There are 32 criteria of a DCMA (Defense Contract Management Agency) certified Earned Value Management System (EVMS). Such a system is mandated for federal contracts greater than $50M and must be considered for contracts greater that $20M where EVM is mandated.
For the non-DoD, NASA, or DOE program there are easier approach to deploying EV. The first place is to start with what is the minimal "good practice". Below are the 32 Criteria from EIA/ANSI-748B
The consensus - lead by Ray Stratton is that they are: 1, 2, 3, 6, 7, 8, 16, 22, 27, 28. There are other opinions, but that is a moot point. These are the starting points for deploying an EVMS for commercial or other domains, where the DCMA is not visiting on an annual basis.
So What are These items
1. Define the WBS
If you don't know what the components of the deliverables are, you can't be expected to know how much it costs and what work needs to be done. The WBS is a Product Breakdown Structure (PBS) and the services needed to produce these products. NOT the functional organizations that deliver these services.
2. Identify the Organization
Who's doing the work? With the WBS and the Organizational Breakdown Structure (OBS) you can build the Responsibility Assignment Matrix (RAM). In the defense business the intersection of the OBS and WBS is the Control Accounts and the Control Account Managers (CAM). This is probably too formal for many domains. But the RAM is always needed. If you don't have a RAM there is no way to hold anyone accountable for the deliverables, the budget and schedule for those deliverables. The notion of a "self managed" team is possible in the presence of a RAM. Without a RAM a self managed degrades into chaos.
3. Integrated Subsystems
Every non-trivial projects has subsystems. Allocating budget and schedule to each subsystem is critical to segregating cost and risk. Decision making around trade off is driven by this type of information.
6. Schedule the Work
When are you going to do what the WBS says is needed to complete the project? What is the order of this work? What is the critical path of the activities for this work? What are the dependencies between the work elements?
7. Identify Products and Milestones
How will you know you are making progress. The presence of a tangible physical outcome from the work effort is the best way to do this. Passing a milestone is not my favorite. The "products" can be intermediate versions of the final product. Prototypes, partially functioning products. The favorite approach for the Agile. In the end only physical evidence should be used to measure progress to plan. And - most importantly - the define of what "done" looks like must be in place BEFORE the works starts, so when the Milestone (which is just a rock on the side of the road) is passed, you can look at the result to see if it was what was planned to be done.
8. Set a Time Based Budget
Money is converted into products. What is the profile of the spend rate for this money? How much money do we need? How much money do we have? How are we planning on spending the money? This is called budget management.
16. Record Direct Costs
How much did we spend? This is the direct cost of doing the work we planned to do.
22. Periodically Collect Control Account Sums
Once a month, collect all these numbers in one place and look at them. Ask hard questions. Did we do what we said we were going to do? Did we spend what we said we were going to spend?
27. Revise Estimate at Complete
If the answer to the questions in #22 is unfavorable, then change the plan.
28. Incorporate Changes
With the changes in hand, make a new plan and execute it for the next period of performance.
That's it. That's all you need (for now) to do earned value. You've got the planned cost (BCWS), the actual costs (ACWP) and the earned Value (BCWP). The only formula you need is
BCWP = BCWS x "Physical Percent Complete"
