The project manager has a number. Accounting has a different number. The owner has a spreadsheet with a third. Every month there is a meeting where the three are reconciled, and every month the meeting takes two hours and nobody leaves confident.
This is close to universal in contractors between twenty and three hundred people, and it is not a software problem in the way it first appears.
Why the numbers differ
They differ because they are measuring different things at different moments, and nobody has written down which one is the number.
Accounting's figure reflects costs that have been coded and posted. It is accurate and it is late. Field costs arrive with the timesheet, supplier costs arrive with the invoice, and the invoice arrives when the supplier gets round to it.
The project manager's figure reflects commitments: what has been ordered, what has been promised to a subcontractor, what the crew actually spent this week. It is timely and it is uncontrolled.
The owner's spreadsheet reflects whichever of the two they trusted last, adjusted by instinct.
None of these people is wrong. There is simply no agreed definition of committed cost, no agreed cut-off, and no single place where the answer lives.
The fix, in order
Define the number before choosing where to keep it. Cost to date, committed cost, cost to complete, and forecast final cost. Four terms, four written definitions, agreed by operations and finance. This is a ninety-minute conversation that most contractors have never held, and it resolves more disputes than any system purchase.
Decide where commitments are captured, and make it the only place. Purchase orders and subcontracts have to enter the system when they are made, not when they are invoiced. This is the single largest source of surprise on jobs, and it is a discipline problem before it is a tooling problem.
Fix the timing of field data. Weekly labour, coded to cost codes, entered by the people who did the work. Daily is better. Monthly is how jobs go wrong quietly for four weeks.
Give the number one owner. Somebody is accountable for the forecast being right, and it should be the person who can change the outcome, which is the project manager, with finance responsible for the integrity of the underlying data.
Then look at the software. Once the definitions and the ownership are settled, the question of whether the current system can support them becomes answerable. Asked in the other order, you buy a system and reproduce the same argument inside it.
What it is worth
The value is not tidier reporting. It is that a job in trouble becomes visible in week six instead of week sixteen, when there is still something to be done about it: a change order to pursue, a sequence to adjust, a conversation to have with the owner before it becomes a claim.
Contractors who get this right describe the change the same way. They stopped finding out about losing jobs at the end of them.
The bonding and banking dividend
There is a second benefit that shows up outside the business. Sureties and lenders read work in progress schedules closely, and an accurate, timely, consistently prepared schedule affects capacity and terms. A contractor whose numbers move around between reporting periods gets treated as a higher risk, correctly.
That relationship is worth more than most contractors realise, and it improves as soon as the internal number stops being contested.