Suppliers Know Your Tolerance Thresholds. Your AP System Does Not Track the Total.
How tolerance policies create cumulative AP leakage in maritime, and why the problem is invisible when checking happens per invoice rather than per supplier.
By Errikos Andreakos
Chief Commercial Officer for MarTrust
This article is part of our series on how to close the control gap in maritime's procure to pay workflow (article five of seven)
How do tolerance thresholds cause AP leakage in maritime?
Tolerance policies allow invoices to clear when variances fall within a defined percentage, typically 1 to 5 percent. The risk in maritime is cumulative: the same supplier across multiple vessels or invoices can price consistently within tolerance on each individual transaction while the total overpayment is significant. Per-invoice checking does not catch this pattern because no single transaction triggers a flag.
Why AP teams use tolerance thresholds and why that matters
Tolerance policies are rational. Chasing a $30 discrepancy on a $1,500 invoice across multiple time zones, with a supplier who holds a signed delivery receipt, costs more than $30. Tolerance bands exist because investigating small discrepancies is operationally irrational.
The problem is what the policy makes invisible.
Standard AP tools check at the transaction level. An invoice that is 1.8 percent above the agreed PO price, within a 2 percent tolerance band, clears. The next invoice from the same supplier, also 1.8 percent over, clears. Six invoices later, across multiple vessels, the same supplier has collected above the agreed rate on every transaction. No single invoice triggered a flag. The cumulative variance is real and it appears nowhere in the system.
This is not necessarily fraud. It may be pricing drift the supplier is managing intentionally, or a legitimate cost increase that was never formally updated in the purchase order. The AP workflow has no mechanism to surface either possibility because the check evaluates each invoice in isolation rather than against what that supplier has charged across the full relationship.
Learn how your team can scale invoice management with the fleet without adding to headcount
Get the guide - closing the invoice control gap in maritime
What happens when there's no PO to check against?
Port agent spend makes the problem harder still. Port agent invoices often arrive without a purchase order. There is no agreed price baseline. Tolerance-based checking requires a reference point. Without a PO, there is no reference. Small variances pass not because they are within tolerance of an agreed rate, but because there is nothing to compare them to.
Both patterns produce the same outcome: cumulative leakage invisible in transaction reviews and only visible in aggregate analysis across suppliers, vessels, and time. By the time it surfaces, it has been accumulating for months.
The controls that address this operate at supplier level rather than transaction level: tracking cumulative variance against agreed rates across the full invoice history, flagging where a pattern of marginal positioning exists even when no individual invoice crosses the threshold. That analysis requires connecting invoice data across vessels and legal entities. Per-transaction checking in a single ERP does not produce it.
Rational at the invoice level. Expensive at the fleet level.
The Maritime Control Gap covers this alongside the other structural gaps in maritime AP control.
Tolerance bands exist because investigating small discrepancies is operationally irrational. The problem is what the policy makes invisible.
Find out why leading ship managers are using Marcura to streamline their entire procure to pay process
Struggling with manual invoice management? We can help
Read more
recent news
New guide
Take control of Procure-to-Pay
Why invoice control breaks down between PO and payment — and how leading ship managers close the gap. Free guide, plus two diagnostic tools.
