BAC
$12.0M
Earned value has a reputation problem in construction: it sounds like something that needs a PMO, a consultant and a six-week setup. It doesn't. If you have a cost budget and a schedule with progress on it, you already have everything EVM needs — and three numbers out the other end will tell you more about your project's true position than the entire monthly pack.
Earned value is the budgeted cost of the work you have actually completed — not what you have spent, and not what you planned to spend by now, but the value of what is physically done, measured on the same scale as your budget. Once you have that number, three ratios fall out of it, and between them they carry most of what a director needs to know.
Research on large project portfolios has found that a project's cumulative cost performance index tends to stabilise early — often cited as by around 20% complete — and rarely improves after that. Whatever CPI you have established in the first fifth of the job is, statistically, close to the CPI you will finish on.
The management implication is the uncomfortable one: EVM's value is front-loaded. The window where measuring can still change the outcome is the very start of the project — which is exactly when most contractors haven't set reporting up yet. Set it up at project start, not when things feel wrong.
Take a $12M civil package, four months into a job that has run for a while. The planned value of the work that should be done by now is $3.6M. The earned value — what is actually complete, in budget terms — is $3.1M. And the actual cost booked against it is $3.4M. Three numbers off the monthly cost report and the schedule; nothing exotic.
Run the ratios and the picture is blunt. CPI of 0.91 means the job is earning 91 cents of work for every dollar spent; carried across the full budget, that is an estimate at completion of about $13.2M against a $12M budget — trending $1.2M over. SPI of 0.86 says the work in place is running about a period behind where the plan wanted it. And this is month four, on a job the site team would almost certainly describe as "feeling fine".
That is the whole case for earned value in one screen. Nothing on the site has gone visibly wrong, no milestone has been formally missed, and yet the numbers are already saying the project is a period late and heading a million dollars over — while there is still two-thirds of the job left to do something about it. A director who sees this at month four has options. The same director who sees it at month ten has an explanation to write.
Drop in your monthly PV, EV and AC and it computes CV, SV, CPI, SPI, EAC three ways, VAC and TCPI, with traffic-light thresholds on each. No signup, no email — it's just a spreadsheet.
Earned value isn't hard maths. Every objection to it is really an objection about data plumbing or client demand — and each one has an answer.
The calculator above is the manual version, and it's genuinely useful — a lot of good decisions have come out of three cells and an honest conversation. But it has the same weakness as every hand-built report: it is only as current as the last time someone had an afternoon to update it, and it stops the month that person goes on leave.
Aegis Command is the same calculation wired to the file you already maintain. It reads your schedule — P6, Microsoft Project or Excel — and generates the full earned value position (CPI, SPI, EAC, plus P50 and P80 completion forecasts from a Monte Carlo run over the network), critical path analysis, and a written monthly Director's Brief, directly from the export you produce anyway. No re-keying, no separate BI project, no per-seat maths: it's unlimited users on every plan, priced on the value under management rather than the headcount, with one project free forever at any size.
Earned value is the budgeted cost of the work actually completed to date — not what you've spent, and not what you planned to spend, but the value of what is physically done, measured in budget dollars. It's the one number that lets you put plan, progress and cost on the same scale.
CPI = Earned Value ÷ Actual Cost. SPI = Earned Value ÷ Planned Value. Both read 1.0 when you're exactly on plan; below 1.0 means you're behind on cost or on schedule respectively.
1.0 or above means every dollar is buying a full dollar of work. A CPI below about 0.95 early in a job is a structural warning rather than noise, because cumulative CPI tends to stabilise early and rarely recovers — so a weak number in the first quarter usually predicts the finish.
BAC ÷ CPI assumes current cost performance continues — the honest default. AC + (BAC − EV) assumes the overrun so far was a one-off and the remaining work runs to budget. AC + (BAC − EV) ÷ (CPI × SPI) assumes both cost and schedule pressure persist, and gives the most conservative figure when a job is both over and late.
In the United States, formal earned value management system (EVMS) requirements commonly apply to federal contracts above defined dollar thresholds. In the UK and Australia, public-sector clients increasingly request earned value reporting in the scope or works information even where no formal standard is mandated. If you're bidding public work, check the specific requirement in the tender documents rather than assuming.
NEC4 programme reporting — how the Accepted Programme becomes a monthly brief the board actually reads, with EVM on top.
P6 reporting — what happens between the .xer export and the report that gets read.
The Director's Brief template (.docx) — EVM position, critical path, risks and decisions-needed on a single page. Or read how it fits NEC4 reporting.
Pricing — banded on contract value, unlimited users, one project free forever.
Put one real job in and see the earned value position on your own project, generated from the schedule you already keep. Free for one active project, any size — the full product, no card.
Run one project free →