The Short Version
The job cost report is the most valuable document in construction finance — and the one most builders either ignore or misread. The report itself isn't complicated. But it requires knowing which five numbers to focus on, what a problem looks like in each one, and what action that problem requires. Builders who check these five numbers weekly catch overruns while they can still do something about them. Builders who don't check them find out a job lost money after the job is done and the money is gone.
Sound Familiar?
Signs your job cost review practice has gaps:
- You check the bank balance more often than you check the job cost report
- Your job cost report shows costs-to-date but not where the job will finish — you're looking backward, not forward
- Change orders approved last month haven't been added to the job budget — so the report shows you under budget when you're actually on budget or over
- You know which jobs are profitable by feel, not by report — and the feel has been wrong before
- The job cost report lives in QuickBooks but nobody knows how to run it, so the PM works from a spreadsheet instead
- You review job costs monthly at best — and by then, there's nothing left to do about a cost overrun
What We Found
The 5 Numbers — What Each One Tells You and Why It Matters
The weekly job cost review isn't about comprehensive financial analysis. It's about five headline numbers, checked in order, that tell you whether a job is on track and what action to take if it isn't. Here's what each number means and why it belongs in the weekly review.
Number 1: Original Budget
The original budget is your baseline — what the job was priced at before any approved changes. This number should never change after the contract is signed. It's the reference point against which everything else is measured. If you're comparing current costs against a budget that's been updated to include approved change orders without tracking the original separately, you lose the ability to assess whether the job is performing against the original estimate.
Why it matters in the weekly review: It tells you the job's original margin target. Everything else is measured against it.
Number 2: Committed Costs
Committed costs are everything you've spent plus everything you're contractually obligated to spend — approved sub invoices not yet paid, purchase orders for materials on order, labor hours accrued but not yet invoiced. This is not the same as costs invoiced or costs paid. It's the full financial exposure you've already locked in on the job.
Most builders track invoiced costs. Committed costs are more accurate because they include obligations that haven't billed yet. A job where framing is 80% complete has committed the full framing sub contract even if only 40% has been invoiced. The committed cost number sees that. The invoiced cost number doesn't.
Number 3: Projected Cost at Completion
Projected cost at completion is where the job will finish based on current performance. It's a forward-looking number built from committed costs plus the estimated cost to complete remaining scopes. It requires judgment — not just math — because you have to estimate remaining work based on current conditions.
A simple calculation: divide committed costs by percentage complete. If you've committed $280,000 and the job is 65% complete, your projected cost at completion is $430,000. If the original budget was $400,000, that's a $30,000 projected overrun — and you know about it now, not after the job closes.
Number 4: Current Projected Profit
Current projected profit is the contract value minus the projected cost at completion. This is the number that answers "will this job make money?" in real time. Not "did it look good when we bid it?" Not "has it been profitable so far?" — those are backward-looking questions. Current projected profit is forward-looking: if everything from here plays out as expected, what does the job end at?
This number should be compared to the target profit from the original estimate. If the original target was 14% net margin and the current projected profit is showing 8%, something has changed — and it's not too late to find out what and whether it's recoverable.
Number 5: Pending Change Orders
Pending change orders are scope changes that have been identified and priced but not yet approved by the client. They represent potential revenue that hasn't been locked in and potential cost exposure that's accruing whether or not the change order gets signed. A job with $22,000 in pending change orders is in a different position than the report shows if those change orders aren't factored in.
The pending change order number tells you: how much revenue is sitting unsigned, what's the risk if the client declines some of it, and does the PM need to close out approvals before the next billing cycle.
How to Read the Five Numbers Together — and What to Do When Something's Off
The five numbers tell you one story when read together. Here's how to interpret the combinations that show up most often:
Scenario 1: Committed costs tracking to budget, projected profit on target, no pending COs
This job is clean. Spend 30 seconds on it in the weekly review: "Job 14, on budget, 12% projected profit, no pending changes." Move to the next job.
Scenario 2: Projected cost at completion exceeds original budget
This is the most important scenario to catch early. There are three causes: cost overrun in the field, scope that wasn't in the original budget has been performed without a change order, or estimating assumptions that were off. Determine which it is. If it's a field overrun, review the specific cost codes that are running high — labor productivity? Material waste? Sub overbilling? If it's unbilled scope, write the change order today. If it's an estimating miss, note it for the next bid and assess what, if anything, can be recovered on this job.
Scenario 3: Committed costs look fine but a large pending CO is unresolved
The job looks healthy but there's $15,000 sitting unsigned in pending change orders. That's a collection risk and a schedule risk. Assign the PM to get approval or a decision this week. Don't let pending change orders sit — the longer they're open, the harder they are to close, and the more leverage the client thinks they have.
Scenario 4: Original budget is outdated — approved COs not reflected
This produces a false picture of the job. The budget shows $400,000, approved change orders add $45,000, but the budget hasn't been updated. Every number calculated against the stale budget is wrong. Fix the budget first, then re-run the review. This is a process failure — approved COs should update the budget within 48 hours of approval.
When to escalate to owner review vs. PM decision:
Any scenario where the projected profit has dropped more than 5 margin points from the original estimate needs owner visibility, not just PM action. At that level, decisions about scope cuts, negotiated settlements, or client conversations may be needed — and those decisions belong to the owner, not the PM alone.
Why Most Builders Don't Review Job Costs Weekly — and What That Actually Costs
I've asked hundreds of builders why they don't run the weekly job cost review. The answers cluster into three categories: "I don't know which report to run," "it takes too long," and "by the time I see it, it's already in the numbers."
All three are fixable — and the fix for each is different.
"I don't know which report to run" — This is a setup problem, not a discipline problem. In JobTread, the report you want is Budget vs. Actuals with committed costs, run at the job level. In QuickBooks, it's a combination of the P&L by Job and the Open Purchase Orders report. If you're not sure which report to run, spend 30 minutes with your accountant or a JobTread consultant setting up a template. The setup investment is a one-time cost. The weekly discipline compounds every week after that.
"It takes too long" — If pulling job cost data takes more than 10 minutes per job, the report setup needs simplification. The review itself should take 2–3 minutes per job: pull the five numbers, check them against expectations, assign actions if needed, move on. If you have 8 active jobs, that's 20–25 minutes total. If it's taking an hour, you're not reviewing — you're auditing. Save the deep audit for a monthly or quarterly review.
"By the time I see it, it's already in the numbers" — This is the real problem for builders who review monthly instead of weekly. A cost overrun that shows up in the monthly review is 3–4 weeks old. With a weekly review, the same overrun shows up with 3–4 weeks of remaining work left on that scope — enough time to tighten production, renegotiate with a sub, or send a change order. The weekly review is valuable precisely because the numbers aren't fully cooked yet.
The financial cost of not running the weekly job cost review is real and calculable. Industry data suggests builders who catch overruns mid-job recover 40–60% of the variance through action — scope cuts, change orders, productivity adjustments. Builders who find out at job close recover 0%. On a $2M company running 5–6 jobs per year, a 5% average overrun rate produces $100,000 in below-target profit. Half of that is recoverable with a weekly review. That's $50,000 per year in margin that literally comes from spending 20 minutes every Monday morning.
Ready to Fix Your Numbers?
Book a $950 Diagnostic. 90 minutes with Grant — he reviews your P&L, job costs, and ops, then gives you a ranked list of fixes with dollar recovery estimates. 312+ builders served.
Book Your $950 Diagnostic →Frequently Asked Questions
JobTread's job cost reports include committed costs (contracted obligations not yet invoiced) and are designed specifically for active job management — they give you a forward-looking picture with budget vs. actuals and change order tracking built in. QuickBooks job cost reports show invoiced and paid amounts by job class — they're accurate for accounting purposes but lag behind reality because they only capture posted transactions. For the weekly review, use JobTread if you have it. Use QuickBooks supplemented by an open POs report if you don't.
Weekly, every week. The 20–30 minute weekly review is not optional — it's the mechanism by which the owner and PM stay aligned on job financial performance while there's still time to act. Monthly is too slow. Daily is operationally impractical. Weekly is the right cadence because it's frequent enough to catch problems before they're entrenched and infrequent enough not to become noise.
Yes, but the review looks different. For fast-turn jobs, check in at three milestones: after the rough work is committed (usually 40–50% into the job), after mechanicals are complete (70–80%), and 2 weeks before scheduled completion. For jobs this size, a weekly meeting may not add value — but a milestone check at each of those three points will catch most overruns while work remains.
Yes. PMs who have visibility into job cost performance make better field decisions — they know when they have room to accommodate a client request vs. when they need to hold the line on scope. Hiding financial data from PMs creates a principal-agent problem: they're managing the job without information that would change their behavior. Share the five numbers, explain what they mean, and hold PMs accountable for the projected cost at completion. That's how you build a team that manages margin, not just schedule.
Almost always one of two things: committed costs aren't captured (only invoiced costs are showing), or approved change orders haven't been added to the budget. Check both. If the framing sub is 80% done but only 30% invoiced, committed costs should reflect the full contracted amount — not just what's been billed. And if you approved a $25,000 change order last week, that $25,000 should be added to the budget today. A report that says you're under budget on a troubled job is a data problem, not good news.
