Most failures are not fraud
It is tempting to read every collapsed transaction as an attempted deception. Most are not. They are capability gaps that neither party surfaced early enough, and the pattern repeats often enough to be predictable.
One — the buyer cannot open the instrument
By far the most common. The buyer is genuine and wants the cargo, and their bank will not issue against their balance sheet. Visible early if you ask which bank will issue and whether the line is in place. Almost invisible if you do not ask until contract stage.
Two — the seller does not control the product
The seller is an intermediary presenting someone else's cargo as their own. It may still be a real cargo, but the seller cannot commit a loading window and the schedule slips repeatedly. Surfaces when you ask for proof of product against a live instrument.
Three — the chain is too long
Four intermediaries between the real buyer and the real seller, each adding a margin and each needing the one before them to move first. Nobody is lying and nothing can complete. Visible in how long each answer takes to come back.
Four — specification mismatch discovered late
The parties agreed on "D6" or "EN590" without agreeing which limits. The dispute arrives at the loading port when the certificate does not match what the buyer assumed. Preventable entirely by specifying against a published standard at the ICPO stage.
Five — procedure disagreement
Both parties want the transaction and each has a procedure that requires the other to move first. Neither is unreasonable; the deal dies in the deadlock. Surfaces immediately if both procedures are exchanged and compared before anything else.