The sub-pages herein are some provide a flavor of the kinds
process considerations that have arose in work done for our clients.
Why Agile Projects Fail - Within an agile environment, certain subtle disfunctional system dynamic can arise that can derail the project. This page describes one such situation bases on a real-world case study.
Mitigating Risk from Scope Uncertainty -
This is a write-up on scope management within Agile project. It
emphasizes getting an early handle on unknown scope that would
otherwise rear its ugly head and cause a crisis later on in the
project. It also describes how to incorporate probability into team
LOE estimates to make them more realistic.
Product Backlog and the Cone of Uncertainty -
speaks more directly to why many Agile project fail. The root cause is
the crisis of unknown scope coming to light after a few iterations.
Typically this scope goes unmeasured and the other parts of the
triple-constraint are not updated to compensate. The result is
generally a crisis near the end of the project where everyone realizes
that the project will not be done on time and time is running out. By
attending to this early, projects can compensate to succeed.
Right-Sizing Iterations discusses
how determine the best length of an Agile development iteration (Scrum
in this example). It also introduces a new metric, the "In Play
Profile" that can be used as an early-warning indicator for when an
iteration is in trouble. I have found this useful in cases where the
development staff is insular and pushes back against allowing anyone to
have direct visibility into their activities.
Sample Metrics Analysis demonstrates
the kind of data that can be pulled together and normalized to compare
the state of one release to previous ones to get a sense of risk to
meeting a release date with a low level of defects. In this example,
the metrics are being captured during a RUP Transition Phase.The
metrics are generally composite and focused on productivity.