Tuesday, March 2, 2010

It's Time to end Waterfall Funding

I grew up in a family with five siblings and learned early on that each of us is bundled with gifts and challenges that cause wonderment, awe, and occasionally - humor for the rest of the population. I knew a guy once that could not tell if the car he was driving was headed up hill or down. Me, I'm great with recognizing voices, but faces - I'm hopeless.

One of my older sisters, being the only doctor in the family, has had to demonstrate her brain power to excel in her chosen field. She is very bright. But give her a map to read and she becomes a driving combination of Helen Keller and Dick Van Dyke. Simple things like, "Sis, start with placing the map in front of you with the north end of the map towards the top." "But, " she says, "what if I'm not going north." "If I turn the map as I'm driving, doesn't that send me in the wrong direction?"

Oh boy.

Reading a map is a lost art I suppose, but whether you're used to MapQuest, Google, Garmin, TomTom, or plain old Rand McNally, one thing is for certain. Most people consult their maps multiple times en-route to a new destination. When you are driving to an unfamiliar location, you plan out your journey, you get started, and several times along the way you go back and monitor your progress, you calculate your arrival time, and you make sure you're still on the right path.

If you've gone astray, you figure out how to get back on path. If you've strayed too far, you might even decide to scrap the trip, start over, or select a new destination. In the vernacular of Architecture, this is called stage gate risk mitigation. You create a plan (a map), you start, monitor your progress, and at certain stages along the way you reassess your progress, your proximity to plan, and possibly even re-evaluate your final goals, i.e. you establish gates to mitigate the risk of going off course.

Any organization that establishes itself as a leader in managing risk, should approach IT projects with a stage gate philosophy. Approvals to move forward should never be a binary manager-take-all go/no-go decision. Rather, approvals should be based upon current state, predictability of the next steps, and a continued re-examination of return on investment.

When a $100,000,000 loan is granted to a developer to construct an office building; only enough funds are dispersed to prep the land to dig a hole suitable for containing a foundation. In this way, the bank can mitigate against the risk of losing a great deal of money through either fraud or incompetence. As each phase of the building is constructed the bank can feel secure in dispersing only enough funding for the next round of work.

In the same way, IT projects can, and should, be financed. This enables the organization to mitigate the risks associated with IT development - a notoriously risky business. The 2009 version of the Standish Chaos report indicates that only 32% of all projects are delivered on time, on budget, with required features and functions, and 24% fail, i.e. are cancelled prior to completion and never used.

Additionally, this stage gate approach provides for a natural re-evaluation of the architecture of the system, something that rarely occurs with waterfall funding. If we rely on common sense, intuition, education, and best practices when taking a road trip, then maybe we should adopt as Standard Operating Practice the process of stage gating our IT projects.

No comments:

Post a Comment

Follow by Email