I was working with a client earlier in the week on their IT process model. In common with many IT shops, their processes were heavy on finding, starting, and managing new stuff. However, they were woefully light on stopping, killing or retiring old stuff!
Many (many!) years ago, I shared an office with a guy who was very structured, probably even “anal retentive” as they say. Anyway, he had a ritual in that if he needed to create a new file (I’m talking filing cabinets, manila folders and hanging green Pendaflex files!) he would always go through the file draw to see what old file(s) could be thrown out. One of our only points of frequent tension was that he could not get me to do likewise (I hate throwing stuff away). However, I did see the merit of his obsession, and I see one form of it today as mandatory IT practice – if you are creating something new, you better figure out how you are going to retire stuff that will (or could) become redundant.
This really is a function of IT Process Management, IT Life-cycle Management and IT Portfolio Management.
IT processes need explicit steps that address the “what can be retired, how can it be retired, how will it be retired” issues. As a leading practice example, I’ve seen clients where the business case for a new IT investment is not considered complete until the costs and savings of retirement opportunities associated with the new investment have been considered, and there is another business case triggered for the retirement project (or program).
All IT investments need to be considered in terms of a defined life cycle – with life cycle costs and benefits estimated, tracked and managed from start to finish. This must become institutionalized behavior – for projects, assets, and services. It really is shameful that IT shops have got themselves into a situation where almost all IT resources are consumed maintaining the stuff that’s already been built – and that this maintenance trap has come as a huge surprise to business and IT leadership – just kind of snuck up on us while we were busy creating!
Finally, the proportion of total IT spend consumed by “keeping the lights on” must be aggressively managed – that means knowing what this proportion is today, defining what it should be in 1 year, 2 years, 3 years, and so on, creating/executing programs and projects to achieve the target portfolio mix, and tracking performance against reaching that target mix.
Professor Peter Weill at the Center for Information Systems Research refers to this as “weed pulling.” It’s a great metaphor – and a critical IT discipline.
Filed under: IT Management, IT Maturity, Portfolio Management
