How to choose the right PSP or acquirer
Most merchants arrive at this decision with a shortlist of three providers and a spreadsheet comparing rates. By the time you compare rates, the important decision has already been made by default.
First, decide which layer you are buying
The term PSP encompasses several different commercial relationships, and merchants regularly sign the wrong one.
A gateway gives you connectivity and nothing else. You still need an acquirer behind it.
A PSP operating as an aggregator boards you under its own master agreement. Onboarding takes days, integration is simple, and the price is usually flat. The trade is control. You are a sub-merchant on someone else’s MID, you have no direct relationship with the schemes, and a category-level decision made in their risk committee can remove you with short notice. For a business under roughly a million in annual card volume, this is often the right answer. For a business planning to triple, it is a ceiling.
An acquirer gives you your own merchant ID and a direct contractual relationship. Onboarding is slower, and the underwriting is real. In return, you get pricing that reflects your performance, direct visibility into scheme monitoring, and a relationship that survives a bad quarter because it was underwritten with that in mind.
A Merchant of Record takes on the contract with your customer, shifting consumer liability, refund obligations, and in most cases VAT collection to them. That solves a genuine problem for digital goods sold across many jurisdictions. It also means the customer relationship and the data attached to it are not entirely yours.
Orchestration sits above all of this and routes between multiple acquirers. It earns its fee at a certain level of volume and complexity. Below that level, it adds a layer, a cost, and another party to a problem.
Pick the layer before you pick the provider. Everything below assumes you are choosing an acquiring relationship.
Write your transaction profile before you talk to anyone
No provider can quote accurately without this, and most merchants supply it in fragments over three calls. It is one document.
Send the same document to every provider. You can compare quotes that answer the same question. Quotes built on different assumptions can’t be compared, which is how merchants end up choosing the provider with the most optimistic sales engineer.
- Annual and monthly card volume, split by month so the seasonality is visible
- Average ticket, and card mix
- Your cardholders’ issuing country, rather than your own location
- Business model: one-off purchase, subscription, trial conversion, deposit, or forward delivery, with the gap between payment and fulfillment stated in days
- Settlement currencies and the currency you want to be paid in
- Refund rate and chargeback ratio for the last 12 months, broken down by reason code
- Target markets for the next two years
Coverage decides more than price
An acquirer can only acquire in jurisdictions where it is licensed and has a local entity or a local scheme arrangement. This matters commercially, not just legally.
If your cardholders are German and your acquirer processes the transaction outside the EEA, the transaction is treated as cross-border and carries scheme fees that domestic processing does not. The same sale, the same card, a different cost base. For a business selling across Europe, the acquirer’s local presence in your largest markets is worth more than a few basis points of margin.
Local payment methods follow the same logic. If a meaningful share of your revenue should be coming through iDEAL, Bancontact, Blik, or SEPA Direct Debit, then either the provider supports them natively or you are running a second integration.
Then check the acceptance technology, because that is where you win approval rates. Network tokens and account updater keep recurring transactions alive when cards are reissued. 3DS2 with proper exemption handling, including transaction risk analysis where the acquirer’s fraud rate allows it, keeps friction off transactions that do not need it. Providers vary widely here, and the difference shows up in revenue rather than in the rate card.
Compare the total cost of acceptance
Ask for IC++ pricing. Interchange and scheme fees are pass-through costs that neither of you controls, so the only number under negotiation is the acquirer’s margin. A blended rate hides that margin and absorbs any interchange savings into the provider’s pocket. In the EEA, interchange on consumer debit and credit is capped at 0.2% and 0.3% by regulation, which means a blended quote materially above those levels is mostly margin and fees.
Scheme fees change twice a year. A contract that lets the acquirer pass through scheme changes is normal. A contract that lets the acquirer change its own margin without your agreement is not.
Then build the full cost, because the headline rate is typically half of it: authorization fees and whether declines are charged, refund fees, chargeback fees and what happens if a representment succeeds, gateway and tokenization fees, 3DS fees per authentication, FX margin on conversion and settlement, settlement and payout fees, monthly minimums, setup costs, and the reserve and its working-capital cost.
Run the arithmetic on the approval rate alongside it. As an illustration, on 20 million euros of annual volume, five basis points of margin is 10,000 euros a year. One percentage point of approval rate on the same base is 200,000 euros of recovered revenue, worth whatever your contribution margin is on it. The numbers will differ in your case, but the ranking rarely does. Providers compete on rate because rate is easy to compare. Spend your negotiating effort on the things that move the larger number.
Underwrite the relationship, not just the deal
The question is whether this provider will still work for you in three years.
Ask who actually underwrites your account and who sponsors it, because an acquirer relying on a sponsor bank inherits that bank’s appetite and its changes of mind. Ask what happens at three times your current volume, whether that triggers a review, and what the volume cap is. Ask what a chargeback spike triggers and at what ratio the conversation starts. Ask how Visa and Mastercard monitoring program breaches are handled and who carries the fines.
A provider who answers these questions precisely is telling you they have thought about your file. A provider that answers reassuringly is telling you it has not.
Read the exit before you sign the entry
Term length, notice period, and automatic renewal. Volume commitments and the cost of a shortfall. Reserve definition, the release trigger, and the release timing. What happens to settlement upon termination: withholding funds for the duration of the chargeback window is common practice and should be a known amount, not a matter of discovery.
Then data portability, which is the real switching cost. If your card credentials sit in a provider-proprietary vault, leaving means either a migration project they have to cooperate with or asking every customer to enter their card again. Network tokens and a contractual right to a vault migration turn a twelve-month problem into a six-week one. Agree it while they are selling to you.
Build the second connection early
A single acquiring relationship is a single point of failure, and the failure modes are ordinary. A risk review, an outage, a category exit, a sponsor bank change. Merchants at scale run a second MID, sometimes dormant, sometimes carrying a small share of traffic to keep it alive and proven.
The integration costs something now. An unplanned migration during peak season costs considerably more, and the provider you approach in that state knows exactly what position you are in.
How PPS runs this
We build the transaction profile, take it to the acquirers whose appetite and coverage actually match the model, and compare quotes on the total cost of acceptance rather than the headline rate. We negotiate the terms that matter later: reserve mechanics, pricing change rights, and data portability. TrustBench structures the documentation so that the same complete file goes to every provider, and underwriting moves in days rather than weeks.
You pay us, not the providers. That means the recommendation fits your business.