Agile won’t work; waterfall is too slow; are we wagile? These are the comments you hear today as new transformational projects are undertaken in financial services—my colleague Ken Barnet even wrote a great blog on the topic last year. While these discussions have merit, my initial thought on the subject is, are we over-emphasizing methodology? As a project manager for over 20 years, the methodologies utilized have been secondary to the core objective of delivering a high-quality project on time and on budget.
As long as we aren’t losing the forest for the trees, I acquiesce that organizations do require a defined methodology to ensure success on large transformational projects. As food for thought, I’ve included a few items to consider when defining a methodology for your project:
In the Project Management Institute whitepaper, Blending agile and waterfall: The keys to a successful implementation, Alex Rodov and Jordi Teixido introduce the concept of the “Golden Mean” to large scale implementations: “…[A]s the Greek philosopher Aristotle indicated long ago, the middle ground between extremes, or the Golden Mean, is probably the right approach to successfully implement the challenging projects ahead.” Reflecting on my career, it’s evident the model I have generally leveraged throughout is an informal hybrid model. In my projects, I have combined aspects of agile (end user engagement, iterative releases, quick wins, etc.) with a waterfall foundation (defined plan, measurable milestones, etc.).
As agile becomes more and more prevalent in the financial services industry, I think it is imperative that agile and waterfall project managers adapt to a hybrid model to be successful.
Data integration work, business process tools, and internal reporting are examples of areas that can, and should, take advantage of collaborative aspects of agile ensuring work is delivered quickly and correctly. To avoid significant rework, data integration builds should not start until downstream data delivery requirements have been captured as much as possible given the projects timeline. On the flip side, client reporting should leverage a more traditional waterfall approach given the complexity and rigidity of most client reporting requirements. Other aspects of a transformational project, core vendor configuration, model offices, conversion tools, etc. should be reviewed to determine the best methodology to leverage for those components.
To close, I return to my original comment; the project methodology leveraged is secondary to ensuring you use all available tools to deliver a project on time and on budget. With that said, and based on my experience, the best model for a large transformational project is a combination of agile and waterfall concepts integrated to create your project’s Golden Mean.