Denver airport study
Case studies at three large, software product companies show the benefits and pitfalls of integrating agile methods with stagegate management models. 0740-7459/05/$20.00 © 2005 IEEE
he project manager enters the room where the pilot agile development team sits. “We have to add a new feature to our next release,” he says. “Can you please put it on the list?”
“Yes, of course,” a team member replies. “Just let us know which feature you want us to postpone in its place.”
The manager gets confused and a little annoyed. “What’s happening? When we used our traditional development methodology, you could squeeze in
T
all the new features we needed. I feel as though
I’m no longer in control here.”
This fictitious dialogue captures the cultural changes and different management and engineering viewpoints we encountered in a study of three large software development organizations that started using agile methods.
We undertook the study in response to questions raised at an industrial seminar on
Extreme Programming (XP)1,2 and agile development methods held in 2003. The Software
Engineering and Research Group at Lund University sponsored the seminar, which included representatives from more than 20 different software companies. The representatives were extremely interested in agile development, but they weren’t sure whether or how it might coexist with their existing project management models. Their interest prompted us to study the fea-
sibility of applying agile methods in the context of large software development projects using stage-gate project management models
(see the related sidebar). The results showed not only the feasibility but also the benefits of this approach as well as issues that must be addressed to ensure its success.
Case studies’ scope
The software engineering community has generally viewed agile methods favorably. The focus on simplicity, programmers, and products empowers the technical staff and gives
he project manager enters the room where the pilot agile development team sits. “We have to add a new feature to our next release,” he says. “Can you please put it on the list?”
“Yes, of course,” a team member replies. “Just let us know which feature you want us to postpone in its place.”
The manager gets confused and a little annoyed. “What’s happening? When we used our traditional development methodology, you could squeeze in
T
all the new features we needed. I feel as though
I’m no longer in control here.”
This fictitious dialogue captures the cultural changes and different management and engineering viewpoints we encountered in a study of three large software development organizations that started using agile methods.
We undertook the study in response to questions raised at an industrial seminar on
Extreme Programming (XP)1,2 and agile development methods held in 2003. The Software
Engineering and Research Group at Lund University sponsored the seminar, which included representatives from more than 20 different software companies. The representatives were extremely interested in agile development, but they weren’t sure whether or how it might coexist with their existing project management models. Their interest prompted us to study the fea-
sibility of applying agile methods in the context of large software development projects using stage-gate project management models
(see the related sidebar). The results showed not only the feasibility but also the benefits of this approach as well as issues that must be addressed to ensure its success.
Case studies’ scope
The software engineering community has generally viewed agile methods favorably. The focus on simplicity, programmers, and products empowers the technical staff and gives
References: 1. K. Beck, “Embracing Change with Extreme Programming,” Computer, Oct. 1999, pp. 70–77. 2. K. Beck, Extreme Programming Explained: Embrace Change, Addison-Wesley, 1999. 3. C. Wallin, F. Ekdahl, and S. Larsson, “Integrating Business and Software Development Models,” IEEE Software, Nov./Dec. 2002, pp. 28–33.