Performance
Measured against baseline version 1.
The end of a reporting period, measured through that close. Earned value is recomputed through that close, against the baseline that was in force at the time, and actual cost is never restated. The plan, the register and the people show today, because Capo keeps no history of those yet and will not pretend otherwise.
- Ensemble Index
- One score from 0 to 100 for how much this plan deserves to be believed, combining delivery performance, plan integrity, stakeholder alignment and team development. 85 and above is holding, 70 to 84 is holding with something to watch, 50 to 69 is under strain, below 50 is not holding. Arithmetic, never written by a model.
- Tuning
- How much the Ensemble Index itself deserves to be believed, from 0 to 100, based on how objective, how fresh and how confirmed the underlying measurements are. A low Tuning never quietly discounts the Index. Both are shown so a confident number resting on thin evidence is visible as exactly that.
- Cost performanceCPI
- The value of work finished divided by what was spent to finish it. 1.00 is a dollar of work for every dollar spent. Below 1.00 the work is costing more than it was budgeted to cost, above 1.00 it is costing less.
- Schedule over timeSPI(t)
- How much calendar the project has earned against the calendar it has used. 1.00 is on the baseline. 0.90 is roughly one working day lost in every ten. The reading underneath converts it into working days ahead or behind.
- Good, watch, off
- Each headline figure carries a state, drawn as a shape with its word so it reads without colour. Good, a circle with a check, is holding: a score of 70 and up, or a performance index of 0.94 and up. Watch, a diamond with a dash, is under strain: 50 to 69, or 0.81 to 0.93. Off, a square with a cross, is not holding: below 50, or below 0.81. A blank figure never takes a state.
- A blank where a number should be
- A figure that cannot honestly be computed is left blank with the reason printed beside it. It is never shown as zero and never filled in with a guess.
- PlannedPV
- What the baseline said would be finished by this date, in money. Drawn as a dashed line.
- EarnedEV
- The budgeted value of what is actually finished. Earned below planned is work running late, earned below actual is work running expensive.
- Actual costAC
- What has really been spent to date. It is never restated after a period closes.
Against a budget of 4.09m.
- If this rate holds
- 5.46m
- If it was a one off
- 4.15m
- If both persist
- 7.64m
- Bottom up
- Not available
Variance at completion is -1.36m. To finish on budget, the remaining work would have to run at a cost performance of 1.01.
- Forecast at completionEAC
- What the whole project is now expected to cost. Four scenarios are listed rather than one number, because the honest answer is a range: if the current rate holds, if the overrun was a one off, if cost and schedule problems both persist, and what the team estimates from the bottom up.
- Budget at completionBAC
- The approved budget for all of the work in the current baseline.
- Variance at completionVAC
- Budget minus forecast. A positive figure is money expected to be left over, a negative figure is the expected overrun.
- To complete performanceTCPI
- The cost performance the remaining work would have to run at to still finish on budget. Well above 1.00 means the plan is asking for a rate this project has not yet achieved.
- Objective measurement69
31 percent of the budget is measured by somebody's judgment of percent complete rather than by a countable fact.
- Schedule signal84
12 percent of the budget is level of effort, which earns planned value by definition and therefore cannot show a schedule variance.
- Freshness0
The newest progress measurement is 64 working days old.
- Measurement coverage16
84 percent of the budget sits in work packages nobody measured this period.
- Cost agreement100
The actual-cost sources agree, so no policy tie-break was needed.
- Human confirmation74
7 of 32 items were filled in by the planning engine and never confirmed by a person.
- Index coverage100
Every dimension of the Ensemble Index could be computed.
- Each factor scores 0 to 100
- Higher means the measurement underneath is more trustworthy. A factor below 60 is printed in rust, and the sentence beside it says what would raise it.
- Tuning
- How much the Ensemble Index itself deserves to be believed, from 0 to 100, based on how objective, how fresh and how confirmed the underlying measurements are. A low Tuning never quietly discounts the Index. Both are shown so a confident number resting on thin evidence is visible as exactly that.