Why isn't there an established methodology for APM?
In my previous post I was remarking on application portfolio management projects that I’ve worked on, now and 7 years earlier, and I noted one interesting fact in particular. This being that, despite the fact that application portfolio management has been around since I was in school (to put that in perspective, Margaret Thatcher was Prime Minister back then), and that the idea of an application portfolio dates back to the early seventies, we still see formal application portfolio initiatives. And yet, I’ve yet to see any formal industry standard methodology for APM.
Yet there are other areas of IT management, such as IT service management or enterprise architecture that date from the same era, have made significant strides in the standards and best practices that exist around them. ITIL is now at version 3; meanwhile, TOGAF 9.1 has almost 50,000 people with the level 1 certification. And in IT governance we have COBIT 5, while the Archimate modeling language is at version 3 now.
So it seems a little strange that an activity as common as application portfolio management, and one that can offer traceable cost savings, hasn’t evolved an established body of best practices and methodologies. It’s instructive to consider why this might be the case.
A while ago on this blog Shane Mortensen called me out for getting lazy and using the terms software and application interchangeably in a post about aliases. While this laziness of mine didn’t affect my argument in that post, he was still right to do so – terminology is important for clarity, and in particular the distinction between application and software lies at the heart of how application portfolio management keeps changing.
What is an application? Is an online service like Google Maps an application? What about a government’s employment authorization verification service? How about a mashup application that pulls information from 5 different sources? Or are they applications in their own right? It sometimes seems like identifying applications is more art than science – to borrow a phrase, “I can’t define what an application is, but I know one when I see it.”
The clue is in the name – an application is when we apply technology to solve business problems. In other words, there’s a component of innovation in an application, in how the technology is applied. This shouldn’t be a particularly controversial assertion – product and service innovation is a recognized discipline in organizations from Silicon Valley to national government departments.
So applications are based on innovation in how technology is applied. But this means that applications will always be an ever-changing target. New technologies emerge on a regular basis; people are now speculating on how areas such as artificial intelligence and augmented reality will change enterprise IT. Unless we think that we’ll reach a point where technological progress ceases, this means that new applications and new types of applications will keep on emerging. Likewise, people will keep finding ways to apply existing technology (or combinations of technologies) in new, innovative ways. That’s human nature.
So innovation is at the heart of applications, and this means that the landscape of applications, even the very nature of applications, will keep evolving and changing shape. Which is why application portfolio management will never be commoditized.