Construction cost management software holds four numbers against every cost code, not one. Budget is what the scope was priced at. Committed is what has been contractually obligated — signed subcontracts, issued purchase orders, approved variations — invoiced or not. Actual is what has been incurred and recorded. Forecast to complete is what the job will finish at. Only the fourth changes a decision in time.
Key Takeaways
- Cost control is not cost accounting, and the Construction Industry Institute says so itself.
- Four numbers matter: budget, committed, actual and forecast to complete. Committed cost leads, actual cost lags.
- Cost is committed at the purchase order or subcontract award, not when the invoice arrives.
- An uninvoiced cost is still a cost. Accruals stop month end flattering the job.
- Forecast to complete comes from a person estimating remaining work, so the monthly review is the control.
Cost Control Is Not Cost Accounting
The Construction Industry Institute defines cost control as analysing budgeted and actual expenditure "for purposes of timely identification of cost trends, problems, and opportunities during the course of the project", then closes with a line no vendor would write: "Cost control is not cost accounting" (CII, RS244-1, 1 February 2009).
Cost accounting answers what has been spent, so it only sees cost that reached the ledger. Cost control answers where the job is heading, working with cost not yet invoiced: the subcontract signed last Tuesday for work starting next month. A builder whose only cost tool is the ledger learns of an overrun when invoices land, usually one to three months after the commitment. The wider discipline is in our guide to project controls.
The Four Numbers, and Which One Moves First
Budget is the scope priced at award and loaded against cost codes, revised as variations are approved. It is why the estimate-to-budget handover matters more than the estimate: if the numbers that won the job do not become the numbers it is tracked against, the budget is a document rather than a control. That handover is where construction estimating software connects to delivery.
Committed cost is the obligation — "project expenses formally agreed through a contract or purchase order but not yet invoiced or paid", committed "as soon as the agreement is signed, rather than when the invoice arrives" (Procore AU). The portion not yet invoiced is the open commitment, and it exists nowhere in a general ledger. Actual cost is what has been incurred and recorded, accruals included. Forecast to complete is cost to date plus estimate to complete.
| Number | Where it comes from | What it tells you | What it cannot tell you |
|---|---|---|---|
| Budget | The priced scope, plus approved variations | The baseline for everything else | Whether the baseline was realistic |
| Committed | Signed subcontracts, issued POs, approved variations | Where the job is heading, earliest | What the work will actually cost |
| Actual | Invoiced and accrued cost in the ledger | What has been incurred | Anything not yet invoiced |
| Forecast to complete | Cost to date plus a human estimate of remaining work | Whether the job still makes its margin | Anything, if that estimate is stale |
A subcontract awarded above the budgeted rate moves committed cost first, the first invoice against it moves actual cost weeks later, and forecast to complete moves only when someone re-estimates. The gap between committed and actual is the window in which a decision was available. Where the four numbers sit in a full system is what a construction ERP answers.
Cost Codes and the Cost Breakdown Structure
A cost code is the address a cost lands at; the cost breakdown structure is the shape of the whole address space. It must pass two tests: can a cost be allocated to exactly one code without argument, and can two jobs with the same scope be compared code for code? Failing the second gives consistent, useless numbers.
Australian builders almost universally run trade-based codes — concrete, framing, lock-up carpentry, electrical, plumbing, tiling — rather than the element-based classification common in the US, where MasterFormat codes organise by building element rather than by who installs it (The Construction Standard). Cost code and cost type are separate dimensions: a wet-weather delay moves labour and plant, a price rise moves material.
Three failure modes recur: codes added mid-job, which split a comparison series; one code doing two scopes, which hides an overrun inside a combined number; and a code set so deep the field codes to the wrong line.
Worked Example: One Job, Four Numbers
The numbers below are illustrative. They are invented to show the mechanics, describe no DDDI or Cyberidge project, and are not a benchmark. A $2,000,000 contract sum excluding GST, mid-tier residential-commercial, month six of twelve, awarded at 11 per cent margin (a $1,780,000 cost budget against the $2,000,000 contract sum).
| Cost code | Budget | Committed | Actual | Forecast to complete | Variance |
|---|---|---|---|---|---|
| Site works and excavation | $140,000 | $142,000 | $138,000 | $142,000 | −$2,000 |
| Concrete and structural | $460,000 | $478,000 | $291,000 | $478,000 | −$18,000 |
| Framing and lock-up | $380,000 | $372,000 | $205,000 | $396,000 | −$16,000 |
| Services | $330,000 | $351,000 | $168,000 | $351,000 | −$21,000 |
| Fit-out and finishes | $290,000 | $268,000 | $112,000 | $312,000 | −$22,000 |
| Preliminaries and site overheads | $180,000 | $176,000 | $102,000 | $184,000 | −$4,000 |
| Total | $1,780,000 | $1,787,000 | $1,016,000 | $1,863,000 | −$83,000 |
Read the framing line. Actual cost is $205,000 against a $380,000 budget — roughly half the scope on a job at its halfway point — so a spend-to-date report shows it under budget. Committed cost is $372,000, which says the budget is nearly fully obligated. Forecast to complete is $396,000, so it finishes $16,000 over. Only the forecast column shows it.
At month six the same job gives two answers. Spend to date is $1,016,000 against $1,780,000 — 57 per cent spent against 50 per cent of programme. Forecast to complete is $1,863,000 against the $2,000,000 contract sum: 6.9 per cent gross margin against the 11 per cent it was awarded at.
Variations, Accruals and Retention
Unapproved variations exist in reality and not in the budget. Work proceeds on a verbal instruction, the cost lands as actual, and the budget it is measured against has not moved, so the variance reads as an overrun on original scope when it is a missing revenue line. That is why unapproved variations are constrained rather than recognised in full when measuring contract revenue under AASB 15 (Duguid; AASB 15 B14–B19).
Approved-but-unentered variations are the quieter failure. The variation is signed and the revenue real, but the cost delivering it never reached the budget, so every later month compares actuals against a budget missing a scope line. This is the commonest reason committed cost exceeds budget.
Accruals are costs incurred but not yet invoiced. A subcontractor who completed a pour but has not issued a claim by month end has cost the job money that month. Percentage complete is most commonly actual cost to date over estimated total cost (Procore AU), so understating cost to date understates it too.
Retention is where cost meets cash: money earned and withheld as security, so a balance-sheet item with a release trigger rather than a cost. The claim mechanics sit with construction invoicing software.
The Monthly Review, and Why Margin Leaks
The control is a cadence, not a screen. Every open job, once a month, with someone re-estimating cost to complete rather than restating the original budget. That estimate is a judgement about work not yet performed, and it is biased low. Unprocessed variations and stale tender rates both push it down (WIP Ready).
Margin erosion rarely arrives as a single event. It accumulates through cost overruns, unrecovered revenue and forecast inaccuracies that each sit below the threshold of visibility (Xpedeon) — which is why the software should hold forecast history, not just the current version.
The output is a work-in-progress schedule: earned revenue against billings, per job, as a contract asset or a contract liability, never netted under AASB 15. Over-billed means the client paid ahead of the work performed; under-billed means work delivered and not billed, a cash hole with a claim owed.
Inside a Delivery Business
Cyberidge Construction, the building arm DDDI Group helped equip, runs a formally gated six-stage process from site assessment to handover, each stage with a defined output, a sign-off condition and a named owner. That structure is where stage-level cost control bites. A gate cannot close meaningfully unless the stage's claim schedule and the commitment register agree, and a gate opened on a register nobody has updated is a milestone, not a control. Two gates carry a direct cost test: where cost becomes committed through subcontract award, and where retention is released. You can see how the framework runs on the Cyberidge delivery record. Our builder enablement work starts from the cost code structure and commitment register rather than a dashboard, because a dashboard over an unclassified job is a faster way to be wrong.
Australian Tools That Hold These Numbers
What separates these platforms is not quality but which of the four numbers the product holds — the difference between tools that track a job's cost and tools that hold committed cost against a budget with a forecast.
| Tool | Typical AU fit | Committed cost | Forecast to complete | Published AU pricing | Checked |
|---|---|---|---|---|---|
| Procore | Mid-tier and large commercial, existing users | Yes, from POs and subcontracts | Yes, in budget and cost management modules | Quote only; priced on Annual Construction Volume, unlimited users | October 2026 |
| Jonas Premier | Mid-market general contractors | Yes, with purchase orders and cost plus claims | Yes, advanced reporting on Premium | Starter AU$349 per concurrent user/month from $15,000 implementation; Premium AU$299 from $25,000; Enterprise from $50,000 | October 2026 |
| Buildxact | Residential builders and renovators | Purchase orders on every tier; change orders from Pro | Actual costings; forecast depth not published | Foundation $199, Pro $399, Master $599 per month excluding GST; $169 / $339 / $509 billed annually | October 2026 |
| BuildLedger | Small residential and renovation teams | Supplier invoice capture; commitment register not published | Not published | Base $150, Pro $299, Premium $474 per month, AUD excluding GST; $125 / $249 / $395 billed annually | October 2026 |
| SiteStride | Trade contractors through to builders | Not published | Cost-to-complete forecasting listed on All-In | Estimating $149, Onsite & Trades $249, All-In $399 per month plus GST; $127 / $212 / $339 billed annually | October 2026 |
| Databuild | Established residential builders, estimating through accounting | Within its own job costing and accounting | Not published | Quote only; no pricing on the vendor site | October 2026 |
| simPRO | Larger trade and service businesses, multi-stage work | Within job costing and reporting | Not published | Quote only; no rate card published | October 2026 |
| Cubit (Buildsoft) | Estimators and quantity surveyors | No — estimating and takeoff only | No | Standard $170, Professional $226, Enterprise $255 per user per month plus GST | October 2026 |
Sources: Procore AU, Jonas Premier AU, Buildxact AU, BuildLedger, SiteStride, Databuild, Simpro and Cubit, all checked October 2026.
The tools publishing the clearest prices and the tools holding committed cost against a forecast are largely different sets: Buildxact, BuildLedger and SiteStride publish rates, and none describes a commitment-versus-budget-versus-forecast model in the detail Procore or Jonas Premier do. "Quote only" is not a sign of a bigger product, since Databuild, simPRO and Procore decline to publish for three different reasons. Cubit marks the boundary — an estimating and takeoff tool that holds none of the four numbers. The full comparison is in our construction management software guide.
Sorting the Cost Model From the Software
Fix the cost code structure first, because it decides whether any tool can produce a variance that means anything. Then name which of the four numbers you cannot see. A builder with budget and actual but no committed cost has a commitment register problem, often a process fix before a purchase. A builder with all three but no forecast has a review cadence problem, and no software generates the estimate to complete.
DDDI Group works with builders on the estimating, quoting and cost systems that sit around the books. Start with builder enablement, review the Cyberidge delivery record, or put a real project through Get Started.
Frequently Asked Questions
What Is the Difference Between Cost Control and Cost Accounting?
Cost accounting records and reports what has been spent. Cost control analyses budgeted and actual expenditure to identify trends and opportunities while the project still runs. The Construction Industry Institute states the distinction in the definition itself: "Cost control is not cost accounting" (CII RS244-1). Accounting works from posted transactions, so it cannot see a committed cost not yet invoiced.
What Are Committed Costs in Construction?
Committed costs are expenses formally agreed through a signed contract or issued purchase order but not yet invoiced or paid. The budget becomes committed when the agreement is signed, not when the invoice arrives (Procore AU). Watching the open commitment against budget is how a builder sees an overrun before the invoices land.
Why Does Forecast to Complete Matter More Than Spend to Date?
Because it is the only cost number that still supports a decision. Spend to date records what cannot be changed. Forecast to complete tells you what the job will finish at, the basis for renegotiating procurement on work not yet committed or repricing the next job. In the example above, spend to date showed 57 per cent spent against 50 per cent complete, while the forecast showed margin falling from 11 to 6.9 per cent.
Can Xero or MYOB Do Construction Job Costing?
Partly, and the ceiling is structural rather than a missing feature. A tracking category tags a past transaction; a job cost needs a budget, a commitment and a forecast compared line by line while you can still act. Which plans include job tracking, and at what cost, is in construction accounting software.
What Is the Difference Between Job Costing Software and Cost Management Software?
Job costing software records what a job has cost, by cost code. Cost management software adds the forward-looking side: committed cost against budget, forecast to complete, and variance reporting that supports a decision. That matches the industry's line between cost accounting and cost control, which is why the tables above treat committed cost and forecast to complete as separate columns.