SOFTWARE — JOB COSTING

QUICKBOOKS ISN'T FAILING YOU. IT WAS NEVER BUILT FOR THIS.

QUICK ANSWER

QuickBooks is excellent general business accounting — and construction is not a general business. The gaps aren't bugs; they're architecture: job costing forced through items, classes, and projects workarounds that collapse past a handful of jobs; no native WIP schedule or percentage-of-completion engine, so revenue follows billing and the P&L lies in both directions; no SOV-based pay application billing, so every G702 gets hand-built in Excel; no cost-to-complete, no committed cost tracking, no retainage management worth the name. Subs make QB work to about $2M–$3M with heroic spreadsheet support. Past that, the workarounds consume more hours than the software saves — and the numbers they produce stop being trusted by the people who need them most: you, your bank, and your surety.

THE PROBLEM ISN'T THE LOGO ON THE SOFTWARE. IT'S THAT YOUR PM CAN'T SEE JOB COST IN 30 SECONDS — AND IN QUICKBOOKS, THEY NEVER WILL.

BY JOSH LUEBKER Published: June 2026 Updated: June 2026

THE GAPS

WHERE THE ARCHITECTURE RUNS OUT.

GAP 01 — JOB COSTING BY WORKAROUND

Items, Classes, and Projects Were Never Cost Codes

QB's job costing is an improvisation: items doubling as cost codes, classes doubling as jobs or divisions, the Projects feature bolting a job dimension onto a system that thinks in customers and invoices. It demos fine with three jobs. At twelve active jobs with two-level cost codes, labor burden allocation, and committed subcontracts, the structure collapses into miscoded costs and a chart of accounts nobody can read — and the reports either say too little (material, labor, other) or so much (1,500 unreadable lines) that nobody trusts them. Untrusted numbers are unmanaged jobs.

GAP 02 — NO WIP, NO POC

Revenue Follows Billing, So the P&L Lies

Construction earns revenue by progress; QuickBooks recognizes it by invoice. There is no native WIP schedule, no percentage-of-completion engine, no over/under-billing math — so a front-loaded SOV makes a break-even month look great, a slow billing month makes a profitable company look broken, and the owner manages by billing noise instead of earned reality. The WIP gets rebuilt monthly in Excel by hand, which means it's usually quarterly, stale, and unreconciled to the books your bank reads.

GAP 03 — PAY APP BILLING DOESN'T EXIST

Every G702 Is a Manual Excel Project

Commercial billing runs on schedules of values, percent-complete-by-line pay applications, stored materials, and retainage withheld and tracked per job. QuickBooks invoices. The gap gets filled with Excel pay apps maintained outside the system, retainage tracked in someone's head or a side spreadsheet, and a billing process slow enough that pay apps miss GC cutoffs — which converts a software gap directly into a working capital problem. Billing velocity is the cheapest cash lever in construction, and QB structurally caps it.

GAP 04 — NO FORWARD NUMBERS AT ALL

Cost to Complete, Committed Costs, and the Questions QB Can't Answer

QuickBooks records what happened. Construction management runs on what's left: cost to complete by line, committed subcontract balances against budgets, projected final margin by job, the variance between estimate and actual at the cost-code level. None of it exists natively — so the highest-value financial questions in the company get answered by spreadsheet archaeology or not at all. The test from Chapter 5 of the playbook: can your PM pull up where a job stands, right now, in 30 seconds? In QuickBooks, the honest answer is never.

THE FIX

WHAT REPLACES IT — AND WHAT IT COSTS.

The fix is construction-native job costing — real cost codes matching your estimate, a live WIP, SOV pay app billing, retainage, committed costs, and CTC — sized for a $1M–$12M sub rather than priced like a $50M ERP. That last part matters: the legacy construction platforms solve these gaps at $30K–$100K+ a year with six-month implementations, which is why subs stay on QuickBooks a decade too long.

SPM runs clients on ControlQore — construction-native job costing and WIP at roughly $150/month per $1M of revenue, so a $4M sub pays about $600/month for the system the big platforms charge ten times for. The software alone isn’t the answer, though: the same gaps reappear in any platform set up by someone who doesn’t know construction. The 60-day SPM install builds the cost structure to match your estimating, migrates the books, and runs the close — software and judgment together. Same business. Better system.

BY TRADE

WHERE QB BREAKS FIRST, TRADE BY TRADE.

Concrete & Structural

Breaks at burdened labor: QB payroll posts wages, but comp by class code, taxes, and benefits never land per job per phase — so labor-heavy concrete jobs look profitable until year-end. The $4.9M sub netting $161K was managing exactly this mirage.

Civil & Equipment-Heavy

Breaks at equipment: no cost-basis rates, no per-machine charging to jobs, idle time invisible. Thirty machines' costs smear into overhead while jobs get billed made-up hourly rates — the $779K balance-sheet swing one civil client found was sitting in this exact gap.

Electrical & Multi-Phase

Breaks at phases and changes: rough-in versus trim versus closeout need phase-level tracking QB's structure fights, and forty small COs need contract-value and SOV management that doesn't exist. Blended job totals hide which work types make money — the visibility that turned a $2.3M electrical sub around.

SWPPP & Multi-Site

Breaks at volume: forty sites as forty QB 'customers' with no per-site costing rollup, no site-level margin, no consolidated per-GC view. The $5.2M erosion contractor netting $24K had QuickBooks. The same company netting $1.1M had per-site job costing.

WHAT CHANGES WHEN THIS IS FIXED

AFTER THE SWITCH, ON THE RECORD.

30 Sec

The PM test, passed. The real benchmark of working job costing: a PM pulls up any job and answers 'where are we spending money' in thirty seconds, live, without asking accounting. That's the daily reality on a construction-native system with the cost structure built right — and the structural impossibility QB was hiding.

8–10 Hrs

What reports used to take. Now instant. A $25M marine GC ran on spreadsheets — every report was an 8–10 hour rebuild and one person was the single point of failure. The system change made reporting instant, unlocked $10M in bonding, and ended the keyperson risk. The same arc plays out at $3M, just with smaller spreadsheets.

~$150

Per month per $1M of revenue. ControlQore pricing scales with the business: a $4M sub runs about $600/month for construction-native costing, WIP, and pay app billing — against $30K–$100K+ annually for the legacy platforms solving the same gaps. The barrier to leaving QuickBooks stopped being price years ago.

Frequently Asked Questions

To a point — and knowing the point matters. With disciplined items-as-cost-codes setup, the Projects feature, and serious spreadsheet support for WIP and pay apps, QB can carry a sub to roughly $2M–$3M with a handful of concurrent jobs. The tells that you've crossed the line: the WIP lives in Excel and gets updated quarterly, pay apps regularly miss GC cutoffs, nobody can state last month's margin by job, and your bookkeeper has invented a workaround vocabulary. Some subs break it at $1.5M with multi-phase work; a few stretch it past $3M with heroics. The question isn't whether QB can be made to limp — it's whether the limping costs more than the fix.

Two costs: software and setup. ControlQore runs roughly $150/month per $1M of revenue — about $600/month at $4M — versus the legacy construction platforms at $30K–$100K+ per year. Setup is the part that decides success: SPM's 60-day install migrates the books back to the start of the last taxable year, builds cost codes that match your estimating, and stands up the WIP, billing, and close cadence — included in the engagement (Core Financial from $1,900/month). The cost most subs never calculate is staying: the spreadsheet hours, the missed billing cutoffs, the bids priced off invisible costs. That bill is already being paid monthly.

No — the migration brings it with you. SPM migrates books back to the start of the last taxable year (and if your tax year ends within three months, through that year-end so the new system starts clean on day one of the new year). Historical QB data stays accessible read-only for reference and audit trail. What doesn't migrate is the workaround structure — the items-as-codes improvisation gets replaced by real cost codes, which is the point. Most owners find the first month's reports cover questions their history never could anyway: margin by job by phase, live, reconciled.

Gently and honestly: this usually isn't a competence problem, it's a tooling problem. A good construction bookkeeper trapped in QB spends their week maintaining workarounds — Excel WIPs, manual pay apps, reconciliation archaeology — and most of them know the numbers aren't quite right better than anyone. The conversation that works: the system change removes the workaround load, not the person; their job becomes coding and reconciliation inside a structure that catches errors instead of hiding them. The ones who resist hardest are usually protecting the workaround knowledge that made them indispensable — which is itself the keyperson risk the switch is solving.

No — and this is the honest part most software pitches skip. The same failure modes that broke QB job costing (codes that don't match the estimate, costs landing late and miscoded, no close discipline, nobody owning the WIP) reproduce perfectly inside any construction platform set up generically. The software removes the structural ceiling; the setup and the operating discipline determine whether you actually get visibility. That's why SPM's model pairs ControlQore with the 60-day install and the ongoing close — the system is software plus structure plus cadence. Any one alone is how subs end up paying for two failed implementations before the third one works.