How cost accounting and control is set up in finance and ERP systems

Cost accounting in an ERP sits in a controlling module that shares master data with the general ledger. Cost centers, profit centers and product cost records are configured up front, costs flow in from postings, allocation cycles redistribute them at period end, and variance and profitability reports read the result.

The structures underneath

Everything depends on the cost objects defined before any transaction posts. The usual building blocks are cost centers for departments and shared services, profit centers for segments of the business that management wants to see as their own small P&L, and internal orders or projects for spend that needs tracking beyond a single cost center. Cost pools collect indirect spend that will later be spread across these objects.

These objects hang off hierarchies. A cost center hierarchy mirrors the management structure. A profit center hierarchy often follows product lines or regions, and it rarely matches the legal entity structure. When the two hierarchies drift apart from how the business is actually run, reports become hard to explain.

Setup also covers closing objects down. Projects end and departments merge. Old cost centers that stay open attract stray postings, so most teams lock them for new entries and keep them visible for history.

How costs get in

Most cost data arrives without anyone in cost accounting touching it. Each general ledger account that carries expense is mapped to a cost element, and the posting carries a cost center or order as an account assignment. Purchase invoices, payroll journals, depreciation runs and travel expenses all land this way.

The weak point is the account assignment at the point of entry. If a buyer or an approver picks the wrong cost center, the error flows straight through. Validation rules and default assignments on vendor or employee records reduce this. They do not remove it.

Production and logistics postings come in separately. Goods receipts, issues to production and confirmations of labor and machine time all generate cost entries automatically, valued at whatever rates the system holds.

Allocation and the period-end run

Indirect costs get pushed out during the close. Systems typically offer two mechanisms. Assessments move costs in bulk under a single secondary cost element and lose the original detail. Distributions keep the original cost elements intact, which helps analysis but produces more lines.

Each cycle needs a basis: headcount, floor space, machine hours, revenue or a fixed percentage table. Statistical key figures hold these drivers, and someone has to update them before the cycle runs. Stale drivers are one of the most common reasons allocations look wrong.

Cycles run in a sequence. IT might charge facilities, facilities charges production, and production costs settle into products. Changing one cycle can ripple into all the ones after it, so order matters as much as the rules themselves. Once the cycles finish, the period is closed in controlling and opened for the next.

Product costing and inventory valuation

Manufacturers build a standard cost for each product by rolling up the bill of materials and the routing. Material prices, activity rates and overhead surcharges combine into a cost estimate. Released, it becomes the standard price used to value inventory.

Inventory then moves at standard. The differences between standard and actual pile up on production orders or in price difference accounts. Some businesses use actual costing, which revalues inventory at the end of each period. It gives truer numbers at the expense of a heavier close.

Cost of sales follows the same logic. When goods ship, the system relieves inventory at its carrying value and posts cost of goods sold, often split by cost component so material, labor and overhead stay visible.

Variances and profitability

At period end, production orders are settled. The system calculates variances and assigns them to categories such as price, quantity, efficiency and lot size, depending on configuration. Variances are then sent to the P&L or, in some setups, partly back to inventory.

Profitability analysis pulls revenue, cost of sales, variances and allocated overhead together by customer, product, channel or region. There are two common designs. One is driven by the general ledger and reconciles easily. The other uses separate value fields and is flexible, but drifts from the ledger unless someone reconciles it every period.

Where controls sit

Controls usually include locks on closed periods, approval for master data changes, restricted access to allocation cycles and a reconciliation between controlling and the ledger. Public sector bodies add funds control, where budget availability checks block postings that would exceed an authorized amount. Auditors tend to ask for documentation of allocation bases and standard cost releases, so keeping that evidence matters.

Questions to ask the people who run it

• Which allocation cycles get reversed and rerun, and what triggers that?
• Where do the driver figures come from, and who updates them?
• Are any costs moved by manual journal because the system rules no longer fit?
• How often are cost centers recoded after posting, and who catches the errors?
• When was the standard cost last revised, and who decides when to revise it?
• Which variance categories does anyone actually read?
• Does the profitability report agree with the ledger without adjustment?
• What spreadsheets sit between the system output and the pack management sees?
• Which old cost objects are still open, and why?
• What would break if a cycle ran out of sequence?

The answers usually reveal workarounds that never made it into the design documents. Those workarounds are where most of the real effort of the process lives.

APQC's Process Classification Framework® (PCF) is an open standard developed by APQC, a nonprofit that promotes benchmarking and best practices worldwide. To download the full PCF or to view definitions and measures, please visit www.apqc.org/pcf.