Construction ERP: Getting Project Costing, BOQ and Subcontractor Billing to Agree

Construction is one of the few industries where a company can be busy, respected and quietly losing money on three out of five sites. The reason is structural. Costs are incurred daily across dozens of heads by people who are not accountants, revenue is recognised on measurement certificates that lag by weeks, and the true margin only becomes visible when the job is finished and it is far too late to do anything about it.

Why do construction projects lose money without anyone noticing?

Three habits cause most of it.

The first is treating the BOQ as the costing structure. A bill of quantities exists to bill a client. It is organised by measurable items at agreed rates: excavation per cubic metre, PCC per cubic metre, reinforcement steel per tonne. Your cost, meanwhile, is organised by material, labour, plant hire, subcontract and site overhead. When a company records cost only against the BOQ line, it can tell you that item 4.2 has consumed forty lakh rupees but not whether that was steel, a subcontractor or three months of a hired crane.

The second is ignoring committed cost. Accounting systems record invoices. A construction project is governed by commitments: purchase orders raised, subcontracts awarded, work orders issued. By the time the invoice arrives, the money is already spent in every practical sense. A project showing sixty percent of budget consumed on invoices and ninety-five percent on commitments is not at sixty percent.

The third is recalculating work in progress once a quarter, from memory. If the site engineer estimates progress rather than measuring it, the WIP figure inherits every optimistic assumption on site.

How should a construction ERP structure a project?

Start with a project hierarchy that survives contact with reality. A workable structure has four levels: project, then sub-project or block, then activity or work package, then cost head. The BOQ maps into activities. The cost heads sit underneath.

Against every activity the system should carry five figures:

1. Budget — the estimated cost from the tender build-up, frozen at award.
2. Revised budget — updated for approved variations and change orders only.
3. Committed — purchase orders and subcontracts issued, less what has already been invoiced.
4. Actual — cost posted from material issues, labour, plant, subcontractor bills and allocated overhead.
5. Forecast at completion — actual plus committed plus a re-estimate of the remainder.

The fifth number is the one that matters and the one most systems omit. Budget versus actual is a report about the past. Forecast at completion is a decision tool.

How does material get costed correctly on site?

Material issued to a site is not consumed the day it arrives. Treating a delivery as an expense inflates cost in the month of receipt and deflates it later, which makes monthly margin meaningless.

The clean sequence is: goods receipt into a site store, issue against a specific activity through an indent, and consumption recorded when the work is executed. Cement, steel and aggregate should be reconciled physically at month end. A steel reconciliation comparing theoretical consumption from the bar bending schedule against actual issues is one of the highest-value controls available to a contractor, and it takes one person a day a month.

Plant and machinery follows the same logic with different units. Own equipment should be charged to activities at an internal hire rate covering depreciation, operator, fuel and maintenance, so an activity that used a crane for nine days carries that cost even though no external invoice exists.

What does correct subcontractor billing look like?

Subcontractor running account bills are where good projects turn bad, because each bill contains half a dozen deductions applied in a specific order. Getting that order wrong overpays early and creates an unrecoverable balance later.

The standard sequence for an RA bill is:

1. Measure work executed to date against the subcontract BOQ.
2. Value it at contract rates to get gross value to date.
3. Deduct the gross value certified in all previous bills to get the current bill value.
4. Apply retention at the contracted percentage of the current bill value.
5. Recover the proportionate share of any mobilisation advance.
6. Recover the value of material issued by the main contractor, at the recovery rate agreed in the subcontract, not at market rate.
7. Apply penalties, debit notes and back charges.
8. Apply GST on the taxable value, and TDS plus any labour cess as applicable.
9. Arrive at net payable.

Every one of those steps should be computed by the system from stored contract terms. Where they are calculated in a spreadsheet by a billing engineer, errors are not occasional; they are systematic, because the same spreadsheet gets copied to the next bill.

Retention deserves particular attention. It is a liability with a release condition, usually half on virtual completion and half after the defect liability period. An ERP should be able to answer, at any moment, how much retention is outstanding, to whom, and when it becomes due. Contractors who cannot answer that are financing their subcontractors without knowing it, or being financed by them.

How does GST on works contracts affect the ledger?

Works contract services in India are treated as a supply of services and generally attract eighteen percent GST, with input tax credit available to the contractor on inputs and input services. Certain government and infrastructure works carry concessional rates, and those must be mapped at the contract level rather than assumed.

Two practical consequences follow. First, the item-level tax mapping on client RA bills has to be right, because a wrongly rated invoice either understates your liability or overcharges a client who will dispute it. Second, input tax credit is only as good as your subcontractor compliance. If a subcontractor does not file, your credit is at risk, so the ERP should flag suppliers whose filings are not reflected before their payments are released.

What separates a working implementation from a shelf-ware one?

Site adoption. A construction ERP fails at the site office, not at head office. If recording a material receipt takes eleven fields and a stable internet connection, the storekeeper will write it in a register and enter it on Saturday, and your data will be a week old forever.

The implementations that stick share four traits. Site transactions are mobile-first and work offline. Measurement books are captured where the measurement happens. Approvals are three taps, not a desktop workflow. And the site team gets something back, usually a live view of their own indent status and pending bills, so the system is not purely an extraction device pointed at them.

SyvaSoft builds its Construction ERP around that BOQ-to-cost-breakdown mapping, with RA billing, retention tracking and subcontractor recovery handled as configured contract terms rather than manual calculations. The same discipline carries over to its RMC Plant ERP, where the concrete supply side of a project needs to reconcile against the same job.

Where should a contractor start?

Do not begin by buying software. Begin by taking one live project and building the five-figure view for a single month: budget, revised budget, committed, actual and forecast at completion, by activity. It will take a week and it will be painful, because the data is scattered. That exercise tells you exactly which processes need a system and which need a policy. Most contractors discover that half their problem is not software at all. It is that nobody was ever made accountable for the committed cost number.