Some projects have considerable uncertainty around project requirements and how to fulfill those requirements using current knowledge and technology. These uncertainties can contribute to high rates of change and project complexity, which can lead to an increase in time and cost. PMI-Agile (2017) suggests using a Uncertainty and Complexity model based on the Stacey Complexity Model, to determine if a project is better suited to a Traditional (i.e. Linear, Waterfall, PMBOK) approach, of whether it needs a more adaptive approach (Agile).
As PMI-Agile (2017) suggest, teams can plan and manage projects with clear and stable requirements and clear technical challenges with little difficulty. However, as the uncertainty in the project increases, the likelihood of changes, wasted work, and rework also increases, which is costly and time consuming. Therefore, projects where there may be a high degree of change and uncertainty, incremental and agile approaches work well for the projects that
involve new or novel tools, techniques, materials, or application domains.
They also work well for projects that may have the following
Require research and development
Have high rates of change;
Have unclear or unknown requirements, uncertainty, or risk; or
Have a final goal that is hard to describe.
The traditional Waterfall approach, therefore, works best where is a clear path and direction with little unknowns. If there a high degree of change or uncertainty, then an Agile approach is more suited.
Figure 15 shows the relationship between time, change, and the cost of change on a traditional project. As time and information about the product increases, the window to initiate change decreases and costs relating to these changes increase.
If we look at a comparison of using the Agile Scrum framework in comparison to a Traditional Waterfall approach as used by PMBOK, we see that any changes made using the PMBOK methodology more than likely will be more complicated, and more costly,
As delivery and testing is on a more frequent basis with regular Sprints, this makes it both easier and less costly to instigate and implement any changes if required. Additionally, as collaboration with the customer is more frequent, these requirements and changes can be relayed more regularly.
Other areas where cost management can be more effective with Agile, according to Layton and Ostermiller (2017), is the ability for the product to generate revenue before the project is actually complete, leading to a possible self funding project. Additional cost-efficient aspects to Agile over Traditional according to Layton and Ostermiller can be seen below.
Despite these differences, there are some commonalities, especially when looking at the tools and techniques used when estimating and determining costs. Both can incorporate techniques such as expert judgement and analogous estimating. The main difference is in how those techniques are used with the bottom up approach on projects managed with the traditional style compared to the top down approach used in Agile. Estimating costs in an agile environment looks more at the project or phase scope as a whole rather then breaking it down into tasks, as per the traditional project management approach.
The agile approach to determining costs also tends to use more group approaches when estimating and allowing the project team to contribute. Techniques such as planning poker and affinity grouping show the agile principle of favouring people and interactions over tools and processes. Traditional approaches puts more of an onus on the project manager to obtain the estimation through calculation or from information on previous projects.