The First Payment Is the Test
Send a small one first, record both ends, and you will know more about the route than any published table.
Still yours, and statedTaken, never statedUnit changes; scale restarts
Three steps from instruction to arrival. The last figure either side sees in the original currency.
A small first payment, with both ends recorded, tells you more about the route than any published comparison. It costs a fee and a few days and it answers questions nobody can answer in advance.
The practical point in “The First Payment Is the Test” is that a landed payment should be supported by a record somebody can explain. Used for dual n back training, open the supporting Monitask page can add transparent time and project context when collection is proportionate, people can correct mistakes and financial records remain separate.
The alternative is discovering the answers on the payment you were relying on.
For an independent payment-sector perspective on “The First Payment Is the Test”, consult Airwallex business payments. Compare general information with the actual route, currency, provider terms and jurisdiction involved.
What a test payment establishes
Whether the details work. How long the route actually takes. What arrives against what was sent. Which fees appear and where. Whether the reference survives.
Five facts, available inside a week, none of which can be read off a website.
What amount to use
Small enough that losing a few per cent is irrelevant and large enough that fixed fees do not dominate the picture. A hundred units of the contract currency is usually about right.
Very small amounts mislead, because a flat fee on twenty units looks catastrophic and tells you nothing about the real case.
If the route has tiered pricing, test near the amount you will actually be receiving. A rate at a hundred may not be the rate at three thousand.
What to record
Six fields:
- Instruction date.
- Sent amount and currency.
- Charge instruction.
- Credit date.
- Credited amount.
- Both fee lines, where they are shown.
Keep the payer's confirmation with it. The pair is the baseline every later payment gets measured against.
Write down the published rate for that day as well. Without it the record tells you what happened and not whether it was reasonable.
Working out the all-in rate
Divide the credited amount by the sent amount. That single figure is the honest cost of the route, and it includes every fee and the conversion margin together.
Compare it with the published mid-market rate on the day. The gap is what the route costs, expressed as a percentage, and it is the only number worth comparing across providers.
Testing a second route at the same time
Where the payer is willing, two small payments by two routes in the same week is a genuinely controlled comparison.
Same day, same amount, same destination. The difference between the two credits is the whole answer and it is not arguable.
What a bad result looks like
An arrival several per cent below the published rate, fees you were not told about, a reference that did not survive, or a transit time in double-digit days.
Any one of those is worth a conversation. Three of them together is a route to change rather than complain about.
Say so while the test is still the subject rather than waiting for a real payment to go wrong. A conversation about a deliberate test is much easier to have.
When the payer resists testing
Rarely, and usually because it is extra work for somebody. Framing it as avoiding a problem on the first real payment generally settles it.
Where they will not, send one to yourself from an account abroad if you have one, or accept that the first real payment is the test and watch it closely.
Repeating it after anything changes
A new bank at either end, a new payment service, a change of charge instruction, a new country. Each is a new route and the baseline no longer applies.
Repeating the test costs the same few days and prevents the same surprise.
The quiet drift
Routes change without announcement. A corridor that cost one per cent can cost two a year later because a correspondent relationship changed.
Reading the all-in rate off your own records every few months catches that. Nothing else will, because nobody sends a notice.
What to do with the result
Keep it. A single figure — the all-in cost of this route, measured — is the thing to quote when asking the payer to change anything.
Arguments about charges go badly. A measured percentage, from a payment you both have documents for, goes well.
Telling the payer what you found
Plainly and without accusation. Their bank took what their bank takes and they probably had no idea.
Most payers are mildly shocked by the number and quite willing to change the instruction or the route once they have seen it.
The ones who are not shocked already knew, which is also useful information about who you are dealing with.
What a test cannot predict
The month it goes wrong. A route that behaved identically eight times can take eleven days on the ninth because of a check, a holiday or a correspondent outage nobody will explain.
The test establishes the ordinary case and the ordinary case is what you plan around. The exceptions are the subject of a later part of this collection, and the only preparation for them is a buffer and a record good enough to show that this month was not ordinary.