Percent complete by cost vs. percent complete by schedule

On a draw request, the GC submits a percentage. On the schedule update, the scheduler submits a different percentage. They're not supposed to match, and knowing why they diverge is half of reading a draw package correctly.

Cost-based percent complete: what the draw request shows

The schedule of values breaks the contract into line items, each with a budgeted cost. Percent complete by cost is just cost-to-date divided by that budget, line by line, rolled up to a job total. The contractor's super or PM estimates how much of each line is done, the draw inspector (if there is one) checks it against what's visible on site, and the number goes on the application for payment.

This method answers one question: how much of the contracted value has been earned. It doesn't care about sequence. A line item for structural steel can show 90% complete because the steel is fabricated, delivered, and stored on site, even if erection hasn't started. The money is committed and largely spent, so the cost percentage climbs, but nothing is standing yet.

That's the soft spot in cost-based reporting. Stored materials, mobilization costs, and front-loaded line items can push the cost percentage ahead of what a walk of the site would show. The method measures money committed. That's a different number from work installed, and the gap is built into how cost percent complete gets calculated.

Schedule-based percent complete: what the baseline says should be happening

Percent complete by schedule compares progress against the critical path schedule, the one with durations, dependencies, and a planned finish date. If the schedule says drywall should be 60% done by week 40 and the update shows 60% done, you're on track regardless of what's been paid out. This method answers the question an owner actually cares about: will the building be done when the loan agreement says it will be done.

Schedule percent complete also exposes float burn. A task can show "on track" in terms of dates while quietly eating the float that was supposed to protect the finish date. Cost-based reporting has no way to show that at all. A line item is either paid for or it isn't.

Where the two numbers pull apart

The gap between the two shows up in a few predictable ways:

• Equipment and materials bought early inflate cost percent complete before anything is installed.
• A trade that's behind on the critical path can still be invoicing on schedule if its line item was structured to pay out on delivery rather than on installed, tested, or accepted work.
• Weather delays or a late submittal can stall physical progress for weeks while the cost percentage barely moves, because the next line item hasn't been mobilized yet.
• Conversely, a contractor working ahead of schedule on a cheap, labor-heavy item (say, interior framing) can look behind on cost percent complete even though the building is visibly moving fast.

None of this means either number is wrong. It means they're measuring different things, and a draw package that only shows one of them is missing half the picture.

Why owners and lenders end up tracking both

A construction loan draw review exists specifically to catch the case where the cost percentage says 70% and the building doesn't look like 70% of a building. That's the whole reason inspecting engineers and draw inspectors get hired. The problem is that a monthly site visit is a single data point, reported after the contractor has already submitted the application for payment it's supposed to check.

Closing that gap takes an independent, dated record of what's physically on the ground, taken often enough to compare against the schedule week by week rather than once a month after the draw is already in. A weekly drone capture measured against the project schedule produces that record: an observed percent-complete to set beside the contractor's cost-based number, so a line item that's flagged 90% complete by cost but is really steel sitting in laydown reads as steel in laydown on the progress map, whatever the pay application says.

If you're reconciling cost-based draw requests against what's built on site, that weekly map is worth a look.