Reading a merchant before the first transaction.

Fictional scenarios and synthetic data. Reviewed 6 November 2026.

A new application lands at Northbank Pay: an iGaming operator, licensed in two jurisdictions, forecasting 60,000 card deposits a month, leaving its current PSP "for commercial reasons". The sales lead wants a yes by Friday. Ines wants the last six months of processing statements, the bonus terms, the affiliate programme and the withdrawal policy. The reader is asked to write the underwriting note.

What is actually being decided.

The PSP is not deciding whether the merchant is good. It is deciding what this merchant will do to a portfolio ratio the PSP reports to its acquirer, Stilling Bank, and which terms make that forecast safe to be wrong about. The PSP carries the acquirer relationship and, through it, the whole portfolio's access to card rails. The acquirer carries the scheme programmes. The merchant carries cash flow under whatever reserve and settlement terms it is given. The merchant's players carry the friction of whatever 3DS and payout rules the PSP insists on, and they never see the underwriting note.

FATF's recommendations describe customer due diligence as risk-based and ongoing, and a PSP's merchant is its customer. The same framing applies: the onboarding file is the start of the record, not the end.

What a PSP reads, and what each thing proves.

Before the first transaction the file contains documents of several kinds, and it pays to write beside each what it proves and what it does not.

Licences and jurisdictions prove the merchant may operate where it says. They do not prove where its traffic comes from; the issuer-country mix in the first month will. Corporate structure and beneficial owners prove who is accountable. Prior processing statements, with monthly volume, fraud reports, chargebacks and refunds, are the single most useful document and the one most often missing or partial. "Leaving for commercial reasons" is an observation about what the merchant said; "terminated for ratio" is an inference, and the statements either support it or do not. The bonus terms, affiliate programme and withdrawal policy are the three documents that describe how money will enter and leave, and they predict more of the first quarter than the financials do. If the welcome offer looks like the one in bonus abuse and multi-accounting and commissions are paid on first deposit as in affiliate traffic quality, the PSP already knows what the first export will look like.

The portfolio arithmetic.

Programme classification depends on the network, region, reporting period, applicable entity and the current rules. Reconcile the acquirer’s event counts and exclusions before comparing a calculated rate with a confirmed threshold. Do not treat a generic floor, ratio or fee quoted in a summary as a universal programme decision.

A synthetic portfolio starts with 400,000 eligible transactions and 1,200 eligible events, a rate of 0.30%. Adding a merchant with 60,000 transactions and 660 events produces 1,860 / 460,000, about 0.40%. If that merchant instead contributes 1,080 events, the combined rate is approximately 0.50%. This measures concentration and sensitivity; it does not establish either party’s current programme classification. Confirm that separately under the applicable rules.

Programme classification depends on the network, region, reporting period, applicable entity and the current rules. Reconcile the acquirer’s event counts and exclusions before comparing a calculated rate with a confirmed threshold. Do not treat a generic floor, ratio or fee quoted in a summary as a universal programme decision.

Terms are controls or prices. Name which.

Every onboarding term either changes what will happen or changes what the PSP is paid for it happening. A higher processing rate is a price; it does not lower a ratio. A rolling reserve is a price with a cushion; it funds losses after the fact. The terms that change what happens are the operational ones: mandatory 3DS on deposits above a threshold, a closed-loop payout rule with named exceptions, a monthly volume cap that steps up on review, affiliate-programme disclosure with a commission horizon the PSP can see, and a descriptor the player will recognise on a statement. Each costs the merchant something, usually cash flow or conversion, and the note should say what. A term the merchant cannot afford is a term that will be renegotiated in week three, which is a worse position than not having set it.

The first ninety days in the export.

The onboarding note is a forecast; the export is the test. Our reading is that a PSP should decide before go-live which columns it will watch, so that the 30-day review is a comparison and not a search. the useful ones are: deposit approval rate and the mix of decline codes; issuer-country share against the licensed markets; share of deposits that went through 3DS; distinct card references per account in the first week; withdrawals requested to a destination that does not match a deposit instrument; refund ratio; early fraud notifications; disputes by reason-code family; and the share of first-time depositors from the top three affiliate references. A merchant whose volume climbs steeply while the withdrawal-to-new-destination column climbs with it is showing the shape the glossary calls bust-out; the shape is an observation, and the review is where the inference gets made.

A failure mode to watch.

Underwriting once. The file is read, the terms are set, and the merchant is next looked at when the acquirer writes, by which point the portfolio ratio has done what the arithmetic above said it would. The second commonest is pricing risk instead of controlling it, so that the note is a negotiation about basis points while the deposit flow is left to the merchant's own judgement. The conversation with the acquirer that follows is covered in working with your acquirer, and it goes better with a 30-day review in the file.

A way to work it.

1. List each document in the file with one line for what it proves and one for what it does not.
2. Do the portfolio arithmetic twice: at the merchant's forecast and stated ratio, and at twice that ratio. Note where the acquirer thresholds sit relative to both.
3. Read the bonus terms, affiliate commission structure and withdrawal policy as predictions of the first export. Write the prediction down.
4. For each proposed term, mark it control or price, and name what it costs the merchant. Drop terms that are prices dressed as controls.
5. Choose the columns for the 30-day review before go-live, with the thresholds at which terms tighten or loosen. Name an owner.
6. At 30 days, compare the export with the prediction, change the terms if the evidence says so, and set the next review. Keep the note; it is the Volume 01 lesson evidence before action applied to a whole merchant.

Take it to the desk.

If this merchant ran at twice the ratio it showed you, which of your onboarding terms would actually change what happens, and which would only change what you charge? Try it: VAMP portfolio call, and the Volume 01 case Dispute spike for the view from inside one merchant. Look up: VAMP ratio and chargeback ratio.

Edited 6 November 2026. Primary references are listed above; fictional scenarios are not claims about Antifrog customers or operating experience.

Fictional learning material. Any numerical examples are synthetic. Use only authorised, appropriately minimised records for workplace exercises; do not upload customer information to this learning site.