Why Companies Lose Money on Invoices (and How to Stop It)
In 2019, Lithuanian Evaldas Rimasauskas was sentenced to five years in prison for a scam in which he tricked Google and Facebook out of more than $120 million. His method? Fake invoices for legitimately delivered goods — except the money went to his own accounts instead of the real supplier's.
This isn't a problem limited to tech giants.
A company orders 2,000 components. The delivery arrives with 1,200. The invoice? For the full 2,000.
This scenario can be prevented, yet most companies still handle it manually. Someone has to dig up the purchase order, find the delivery note, and compare it to the invoice. Three documents, five minutes of work, and a high chance of error.
The scale of the problem is staggering. The Association of Certified Fraud Examiners (ACFE) estimates that organizations lose 5% of their annual revenue to fraud, with billing schemes being the most common type of asset misappropriation. For a company doing $50 million in revenue, that's $2.5 million walking out the door.
What 3-Way Matching Means in Practice
3-way matching is a process that automatically compares three documents:
Purchase order – what the company ordered and at what price
Delivery note – what physically arrived in the warehouse
Invoice – what the supplier is billing
When all three match in quantity, price, and items, the invoice can be approved. If not, the system flags it, and someone has to find out why.
It's not revolutionary technology — it's simply the automation of what every person approving invoices should be doing. But in most cases, they don't do it consistently.
The strength of 3-way matching lies in its simplicity. Each document acts as a checkpoint. The purchase order confirms authorization ("did we agree to buy this?"). The delivery note confirms receipt ("did we actually receive it?"). The invoice confirms the financial claim ("is the amount correct?"). When any link in this chain breaks, the discrepancy surfaces immediately rather than hiding in spreadsheets.
Some organizations extend this to 4-way matching, adding the inspection or quality check as a fourth verification point. This is particularly common in manufacturing and pharmaceutical procurement where quality specifications are contractually binding.
Where the Money Gets Lost
According to studies, roughly 3-5% of B2B invoices deviate in some way from the purchase order. For companies processing thousands of invoices monthly, these deviations add up fast. The most common cases are:
Price differences – The company agreed on a discount, but the invoice uses the old price. Or the supplier simply charges more than agreed. This is especially common with long-term contracts where prices were renegotiated but the supplier's billing system wasn't updated. Research from Ardent Partners shows that invoice price exceptions account for 39% of all AP processing delays.
Quantity differences – Fewer items arrived than are listed on the invoice. Someone in the warehouse miscounted, or the supplier hoped no one would notice. Partial shipments are a major source of quantity mismatches — the supplier ships what's available but invoices the full order.
Duplicate invoices – The same invoice arrives by email and by post. Or the supplier sends it twice "just to be sure." Result: double payment. Studies suggest that 0.1-0.5% of all payments are duplicates, and recovering them takes an average of 60-90 days once discovered.
Invoices for cancelled orders – The order was cancelled, but the invoice still came through. In organizations with disconnected systems, cancellations may not propagate to accounts payable until after payment is made.
Unauthorized charges – Line items that were never part of the original agreement: rush fees, handling charges, or "administrative surcharges" that appear without prior discussion.
For 500 invoices per month with an average value of EUR 5,000, that means a potential loss of EUR 75,000 to EUR 125,000 per year. Not every discrepancy is fraud — but every discrepancy costs money. And the cumulative effect over years can be enormous.
The Hidden Cost of Late Payments
Beyond direct losses from discrepancies, manual invoice processing creates a secondary cost: late payments. When invoices sit in approval queues for weeks, companies miss early-payment discount windows. A typical 2/10 net 30 discount equates to a 36% annualized return. Missing that on $1 million in monthly spend costs $20,000 per month — money left on the table simply because the process was too slow.
Late payments also damage supplier relationships. Suppliers who aren't paid on time may increase prices, reduce priority during shortages, or add payment-risk premiums to future quotes. The cost of a strained supplier relationship rarely shows up on a balance sheet, but it's real.
How It Works in a System
Modern procurement systems have 3-way matching built in. The process works like this:
The purchasing department creates a purchase order — the system records the items, quantities, and prices.
The warehouse receives the goods and logs the delivery note — what actually arrived, with timestamps and photos if needed.
The accountant uploads the invoice into the system (as a scan or PDF). OCR technology extracts the data automatically, eliminating manual data entry.
The system automatically compares all three documents within seconds.
If everything matches within defined tolerances (e.g., +/- 2% on price, +/- 5 units on quantity), the invoice goes for approval. If not, the system highlights the specific difference:
"Invoiced 2,000 pcs, delivered 1,850 pcs"
"Invoiced price EUR 44.50, ordered price EUR 42.00"
"Item XY123 not found in purchase order"
Someone in the company must then explain the difference and confirm whether the invoice is valid or not. The system creates an audit trail of every decision — who approved what, when, and why.
Advanced systems also learn from patterns. If a particular supplier consistently invoices 3% above agreed prices, the system can flag the vendor for contract review before the next renewal.
Why It's Rarely Done in Practice
The main reason is simple: manually, it's a pain.
The accounting team gets the invoice. They have to find the purchase order (maybe in another system, maybe in an email). They have to find the delivery note (often on paper, somewhere in the warehouse). They have to compare everything line by line.
With 50 invoices per month, that's manageable. With 200+, it becomes days of work. So it's done only randomly — or not at all.
The result: invoices get approved simply because "it looks fine." That's not control — that's gambling.
There are additional barriers beyond time:
System fragmentation – Purchase orders live in one system, delivery notes in another, and invoices in a third. There's no single source of truth.
Staff turnover – Institutional knowledge about pricing agreements and supplier terms walks out the door when experienced AP staff leave.
Volume spikes – Month-end and quarter-end processing surges overwhelm teams, leading to shortcuts.
Lack of standardization – Different suppliers use different invoice formats, making automated comparison nearly impossible without OCR and normalization.
What It Actually Solves
The main benefit isn't just catching errors or fraud. It's using time effectively.
Instead of copying numbers from PDFs into Excel and comparing them manually, the accounting department can focus on real issues: why a supplier keeps under-delivering, why prices fluctuate, how to optimize orders.
And for companies where approval takes weeks, automatic matching means invoices that are correct pass through the system in hours, not days.
Specifically, automated 3-way matching delivers:
Fraud prevention – Catches fictitious invoices, billing for undelivered goods, and price manipulation before payment is made.
Cash flow visibility – When you know exactly which invoices are clean and which need review, you can forecast cash outflows with confidence.
Audit readiness – Every match, mismatch, and resolution is logged. When auditors come, the evidence is already organized.
Supplier accountability – Data on invoice accuracy by supplier becomes a powerful negotiation tool. Suppliers who consistently send incorrect invoices can be held to performance standards.
Faster close cycles – Finance teams spend less time reconciling at month-end because discrepancies are caught in real time, not discovered during close.
How It Works in ProcureSwift
We built our procurement application on exactly this principle — automatic 3-way matching running in the background without burdening the accounting department.
The system works simply: if the documents match, the invoice goes for approval. If not, the system clearly shows where the problem is — whether it's a difference in quantity, price, or an item that doesn't belong there. Then all that's left is to find out why and decide whether to approve or reject the invoice.
The key is that all three documents are digital and stored in one place. No Excel sheets, no email threads, no papers in drawers.
ProcureSwift's matching engine handles multi-line invoices, partial deliveries, and multi-currency transactions automatically. Tolerance thresholds are configurable per supplier or category, so you can set tighter controls on high-value items while allowing minor rounding differences on small purchases.
Conclusion
3-way matching is a basic practice every company should follow. It's not about distrusting suppliers — it's about controlling your own money.
Companies without automatic matching lose both time and money: time on manual checks that are often skipped anyway, and money on discrepancies that go unnoticed.
The ROI is straightforward: if you process 500+ invoices monthly, automated 3-way matching typically pays for itself within 3-6 months through caught discrepancies, captured early-pay discounts, and reduced AP labor costs.
ProcureSwift automates this entire process to run seamlessly in the background. No extra training, no complex setup — just precise verification of every invoice before it's paid.