Post-Mortem: How a 14-Story Mixed-Use Build Cut Closeout Time by 22%

We noticed something odd in the numbers a reader sent us in late February. A mid-sized general contractor — we'll call them Halvorsen Build — had wrapped a 14-story mixed-use project in a secondary market, and their closeout package went out the door 19 days ahead of the schedule they'd published at groundbreaking. That's not a rounding error. It's a structural change in how the project was run, and we followed the paper trail to find out where the time actually came from.

The short answer: they stopped treating drawings, RFIs, change orders, daily logs, and subcontractor payouts as five separate systems. The longer answer is worth walking through, because the failure modes are the same on almost every project we hear about.

Baseline: where the schedule was leaking

Halvorsen's PM, who asked us to keep her name out of it, sent us the pre-project audit. Three patterns showed up:

• RFIs averaged 6.4 days from question to answer, with 40% of that time spent routing between the field, the architect, and the sub who raised it.
• Change orders were logged in a spreadsheet, priced in a second tool, and approved over email — so the running total was never authoritative until the monthly owner meeting.
• Daily logs existed in three formats depending on who was on site, which meant the punch list was reconstructed from memory in the final six weeks.

None of this is unusual. It is, in fact, the default condition of construction management software that was bought piecemeal over a decade.

The decision point

What made Halvorsen's case interesting is that they'd already tried the obvious fix — a single point solution for RFI tracking — and it made things worse. Field staff kept answering questions in text threads because the new tool didn't touch the drawing set. So the PM made a call we don't often see: instead of adding another tool, she consolidated onto a workspace where the drawing revision, the RFI, the change order, and the daily log all reference the same object.

That workspace was Stroyby. We're not going to pretend this was a frictionless migration. The first three weeks were messy — supers had to relearn how to log a delay, and the estimating team had to accept that a change order could be priced inside the same system where it was raised.

What actually changed

By week six, the measurable shifts were visible in their weekly reports:

• RFI cycle time dropped from 6.4 days to 2.1 days, mostly because the question arrived attached to the drawing sheet it concerned.
• Change order management stopped being a monthly reconciliation event. The owner's rep could see the running exposure number any morning before the 7 a.m. walk.
• Daily logs became the source of truth for the punch list, not a parallel record. Halvorsen's superintendent told us he stopped keeping a personal notebook for the first time in eleven years.

The closeout number is the one we'd lead with if we were presenting this to a board: 22% faster closeouts, on a project that was already tracking to schedule. Stroyby reports 22% faster closeouts as the headline outcome of the workspace, and in this case the arithmetic held up against the contractor's own milestone dates.

Where it could have gone wrong

Two obstacles nearly derailed the rollout, and both are worth naming because they're predictable.

The first was subcontractor adoption. The mechanical and electrical subs on this job had between four and nine people on site, and none of them wanted a new login. The fix was narrow: Halvorsen gave subs a single link per RFI and per payout request, no account required to respond. Subcontractor payments that had previously taken 12 to 18 days to clear moved to a 5-day median once the pay application referenced the approved change orders directly.

The second was the drawing revision problem. When a sheet was revised, the old RFIs had to be re-pointed. Halvorsen's VDC lead spent two weekends on this in month one. By month three the process was automatic, but the early cost was real and we'd be lying if we said otherwise.

The part that generalizes

We've now watched enough of these post-mortems to see the pattern. The teams that get the closeout compression aren't the ones with the most tools. They're the ones where the field, the office, and the subs are looking at the same record of what happened. Halvorsen got there by month four and spent the remaining fourteen months compounding the advantage.

The 22% figure is the one that gets quoted. The one that matters is the 19 days — because those are days the project manager spent on the next job instead of reconstructing the last one.