Here's a pattern that will be familiar to most Australian service and trade businesses. A job wraps up mid-month. The client is happy, the crew has moved on, the invoice goes out. Then, three or four weeks later, the bookkeeper finishes reconciling the month and someone finally asks the question that should have been answered weeks ago: did we actually make money on that one?
Sometimes the answer is yes. Sometimes it's “barely”. And sometimes it's a quiet, uncomfortable no — a job that looked fine on the quote but got eaten alive by extra hours, a parts price that had gone up since the last order, and a variation nobody wrote down.
By the time you know, the job is long closed. You can't fix it. Worse, you've probably quoted several more jobs in exactly the same way while you were waiting to find out.
This post is about why that delay happens, why it's a systems problem rather than a bookkeeping one, and the realistic options for seeing margin while a job is still running — most of which don't involve buying anything.
Why Job Margin Only Shows Up at Month-End
Margin on a job is a simple calculation: what you charged, minus what it cost you to deliver. The reason it takes a month to produce isn't that the maths is hard. It's that the inputs live in different places, entered at different times, by different people.
In a typical 5–50 person service business, a single job's financial picture is scattered across something like this:
| Piece of the Puzzle | Where It Usually Lives | When It Becomes Visible |
|---|---|---|
| The quote & agreed price | A PDF in email, a CRM, or a quoting tool | At the start — then rarely looked at again |
| Labour hours | Timesheets, a rostering app, or handwritten notes | Payroll run — often fortnightly, after the fact |
| Parts & materials | Supplier invoices and statements | When the supplier bills you — days or weeks later |
| Subcontractor costs | Subbie invoices, sometimes verbal agreements | Whenever the subbie gets around to invoicing |
| Variations & extras | Text messages, phone calls, memory | Sometimes never — they just get absorbed |
| The invoice | Xero, MYOB or QuickBooks | When someone raises it |
None of these systems knows about the others. So once a month, a human — usually the bookkeeper, sometimes the owner on a Sunday night — pulls all of it together, matches costs to jobs as best they can, and produces a picture of what happened. It's accurate-ish, and it's always late.
We've written before about why small businesses end up running on half a dozen disconnected systems. Job costing is where that fragmentation stops being an annoyance and starts costing real money, because the delay isn't cosmetic — it directly shapes your pricing.
It's a Systems Problem, Not a Bookkeeping Problem
When margin reporting is slow, the instinct is often to push the bookkeeper: can we get the month closed faster? Can we reconcile weekly?
That misses the point. Your bookkeeper is almost certainly doing the job correctly. The delay is structural: they can't allocate a cost to a job until the cost arrives in a system they can see, and most of the costs arrive late by design. Supplier invoices lag the purchase. Timesheets lag the hours. Variations lag reality by however long it takes someone to remember them.
Reconciliation, by definition, happens after the fact. You can make after-the-fact faster, but you can't make it during. Getting margin visibility while a job is open requires something different: costs recorded against the job at the moment they're incurred or committed, not when the paperwork catches up.
That's a question about how information flows through your business — which is why it belongs in a conversation about systems, not a conversation about bookkeeping effort.
What the Delay Actually Costs You
The direct cost of a losing job is obvious. The hidden cost is what happens during the blind spot.
The real damage is repetition. By the time month-end reveals that Job A lost money, you've usually quoted Jobs B, C and D using the same rates, the same labour estimates and the same parts assumptions. A one-off bad job is a lesson. A month of invisibility turns it into a pattern you paid to repeat.
A clearly illustrative example — not real numbers, just the mechanics: suppose a trade business quotes a recurring type of job assuming two days of labour, and in practice those jobs have been averaging two and a half days because site conditions changed. If margin only becomes visible at month-end, every similar job quoted during those weeks carries the same underestimate. Nobody did anything wrong. The information simply arrived after the decisions that needed it.
There are second-order costs, too:
- No chance to intervene. A job trending over budget at the halfway point can often be rescued — a variation raised, scope clarified, a cheaper supplier found. A job discovered to be over budget after close cannot.
- Variations leak. When extras are agreed verbally and reconciled later, some never make it onto the invoice. If the only record is memory, you're billing from memory.
- Pricing drifts on vibes. Without timely per-job numbers, quoting defaults to “what we charged last time” — which bakes in last quarter's costs during a period when materials and wages have been anything but stable.
- You can't tell good work from busy work. Some job types, clients or crews are reliably more profitable than others. Month-level averages smooth that signal away.
What “Seeing Margin While the Job Runs” Actually Requires
Strip away the software branding and live job costing needs four things:
- One record per job that the quote, the hours, the purchases and the invoice all attach to — however loosely.
- Costs captured at the time, not at reconciliation. Hours logged the day they're worked. Parts recorded when ordered, not when the supplier statement lands.
- Committed costs counted, not just billed ones. If you've ordered $4,000 of materials, the job has spent that money even though no invoice exists yet. (Figure illustrative.)
- A running comparison against the quote that someone actually looks at — ideally the person running the job, not just the office.
One important caveat: how you cost jobs — how overhead is allocated, how labour is rated, how WIP is treated — is an accounting method question. Agree it with your accountant before you build process or buy software around it. A fast number calculated the wrong way is worse than a slow number calculated correctly.
Your Realistic Options, From Cheapest to Most Involved
1. Tighten process inside the tools you already have
Often the biggest single improvement costs nothing: put a job number on everything. Every timesheet line, every supplier order, every invoice line in your accounting file carries the job reference. Then run a short weekly “open jobs” review — fifteen minutes comparing hours and known purchases against each open quote. It's manual, and it won't catch everything, but it shrinks the blind spot from a month to a week.
2. Use the job costing features you're already paying for
Xero, MYOB and QuickBooks all offer project or job tracking in some form, and many businesses pay for tiers that include it without ever switching it on. If your team will reliably code transactions to jobs, this can take you a long way. Its limit is usually the labour and committed-cost side — accounting software only knows about costs once they're entered, so the discipline problem remains.
3. Add a dedicated job management tool and connect it
Field service and job management apps exist precisely for this, and connecting one to your accounting file can work well. The trade-off is another subscription, another login for the crew, and an integration to maintain — we've covered the honest trade-offs in whether to integrate your systems or replace them, and the ongoing cost side in what business software actually costs an Australian SMB. Per-seat pricing deserves particular attention here, because job costing only works if everyone who incurs costs has a login.
4. Consolidate onto one operational record
The most structural fix is to stop splitting a job across systems at all: the quote, the hours, the purchases, the variations and the invoice live on one record, so margin is a lookup rather than a reconciliation. For some businesses that's an off-the-shelf platform. For businesses whose workflow doesn't fit a template, it can be a built-for-you system — this is what our iTec HQ platform does, built around how your business already works rather than the other way round, with modular pricing (you only run and pay for the modules you need, and it isn't priced per seat) and sign-in through your own Microsoft 365 tenant.
If your job costs are spread across a CRM, a job book, timesheets and spreadsheets, it's worth seeing what one record per job looks like. Most builds are live in two to four weeks, hosted in Sydney.
See iTec HQ →How to Decide Which Option Fits
A simple framework:
- If costs aren't being coded to jobs at all — start with option 1. No software fixes a discipline gap; it just makes it more expensive.
- If coding happens but the picture is still slow — the lag is probably labour and committed costs. Look at options 2 and 3, and ask specifically how each handles hours and purchase orders, not just bills.
- If the picture exists but nobody trusts it — you likely have the same job represented differently in two systems. That's a consolidation problem, which points at option 4.
- If the tools work but nobody looks — that's a management rhythm problem. Put a weekly open-jobs review in the calendar before touching anything else.
And whichever path you take, involve two people early: your accountant (on costing method) and whoever handles your IT (on how the pieces connect and who can see what). If you don't have the latter, this is exactly the kind of question a managed IT partner should be helping you think through — it's a workflow and systems decision wearing a software costume.
The Bottom Line
Finding out a month late whether a job made money isn't a sign of bad bookkeeping. It's the predictable output of job data scattered across systems that only meet at reconciliation. The cost isn't just the occasional losing job — it's every job quoted on stale assumptions while you waited for the numbers.
Start with discipline: job numbers on everything, a weekly open-jobs review, and a costing method your accountant has signed off on. Then, if the blind spot persists, fix the structure — whether that's switching on features you already pay for, connecting a job management tool, or putting the whole job on one record. The goal is the same in every case: margin you can see while there's still time to do something about it.