Project management methodologies are very rarely one size fits all for every project. Program managers often must customize methodologies to suit their needs, occasionally even blending methodologies to satisfy all project and stakeholder requirements.
Agile EVM is one such blend. Project management professionals Bijan Nikravan and James B. Forman note in their PMI Global Congress paper that an agile EVM approach helps determine an agile project’s true status by detecting early signs of trouble and providing end-state forecasts based on current project data.
How did this blended methodology come about? Companies use agile methodologies to encourage innovation in projects and to adapt more seamlessly to changes by focusing on the performance of the team in meeting incremental progression goals. Agile, which we addressed in depth in our previous post on organizational agility, focuses on the process and deliverable benchmarks, which are all key data points on the progression of a project.
What’s missing, however, are the monetary and schedule benchmarks. Agile does not concern itself with budgetary performance, and it focuses on sustainable pacing rather than hard deadlines.
Company executives and customers, however, do concern themselves with those project benchmarks. These stakeholders expect to be able to track and predict cost and schedule performance throughout the lifetime of a project to determine whether deliverable goals are being met.
That is where earned value management comes in. EVM provides specific data points based on the value of work done and forecasts future progression based on past work. Visit our previous post for more details on the beneficial use of EVM in project management.
Though not congruous by definition, the fact that agile and EVM both help manage change in a project has led to the evolution and adoption of agile EVM.
How Earned Value Management And Agile Work Together
On the surface, earned value management and agile methodology do not appear to work together.
John Zyskowski, senior editor at Federal Computer Week, explains the difference between these two frameworks. With EVM, program managers create a project baseline at the beginning of projects that includes a detailed scope, a schedule and a budget. As projects progress, EVM compares the values of the work completed against the baseline to catch and correct deviations.
When using agile, on the other hand, program managers only have to have a high-level definition of project requirements to work from, as well as a schedule and a budget. The framework assumes there will be changes over time, so the requirements are refined along the way based on testing and feedback, not actual performance based on a project baseline.
So, how do these two methodologies converge? According to experienced scrum trainers Tamara Sulaiman Runyon and Hubert Smits, agile EVM uses the scrum framework artifacts as inputs with traditional EVM calculations, and the resulting output is expressed in traditional EVM metrics. This is the foundational application of agile EVM.
It is important to understand the data points of each discipline separately to understand how they work together.
Primary Scrum Data Points
Scrum is an agile practice that breaks projects up into sprints, or iterations, in which the team has a certain amount of time to complete the work for each sprint. Progress is regularly checked via these primary data points:
- Product backlog — a features list for project requirements
- Story points — units of measure for expressing the estimate of the effort required to deliver a product backlog item
- Velocity — a measure of the amount of work a team can tackle during a sprint
- Sprint backlog — a list of tasks to be completed during a sprint
- Release burndown charts — tracks the amount of work remaining at the start of each sprint
None of these data points manage or track project costs or performance and cannot provide metrics to support management decisions, such as changing requirements.
Primary Data Points in EVM
Earned value management metrics find variances in projects based on comparison of work performed and work planned. It is used in cost and schedule control as well as project forecasting, project management professional Umesh Dwivedi notes. The earned value management methodology has three basic metrics that serve as the baseline for the other calculations:
- Planned value — a time-phased budget baseline as an immediate result of the baseline schedule
- Actual cost — the cumulative actual amount spent at any given time
- Earned value — the amount budgeted for performing the work done by a given date
The ability to calculate these metrics lies in EVM’s rigid baseline requirements.
How They Merge To Create Agile EVM
Agile EVM is the result of combining those scrum and EVM data points to enhance project performance feedback. The combination of the two is stronger than their individual selves. Sreedhar Koganti, experienced software and system developer, notes that in agile EVM the metrics are used to measure the progress at the end of each sprint, when actual velocity and costs are known. This is in direct contrast to the predetermined intervals set in the traditional EVM baseline plan, Koganti says.
Agile EVM works by comparing the current release plan against actual work performed. In agile EVM, program managers need to know these project data points to establish a baseline from which to calculate performance metrics. Those points include:
- The number of sprints a project has
- The number of sprints that have occurred previously
- How much work is required to complete a project
- How much work has been done
- The initial estimated cost of a project
- The current actual cost of the project
Once this data is collected, program managers can then calculate the foundational agile EVM values of expected percent complete and actual percent complete. The important thing to keep in mind when calculating these percentages is that in agile EVM a story point is either done or not done. There is no percentage of completion for these values, Runyon and Smits note. These values are then used to calculate agile EVM Planned Value and Earned Value. From those metrics, PMs can calculate:
- the Schedule Performance Index (SPI), which shows whether a project is on schedule
- the Cost Performance Index (CPI), which highlights whether or not the project is on budget
- the Estimate At Complete (EAC), which is the forecasted cost of the project as the project progresses
In agile alone, these metrics are not available. The value of agile EVM is in the ability to use the shorter sprint cycles of agile in conjunction with the linear project performance metrics of EVM.
The Benefits Of Using Agile EVM
Though the adoption of agile EVM methods require a shift in thinking for both disciplines, and extra effort to properly implement, organizations that are able to do it realize significant benefits.
The Ability to Re-Baseline
The biggest benefit to blending the methodologies is the ability to re-baseline after every sprint in agile EVM.
EVM does not allow for changes to the baseline once it is set. Its metrics calculate performance based on that baseline. In agile, the baseline is loosely defined, and changes are usually encouraged and made at the end of each sprint.
Agile EVM allows for the merging of these two ideas. By re-computing the baseline after every iteration, project managers are validating the modified backlog against the release plan, a confirmation that the work planned for the release will be completed on time and on budget.
Make Better Business Decisions
The power of agile EVM lies in its ability to provide data on cost, performance and schedule, making it more informative than traditional agile burndown charts.
Agile EVM metrics can compare monetary expressions of value planned, earned and consumed at any given time; calculate cost and schedule efficiency; and make end-of-project forecasts on time and cost.
PMP and scrum coach Tamara Sulaiman Runyon explains that agile EVM metrics do not replace agile metrics such as burndown charts. The two sets of data can be used in conjunction to provide stronger data on performance and business value, which helps management make better business decisions.
As Nikravan and Forman explain, agile sprint cycles help reduce risk and uncertainty through quick delivery, keeping product backlog data dynamic and the data relevant. Nevertheless, EVM is still required to communicate bigger-picture data to non-agile stakeholders for project control and executive decision making.
Technology is Paving the Way for the Merger
Technology is evolving with the idea of blending agile and EVM. Managers operating at project and programs levels are overcoming traditionally inflexible processes and adapting to the need for continuous change and improvement. As such, innovative software solutions are being developed to aid program managers and industry professionals in this transition, as we presented in our previous post on the role of technology in continuous change management.
Joe Kerins, chief defense industry strategist here at Artemis, notes that organizations are adopting these new technology solutions and integrating them in a way that maintains that structure and processes of an organization but is flexible enough to allow for change. The inherent flexibility of the software allows for flexibility in the processes and operations of an organization.
Agile EVM Helps Program Managers Meet Stakeholder Demands
Clients care about the end product being on time and on budget. Executives and project teams are concerned with milestones along the way that signal whether their internal processes are allowing them to meet project cost and schedule requirements.
By using agile EVM, program managers can deliver the data each stakeholder group is looking for while better managing projects based on performance milestones in sprints and overall baseline performance.