To Agile, Or Not to Agile?
Agile is the solution, or is it not? As with many frameworks and tools, it always depends on the application. Just think about the "real world": would you use a screwdriver to drive a nail into a wall?
Agile and more specifically SCRUM have taken the IT world by storm - and for good reasons. Running your IT projects using an agile framework has many advantages and actually feels much more "natural" than using a waterfall approach. However, for the same reasons why agile is so great, e.g., for software development projects, it is a nightmare for certain other kinds of projects.
The major reason for this is simply that projects where agile is the perfect fit share certain criteria, such as: there is no complete blueprint available (realistically) at the start of the project, many details are uncertain, requirements can change as the demands are sometimes highly volatile, the technological risks are big, since many problems can only be really identified when you are already halfway through the project, and last but not least software projects are very hard to estimate correctly.
Because of these criteria, I still see a lot of growth for agile: just think of the early prototyping and design projects in the automobile and aerospace industries, the early stages of drug development projects in the pharmaceutical industry, or any other early-stage high-tech project. BUT what if the project criteria are totally different?
What if you need to plan for explicit deadlines based on concrete milestones? What if dependencies between tasks are mission critical? What if you need to do explicit resource planning, because you are scheduling 99% key personnel that is not interchangeable? What if you need an upfront cost estimate for a fixed-price proposal?
In this case, the answer is really simple: then agile is not for you. Do not try to adopt agile just for the reason of running with the pack. Agile is great and we use it at Onepoint since the beginning - for our design and software development projects. But I would never use agile for a fixed-price customer project, or for my marketing plan - it simply does not make any sense. Don't try to use a screwdriver to drive that nail into your wall, simply use a hammer :-).