Earned value management in construction: a project manager's guide

Why the budget-versus-actual report is not enough

Most monthly cost reports put two columns side by side: the approved budget and the cost to date. On a $6.5M job, seeing $4.03M spent tells you almost nothing on its own. If the building is 70% done, you are doing well; if it is 48% done, you have a serious problem. The missing piece is the value of the work actually in place, and that is exactly what earned value adds.

Earned value is the method the PMBOK Guide and the PMP exam use for cost and schedule control. It is not reserved for megaprojects: on a mid-size building, all it needs is a budget broken down by work package, a schedule with a baseline, and an honest percentage complete for each package.

The three base values

Planned value (PV)

The budgeted cost of the work scheduled to be done by the status date, read from the baseline schedule.

Earned value (EV)

The budgeted cost of the work actually done: budget × percent complete, package by package.

Actual cost (AC)

What the work done has cost to date: invoices from subcontractors and suppliers, labour, equipment, accruals for work in place not yet billed.

Budget at completion (BAC)

The total approved budget, including approved change orders.

On a job site, the trap is AC. Actual cost must cover the same scope as EV on the same date. If the rebar is installed but the supplier's invoice arrives next month, accrue it; otherwise your CPI looks better than reality for four weeks, then drops all at once.

Variances and indices

Cost variance (CV)

EV − AC. Negative means the work in place cost more than budgeted.

Schedule variance (SV)

EV − PV. Negative means less work is done than planned, expressed in dollars.

Cost performance index (CPI)

EV ÷ AC. Below 1.00, each dollar spent produces less than a dollar of work.

Schedule performance index (SPI)

EV ÷ PV. Below 1.00, the job is producing less work than planned.

Estimate at completion (EAC)

BAC ÷ CPI, when the current cost performance is expected to continue.

Estimate to complete (ETC)

EAC − AC: what remains to be spent.

Variance at completion (VAC)

BAC − EAC. Negative means a forecast overrun.

Worked example: a $6.5M mid-rise building

A four-storey wood-frame residential building with a $6.5M construction budget. At the end of month 7, the baseline says 55% of the work should be done. Progress measured on site, package by package and weighted by budget, gives 48%. Cost to date, with accruals, is $4.03M.

Read it the way an owner's representative would: the job has spent $4.03M to produce $3.12M of work. At that rate, the building will cost about $8.4M, nearly $1.9M over budget, and it is also behind schedule. A budget-versus-actual report would have shown $2.47M “left to spend”, which is simply wrong.

Reading CPI and SPI together

• CPI below 1 and SPI below 1: over budget and behind. The most common pattern on a struggling job, often a productivity issue (crews waiting, rework, poor sequencing).
• CPI below 1 and SPI above 1: ahead but paying for it. Overtime or extra crews are buying schedule; check that it was a decision, not an accident.
• CPI above 1 and SPI below 1: under budget because work is not getting done. Usually a warning, not good news: the cost will come with the work.
• Both above 1: verify the percentages complete before celebrating. Optimistic progress is the most common source of a flattering CPI.

SPI has a known limit: as the job nears completion, EV catches up with PV and SPI drifts back towards 1.00 even on a late project. In the last third of a job, read the schedule from the critical path and the projected finish date, not from SPI alone.

Making it work on a real job site

1. Measure progress by work package, never by eyeballing the whole building. Use rules everyone understands: units installed, square feet of drywall hung, milestones within a package (for example 0% / 50% at rough-in / 100% at final).
2. Weight progress by budget. A package worth $900,000 at 40% moves the project far more than a $30,000 package at 100%.
3. Lock a baseline before you start measuring. Without a frozen reference, PV moves every time the schedule slips and SPI stays at 1.00 on a late job.
4. Add approved change orders to BAC and to the baseline at the same time, or the indices compare apples with oranges. See managing change orders.
5. Report the same day each month, with accruals, so the trend is meaningful. A single CPI matters less than three months of CPI moving in one direction.

Other ways to forecast the EAC

BAC ÷ CPI assumes the job keeps performing as it has. The PMBOK Guide lists other formulas: AC + (BAC − EV) when the overrun was a one-off and the rest will go as budgeted, or AC + (BAC − EV) ÷ (CPI × SPI) when a late schedule will also drive cost. On a building job, a bottom-up forecast by cost code, using committed subcontracts and quotes for what remains, is often more accurate than any formula. Use the formula as a check: if your bottom-up forecast is far better than BAC ÷ CPI, explain why.

Applying it in Teyvor

In Teyvor, each budget line is tied to a task of the schedule. Earned value is the line's budget times the task's progress, planned value is spread over working days between the task's baseline dates (its current dates until a baseline is locked), and the indices use the actual cost of the measured lines only, so an unmeasured line does not drag CPI down. The EAC applies CPI to the measured lines and keeps any unmeasured line at its budget. The cost control screen shows CPI, SPI, EAC and VAC next to the approved budget, committed and actual cost of each cost code, with the cash flow curve. The definitions are the ones described above; the arithmetic is done for you every time progress or a cost is entered.