As an Agile practitioner, I wanted to capture some of the great learnings my team had from a recent project in a series of blogs that look at how the mechanics of a large Lean/Agile project evolved. My role on the project could best be described as the Agile Manager. I want to write from this perspective in the hope I cover at least some new ground in the techniques of adapting project approaches from traditional waterfall to Agile.
The point is not to prove what we did or how we did it is the right way, but more of what worked and didn’t work and under what conditions. The blogs will be a combination of concrete field reports with the occasional digression into ideas that help define a challenge or seed a change. I want it to be apparent that it was the various changing contexts of the project that drove the choice of practices rather than the ebb and flow in popularity of the various flavors of Agile. I’ll outline those changes in the next blog and refer back to them often.
About the Project
Size is relative but our project had an eight figure budget and between the client and the ICF-Ironworks numbered 90 personnel in, around and including the delivery teams. It was challenging in that the requirements and underlying goals changed over time, resulting in several distinct configurations of the team and techniques that are worthy of reflection. It is possible to look at the drivers for change without having to know too much about the application or its technology. However it would help to know something of the project only to understand the challenges and opportunities I’ll talk about later.
The main goal of the project was to migrate the main marketing site of a large financial institution from a legacy PHP based solution to one based on our Oracle’s WebLogic Portal (WLP). The Portal application was to be built on the target WLP architecture responsible for delivering the security and scalability needed by the enterprise’s Portal and portal like applications. The application itself was expected to leverage shared frameworks to enable faster time to market for later capabilities.
How the Project Began
The ICF-Ironworks Portal team was involved in the initial discussions and asked to create a strategy that would help guide future planning of the architecture, platform governance and application development.
Analysis revealed some 3000 content heavy pages that would need to be migrated in some form to the target architecture. The pages were very dynamic, enabling the organization to market to different market segments with the application being expected to determine a customer’s likely segment from geography and third party targeting services. The look and feel was standardized which ended up being a good thing as the design of the application made use of this consistency to develop tools that optimized time to market of new pages and content while minimizing the specialized skills and size of the team needed to maintain the site.
ICF-Ironworks was then asked to develop the critical content management system and interface to the Portal application which was being developed by one of the organization’s existing Portal application partners.
Rebuilding for Business Value – Metrics Matter
Some time after the initial rollout of the lowest risk portions of the site, the organization reached back out to ICF-Ironworks requesting a review of the high value portions of the site, now in the immediate pipeline for conversion to the new platform.
The application was showing signs of not delivering on the SLAs for changes to the site, in some cases underperforming the throughput for updates on the legacy platform. A study conducted by the ICF-Ironworks team determined that the current application architecture would never be able to meet the expectations on time to market and consistent collection of analytics.
The decision was made to rebuild the application and supporting tools with renewed focus on the operational metrics in addition to the typical application performance expectations for a high traffic site in the top 100 busiest in the US and top 500 in the world.
It is from this small harbor of understanding; I set sail for a series of discussions aimed at showing how the team and its practices evolved over time and in response to shifting tides of change.