WHERE AND WHEN TO USE THE WATERFALL MODEL?
Waterfall methodology is ideal for projects with clearly understood, fixed, and documented requirements, well-defined technical tools, architectures and infrastructures, and a short life cycle.
Use Cases Where Waterfall Works Well
In general, traditional waterfall project management is a good fit for projects that:
▪️ Can be planned from beginning to end before they start;
▪️ Don't require work on multiple phases at the same time;
▪️ Have a clearly defined product and process.
The Healthcare industry is a good example where waterfall methodology is a nice choice. Scientific research is naturally an orderly practice, and the end product is clearly defined. To develop a new drug, scientists form hypotheses and proceed through a rigorous set of steps. Each time they fail, they start over, forming adjusted hypotheses to explore.
Here's a brief overview of how a waterfall model might go in a pharma context:
1.Planning. Scientists research the disease they're trying to cure, including studying it in a lab and interviewing patients affected by it. They form a hypothesis on a potential cure.
2.Designing. The scientists develop a waterfall project plan as to how they will explore the hypothesis and what resources they need.
3.Implementation. The scientists follow their plan and develop a drug that potentially cures the disease.
4.Testing. The scientists perform relevant testing to verify the drug efficiency. If it doesn't work, they start over.
5.Maintenance. The scientists reflect on the process, identify lessons learned, make changes to their hypotheses and development process, and document all these aspects to optimize their next drug development project.
Use Cases Where Waterfall Doesn't Work Well
In general, traditional waterfall project management would not be a good fit for projects that:
▪️ Necessitate different phases or tasks be worked on simultaneously;
▪️ Require feedback at multiple points throughout the project;
▪️A working prototype is more important than quality (eg. you first need to test if there's a market demand);
▪️ Don't have a clear picture of how the final product should look like;