CFOtech Ireland - Technology news for CFOs & financial decision-makers
Ireland
The Cost Curve Finance Forgot to Model

The Cost Curve Finance Forgot to Model

Tue, 17th Nov 2026 (Today)
Vijay Vidiyala
VIJAY VIDIYALA Chief Product & Technology Officer PairSoft

Every growth plan a CFO approves rests on assumptions about customers, margins and capital. Few state how many invoices the larger business will generate or how many people it will take to process them. In a manual accounts payable (AP) function, that omission is costly. Invoice volume rises with revenue, and headcount rises with volume, so a department that earns no revenue becomes one of the fastest-growing lines in the cost base.

Scaling or merely growing?

Finance leaders watch interest rates, inflation and input costs, but those are largely beyond their control. Cost per invoice, cost per payment and AP cost as a share of revenue are within their control, yet they get far less scrutiny. APQC's cross-industry benchmarking puts the median cost to process an AP invoice at $6.00, based on data from 5,846 organizations.

The model is simple. AP cost equals invoice volume times cost per invoice. Volume tracks revenue. In a manual function, cost per invoice barely moves, so the line is straight: double revenue, double the cost. A company processing 50,000 invoices a year at the $6.00 median spends $300,000. At 100,000 invoices it spends $600,000, unless cost per invoice falls. APQC's top-quartile organizations run at roughly $2 per invoice, so the same 100,000 invoices cost about $200,000. The gap between those two lines is what the growth plan left out.

A benchmark tells you where you stand today. To see whether your own AP function is scaling, chart three years of AP headcount and invoice backlog against revenue. If the lines rise together, the function is growing alongside the business rather than scaling with it.

Where the cost accumulates

That cost is easy to miss because it is spread across the organization, and it begins with delay. Invoices sit in the inbox of a traveling manager or the wrong cost-center owner, and even the right approver often cannot see whether the invoice matches the purchase order, how much budget remains or whether the price agrees with the contract. The delay has a price that never appears on a statement. A missed early-payment discount leaves no ledger entry, yet on standard 2/10 net 30 terms, forgoing the 2% discount to pay 20 days later is equivalent to passing up an annualized return of about 37%. Suppliers notice the pattern as well. They remember who pays reliably, and reliable payers tend to earn larger credit lines, priority when supply is tight and better terms at renewal.

The rest of the cost falls on people and controls. Each mismatch between purchase order, goods receipt and invoice sends someone searching for the cause, and across thousands of exceptions, analysts end up reconciling data instead of analyzing it. The same pressure wears down oversight. Manual checks rely on people noticing duplicate invoices, altered bank details and out-of-policy spend, and attention is the first thing to give way when volume spikes.

No single report totals these costs, which is why they rarely reach the CFO as one number.

From cost center to working-capital lever

Judging AP only by what it costs to process an invoice undersells the function. The larger opportunity is on the balance sheet. When invoices are approved within days of arrival, the CFO has a real choice on each one: take the discount where its implied return exceeds the cost of capital, pay strategic suppliers early when doing so secures better terms or supply priority, or pay everyone else on the due date, not before.

Payment method belongs in the same decision. Moving from paper checks to ACH lowers processing costs and closes off a channel that fraudsters favor. Same-day ACH covers urgent payments without wire fees, and virtual cards can earn rebates and extend the cash cycle.

When every approved and pending invoice sits in one system, treasury sees upcoming obligations as they build, and cash forecasts improve. Days payable outstanding becomes a figure the CFO sets deliberately, rather than one reported after the fact.

Sequence the fix

Order matters here. Companies that skip steps tend to automate their problems instead of solving them. Start by measuring: cost per invoice, cost per payment, discount capture rate and touchless rate. Touchless rate, the share of invoices that move from receipt to approved for payment without anyone handling them, is the clearest indicator of whether AP can absorb growth. Set a target for each metric, report it monthly and review the invoices that fall out of the flow.

Next, repair the process. Automating a broken workflow only produces the same errors faster. Group a quarter's worth of manual interventions by root cause, such as missing purchase orders, price mismatches or duplicate supplier records. Enforce "no PO, no pay" where it applies, and put your payment policy covering discounts, early payment and payment methods in writing.

Only then should you choose technology, and specifically tools built for AP. General-purpose automation rarely handles three-way matching, GL coding, vendor master data or live ERP connections well. The right software applies company policy, learns from the organization's own approval and coding history, verifies supplier bank details, stops duplicates before they are paid and keeps a complete audit trail. Count integration in the model too. Tools that need middleware or custom connectors carry a maintenance cost that returns with every ERP upgrade; tools that integrate natively with the ERP do not.

Software should handle routine matching, coding and routing. People should keep the approvals that commit spend, the exceptions that require judgment, supplier relationships and the timing of payments. As automation resolves more exceptions reliably, finance teams can extend its remit, but each step should be justified by the audit trail rather than assumed.

Put the invoice line in the model

The question for the next growth plan is not whether AP will cost more as the business grows. It is whether the invoice line is in the model at all, and whether the plan assumes cost per invoice stays flat or falls. Three years from now, the chart of AP headcount against revenue will show which assumption was right.