The weighted milestone method is an earned value measurement technique where you assign a budget weight to each milestone and only claim EV credit when that milestone is fully complete. No partial credit. No 90% done. Either the milestone is achieved and you book the value, or it isn't and you book nothing. It's the most honest EVM technique for construction because it eliminates the subjectivity that plagues percent-complete estimates.
This term is part of the earned value definitions glossary. For understanding how different measurement techniques compare, see the progress measurement page.
How It Works
You break a work package into discrete milestones, assign each a budget weight (usually based on the cost or effort proportion), and measure progress by counting completed milestones. The cumulative EV at any point is the sum of budget weights for all achieved milestones.
It's binary. Done or not done. That's what makes it powerful.
WEIGHTED MILESTONES – HOW EV ACCUMULATES ============================================= Work Package: Structural Frame – BAC = £4,000,000 Milestone Weight Cum. EV Status ──────────────────────────────────────────────────── M1 Excavation complete 10% £400,000 DONE ✓ M2 Foundations poured 15% £1,000,000 DONE ✓ M3 Columns to level 2 20% £1,800,000 DONE ✓ M4 Slab level 2 cast 20% £2,600,000 IN PROGRESS M5 Columns to roof 20% £3,400,000 NOT STARTED M6 Frame complete 15% £4,000,000 NOT STARTED ──────────────────────────────────────────────────── Current EV = M1 + M2 + M3 = £400K + £600K + £800K = £1,800,000 (M4 is 70% done but gets ZERO credit until 100% complete) EV (£M) | 4.0 ┤ ────── M6 ✓ | ╱ 3.4 ┤ ──── M5 ✓ | ╱ 2.6 ┤ ──── M4 ✓ | ╱ 1.8 ┤ ──── M3 ✓ ← WE ARE HERE (EV = £1.8M) | ╱ 1.0 ┤ ── M2 ✓ | ╱ 0.4 ┤ M1 ✓ |╱ 0.0 ┼────────────────────────────────────── Time Jan Feb Mar Apr May Jun Staircase pattern = earned value only jumps at milestones Flat sections = work in progress, no EV credit yet
The staircase pattern is the hallmark of weighted milestones. Between milestones, EV is flat. When a milestone completes, EV jumps. This creates a lumpy S-curve rather than the smooth curve you'd get from percent-complete measurement, but it's a more truthful representation of progress.
Why It's the Best Method for Construction
I've used three EVM techniques on construction projects: percent complete, fixed formula (0/100 or 50/50), and weighted milestones. Weighted milestones wins. Here's why.
Percent complete is subjective. Ask three people how complete the M&E rough-in is and you'll get three answers. The site foreman says 75% (based on what's installed). The QS says 60% (based on cost). The planner says 65% (based on programme). Who's right? Nobody knows, and the disagreement burns an hour of every progress meeting.
The fixed formula method (50/50 or 0/100) is too crude for packages longer than a few weeks. A 6-month M&E package using 0/100 would show zero EV for 5.5 months then full credit at completion. That's useless for monthly reporting.
Weighted milestones hit the sweet spot. They're objective (the milestone is either done or it isn't), granular enough for monthly reporting (assuming you've set 4-8 milestones per package), and they force the team to define what "done" actually means before the work starts.
Setting the Right Milestones
The quality of weighted milestone measurement depends entirely on how you define the milestones. Too few and you get the same problem as 0/100, long flat periods with no EV movement. Too many and you're basically doing percent complete with extra steps.
The rules I follow:
- 4 to 8 milestones per work package. Fewer than 4 creates reporting gaps. More than 8 starts to feel like micromanagement.
- Each milestone should represent a physical deliverable. "Foundations poured" is a milestone. "Design 50% complete" is not. That's just percent complete wearing a disguise.
- Milestone weights should reflect cost proportion. If excavation is 10% of the work package budget, the excavation milestone gets a 10% weight. Don't assign equal weights unless the milestones genuinely represent equal effort.
- Milestones must be objectively verifiable. Can someone walk on site and confirm it's done without a subjective judgement? If yes, it's a good milestone. If it requires a committee discussion, it's not.
Worked Example: £4M Structural Package on a Mixed-Use Scheme
Worked ExampleScenario: A £4M structural frame package on a £30M NEC4 Option C mixed-use development in Bristol. The package has a 6-month duration from January to June 2026. The CAM sets up weighted milestones for EV measurement.
Milestone definition:
| # | Milestone | Weight | Budget Value | Planned Completion |
|---|---|---|---|---|
| M1 | Excavation complete | 10% | £400,000 | 31 Jan 2026 |
| M2 | Foundations poured and cured | 15% | £600,000 | 28 Feb 2026 |
| M3 | Columns to level 2 complete | 20% | £800,000 | 21 Mar 2026 |
| M4 | Level 2 slab cast and struck | 20% | £800,000 | 18 Apr 2026 |
| M5 | Columns to roof level complete | 20% | £800,000 | 16 May 2026 |
| M6 | Frame complete (roof slab cast) | 15% | £600,000 | 13 Jun 2026 |
| Total | 100% | £4,000,000 |
Monthly progress report (end of March 2026):
| Milestone | Status | EV Claimed |
|---|---|---|
| M1 Excavation | Complete (27 Jan) | £400,000 |
| M2 Foundations | Complete (3 Mar, 3 days late) | £600,000 |
| M3 Columns to L2 | Complete (20 Mar, 1 day early) | £800,000 |
| M4 Level 2 slab | In progress (formwork 70% done) | £0 |
| M5 Columns to roof | Not started | £0 |
| M6 Frame complete | Not started | £0 |
EV at end of March = £400K + £600K + £800K = £1,800,000
Note: M4 is 70% through formwork, but the weighted milestone method gives zero credit until the slab is cast and struck. This is conservative, but honest. The PV at end of March (per the baseline) includes M4 as scheduled for 18 April, so PV at end of March = £1,800,000.
SV = £1,800,000 - £1,800,000 = £0. Exactly on programme.
But here's the nuance: M2 was 3 days late. That 3-day slip was absorbed by completing M3 one day early and starting M4 formwork before the planned date. The milestones show on-programme, and the daily recovery effort is invisible in the numbers, which is exactly how it should work. The CAM managed the slippage within the package without escalation.
Common Mistakes
- Defining milestones that aren't binary. "Steel erection 50% complete" is not a milestone. It's a subjective progress estimate dressed up as a milestone. Real milestones are binary: "Steel erection to level 3 complete" can be verified by standing on site and looking up. Either the steel is there or it isn't.
- Weights that don't match cost. Assigning equal 20% weight to five milestones when the first one represents 5% of cost and the last represents 40% distorts the EV curve. Weight each milestone according to its proportion of the work package BAC.
- Too few milestones on long packages. A 9-month M&E package with only 2 milestones (rough-in complete, commissioning complete) means you'll have months of zero EV movement. Split it into 6-8 milestones covering first fix, second fix, testing, and commissioning stages.
- Changing milestone definitions mid-package. Once the baseline is set, the milestones are locked. If you redefine "foundations complete" halfway through to include backfill (which wasn't originally included), you've corrupted the baseline comparison. Any changes must go through the formal re-baselining process.
How Gather helps. Gather's AI reads your site diaries daily and maps progress against your cost-loaded programme, giving you accurate earned value data without manual spreadsheet updates. Book a demo to see it working on a live NEC4 project.
Frequently Asked Questions
When should I use weighted milestones instead of percent complete?
Use weighted milestones whenever you can define clear physical deliverables, which is most construction work. Percent complete is better suited to continuous activities where there's no natural milestone, like site management or design coordination. For discrete packages (structural, M&E, externals), weighted milestones are almost always the better choice.
What happens to EV reporting during long gaps between milestones?
EV stays flat. That's by design. If there's a 6-week gap between milestones, the S-curve will show a plateau. This can make progress look stalled in dashboard reports. The solution isn't to abandon weighted milestones. It's to break the package into more milestones so the gaps are shorter. Aim for monthly milestone completions at minimum.
Can weighted milestones be combined with other EVTs on the same project?
Absolutely. Most projects use a mix. Weighted milestones for construction packages, level of effort for site management and prelims, and percent complete for design activities. Each work package gets the technique that best fits the nature of the work. The EVM implementation guide covers how to select the right technique for each package type.
.webp)
