You’ve already made the decision before the session starts
A driver taps a card at an unattended charger. Before a single electron moves, you have already made the most consequential decision in your payments stack — and most operators make it without noticing.
The conversation leading up to that tap is almost always about machinery: which terminal, which CSMS, which acquirer, how the pre-authorization hold works. Those are real questions. None of them is the one that matters.
The one that matters is quieter, and rarely asked out loud: when the driver taps, whose money is it?
That single answer — who is the merchant of record — propagates outward into your cash flow, your unit economics, who owns the customer relationship, and how hard it is to leave your vendor later. It is the most load-bearing choice in the entire stack, and it tends to get made by accident: a side effect of picking a product rather than a decision made on purpose.
Two architectures, one fork in the road
There are really only two shapes.
In the merchant-of-record (MoR) model, your payments vendor becomes the legal merchant for the transaction. The driver’s card is charged to their merchant account. The funds land in their bank. They then pay you out — net of their cut, on their settlement timeline. On paper this is the low-effort path: they own compliance, they own the acquirer relationship, they own the card data, and you get a clean payout with less to build.
In the non-MoR model, you stay the merchant. The driver’s card is charged to your merchant account, and funds settle directly to your acquirer and your bank. A vendor still does the hard part — orchestrating the pre-authorization, the capture, the refund of the unused hold, the country-specific fiscal round-trip — but it never sits in the flow of money. It moves messages, not funds.
Same driver, same tap, same charger. Completely different business underneath. These are not “the same thing with different paperwork.” They diverge on four axes, and the axes compound.
Axis one: cash flow
Under MoR, the money is the vendor’s until they decide to release it. Your cash is gated by someone else’s settlement cadence, their reserve policies, their reconciliation. For an operator running thin margins on slow AC sessions, a payout that arrives on a vendor’s schedule — with a rolling reserve held against chargebacks you never saw — is a working-capital problem you imported without negotiating it.
Under non-MoR, the money was always yours. It settles to your bank on your acquirer’s terms — terms you can negotiate, because the relationship is yours. There is no intermediary deciding when you get paid for electricity you already delivered.
Axis two: the economics of a physical tap
This is the part that quietly eats small sessions.
A charge where the driver physically taps is card-present — lower risk, lower processing cost. That is the natural state of an AFIR-aligned terminal at an unattended charger: the card is right there, EMV contactless, the driver standing at the unit.
Route that same transaction through a third-party merchant of record and it frequently re-classifies as card-not-present — the markups designed for remote billing, e-commerce, and the fraud risk that comes with not seeing the card. Those remote-billing markups are tolerable on a high-value DC fast-charge. On a small, frequent AC session they are a meaningful slice of a thin margin, taken on every transaction, forever.
Non-MoR keeps the transaction what it physically is: settling through your own acquirer, priced as the in-person payment it actually is. You don’t pay the e-commerce rate on a driver standing two feet from the reader.
Axis three: who owns the customer
Under MoR, the driver is — in a real legal sense — the vendor’s customer. The vendor’s name can appear on the statement. The vendor holds the card data and the payment relationship.
Follow that through one unhappy session. A charge drops mid-flow, the hold doesn’t release cleanly, and the driver disputes it with their bank. Under MoR, that chargeback lands in the vendor’s support queue, against the vendor’s merchant account, resolved on the vendor’s timeline — but the driver doesn’t know any of that. They stood at your charger, under your signage, and yours is the brand they’re angry at. You absorb the reputational hit for a dispute you can’t see, can’t touch, and can’t expedite. The refund that would have rebuilt trust gets issued by someone whose name the driver never read.
The same asymmetry runs through every moment that actually builds a relationship:
- a refund handled gracefully
- a receipt the tax office accepts
- a question answered fast
If those flow through the vendor, you don’t own the trust — you only own the blame when it goes wrong.
It also means your customer data and payment history live in someone else’s system. The day you want to do something with that relationship — loyalty, a direct contract, anything — you’re asking permission. And that history is personal data, so GDPR obligations follow whoever holds it — one more reason to know exactly whose system it lives in.
Under non-MoR, the driver is your customer paying you. The statement is yours, the fiscal receipt is issued under your identity, the dispute is yours to resolve, the relationship is yours to build on. The vendor’s job is to make that flow work flawlessly and then disappear.
Axis four: lock-in
This is the axis people feel last and regret most.
When your vendor is the merchant of record, they hold the merchant account, the acquirer relationship, the card data, and the customer. Switching vendors is not a configuration change — it’s re-papering your entire payment relationship, re-onboarding the customer, and migrating data you don’t control. The exit cost is the whole point of the model, even when no one says so.
Non-MoR inverts that. Because you keep your own acquirer, your own bank, and your own CSMS — your chargers speaking OCPP underneath, payments bridged over OCPI above — the orchestration layer is the thinnest thing in your stack, not the thickest. The vendor bridges the money flow — pre-authorization strategy, partial captures, refunding the unused hold, the per-country fiscal signature round-trip, reconciling when the charger’s record and the acquirer disagree — but you can swap the terminal, swap the acquirer, change the country you operate in, and keep everything else.
The layer that orchestrates is replaceable by design. That is the opposite of lock-in: it is leverage.
Why neutrality is an architecture, not a slogan
It’s tempting to read “non-MoR” as a marketing posture. It isn’t. It’s a structural commitment, and it shows up in where the money physically goes.
A neutral payments layer means the operator stays the merchant and funds never route through the vendor’s accounts. That is verifiable. Ask two questions:
- Whose merchant account does the driver’s card hit?
- Whose bank does the settlement land in?
If the answer is “yours” on both, you have non-MoR. If it’s “the vendor’s,” you have MoR — whatever the contract calls it.
This is the line Bolt holds deliberately. Bolt orchestrates the entire money flow for a charge — across any terminal, any acquirer, any CSMS — while the operator keeps card-present economics, keeps their bank, keeps their customer, and keeps the freedom to swap any piece. The engine runs the money; it never becomes the money.
Make the decision on purpose
There are operators for whom MoR is genuinely the right call — typically when payments sit far from the core business and offloading the entire merchant relationship is worth the margin, the cash-flow timing, and the customer ownership you give up. That can be a rational trade. The failure mode isn’t choosing MoR.
The failure mode is choosing it without knowing you did — picking a turnkey product because the demo was smooth, then discovering a year in that your cash is on someone else’s clock, your small sessions are paying e-commerce markups, your drivers belong to your vendor, and leaving means rebuilding the whole thing.
So make it the first decision, not the last. Before you compare terminals or pre-authorization strategies, answer the one question that sets all four axes at once: when the driver taps, whose money is it? Everything else is downstream of that answer — and you only get to choose it cheaply once.