly because my web ways (full of web governance, web 2.0, social media, and open source) don’t compare to thing like COBOL, a 50 year old language that still accounts for 80% of the currently active computer code (i.e. 250 billion lines of it) worldwide. I honestly feel like I have entered a parallel universe. In its own right, COBOL certainly seems to be.

I get that COBOL was revolutionary, easy to use and allowed the business process to be the leading act. I don’t get why it has been allowed to become its own ecosystem where all was well until an imbalance occurs. And now with a huge portion of the COBOL workforce retiring (i.e. baby boomers) there is a sense of panic as the ecosystem is threatened by a looming shortage of staff and a scramble to snatch up those few green screen programmers remaining. So here are my questions: Did nobody see this coming? Was anybody paying attention? And what lessons will we learn from this that we can apply to our own little ecosystems of java, PHP, Ruby, and .NET?

For me this has been a case study supporting the argument for planning, analyzing, and most importantly managing our ecosystems. We should focus on project management for our project releases, upgrades and enhancements, but we must have a capital planning aspect to our investments, and we must apply a management technique (i.e. Application Portfolio Management, or APM) to understand whether we are in balance, or whether our applications will go the way of COBOL with that sudden threat to their existence and a need for immediate reaction.

So, here are some suggestions of what to consider around APM:

1. Build an inventory of all of your applications, including the platforms, requirements details, and priority and business value from your stakeholders’ perspective (for a detailed discussion on what to collect see the Application Portfolio Management site at http://www.applicationportfoliomanagement.com/).
2. Start collecting and understanding the costs associated with the inventory of applications you have built, ensuring you have a mechanism in place for monitoring the changes and understanding what increases or decreases the cost.
3. Build out a business process (light and with low overhead) to measure the business value from the applications you have in house, and tie them back to the triggers for increasing or decreasing the cost (as described in #2).
4. Benchmark your portfolio against industry standards, or at the very least, understand at what point your costs are going to start outweighing the benefits, and consider what your response plan will be to that scenario. In essence, you are building a risk pattern and preparing to mitigate for those risks.
5. Analyze your portfolio (every six months or year, or whatever frequency makes sense based on the changes of cost and value you are noticing) in order to start your application portfolio management journey.