Every payments vendor in EV charging says the same two things: “open ecosystem” and “no lock-in.” The words are free. They cost nothing to print on a slide, and they survive exactly until the contract gets specific.
So stop reading the claim. Define the test instead.
The only question that matters
Here is the litmus test for a neutral payments layer: can you rip it out?
Not “is the API documented.” Not “do they support OCPI.” Can you remove the vendor — this quarter, without re-platforming your charging network, without changing banks, without losing your customer relationships — and keep operating?
If the answer is yes, you have a layer. If the answer is “well, technically, but your settlement would break, your receipts would stop, and your terminals would brick,” you don’t have a layer. You have a dependency wearing a layer’s clothes.
The rest of this is that one question, pointed at the three places a payments vendor can quietly become load-bearing:
- Your money — who settles the funds.
- Your charging stack — whose protocol the integration speaks.
- Your hardware — how hard it is to swap a terminal, an acquirer, or a market.
Test 1 — Who is the merchant?
This is the test the marketing language is designed to make you skip.
In a neutral setup, you are the merchant of record. The driver’s card is charged on your behalf. The funds settle to your acquirer, into your bank account. The payments vendor orchestrates the flow — the pre-authorization, the capture, the refund of the unused hold — but the money never lands in their account, never routes through their balance sheet, never waits on their payout schedule. They are never the merchant of record. They never hold your cash.
The alternative looks friendlier on day one and is a trap by day ninety. The vendor steps in as the merchant. They collect the driver’s payment, take their cut, and pass you the remainder on their timeline. Now ask the rip-out question. You can’t leave, because they own the cash flow. Your money is on the wrong side of the relationship. Worse, they own the driver — the card data, the transaction record, the support relationship, the receipt. You’ve outsourced your customer to your supplier.
There’s a second, quieter cost to handing the merchant role away: economics. When you stay the merchant with your own acquirer, your sessions clear as card-present transactions — a card physically tapped at a terminal on the charger. Card-present is the cheaper, lower-risk lane. Route the money through an intermediary platform instead and your sessions can start being treated as card-not-present, which tends to carry the markups built for e-commerce risk. On a small roadside top-up — a few kWh, a low-value tap — those markups are not a rounding error. They’re the difference between a session that makes money and one that doesn’t.
A neutral layer keeps card-present economics because it keeps you as the merchant. That isn’t a feature you bolt on. It’s a consequence of who holds the money.
Test 2 — Do you own your charging stack?
Your chargers talk to your CSMS over OCPP. Your CSMS talks to the outside world — roaming partners, regulators, and now payment terminals — over OCPI. That’s the open spine of the industry, and a neutral payments layer should bolt onto that spine, not replace a vertebra.
This is where the wrong architecture hides. There’s a tempting shortcut: bolt a POS terminal directly onto one charger model and wire it straight into that charger’s firmware. It demos beautifully. It’s also the most locked-in thing you can build.
A charger is not a cash register, and it was never designed to be one. A charger maker who is excellent at building chargers has no reason to also own your payment plumbing — and the moment a terminal is wired directly into a charger over a bespoke charger-to-terminal-to-CSMS path, you’ve written a custom extension that sidesteps the open protocols. It’s a proprietary side-channel that only works for that charger model from that one vendor. Real operators run mixed fleets: different makes, different vintages, AC and DC side by side. A per-charger integration doesn’t scale across that fleet; it fragments it. Now you have one payment integration per hardware SKU — each one a maintenance liability, each one a reason you can never swap that hardware.
The neutral answer is to bridge payments over OCPI — the same protocol your CSMS already speaks. The payments layer doesn’t care what charger sent the session or which CSMS is running it. Any charger, any CSMS, any terminal. Swap any piece whenever you want, because the contract between them is an open standard, not a private cable.
Apply the rip-out test. With the OCPI-bridged approach, you can change your CSMS and your payments layer keeps working, because it was only ever talking OCPI. With the bolted-on POS, changing anything means rewriting the integration.
One of these is a layer. The other is a weld.
Test 3 — Can you swap the hardware and the bank?
Neutrality isn’t a static property. It’s the freedom to change your mind later.
A genuinely neutral layer lets you mix terminal vendors across your network and swap them when a better one ships — because it speaks EMV to any modern terminal, not a proprietary handshake to one. It lets you change acquirers when your rates or your coverage demand it, because you’re the merchant and the acquirer relationship is yours to renegotiate. And it lets you operate across countries and currencies without re-architecting, routing each session to the right acquirer for the right market.
The test is the same every time. Pick the piece — terminal, acquirer, country — and ask: to change it, do I call my payments vendor and ask permission, or do I just do it? If swapping a terminal vendor requires a new firmware project, you’re locked into the hardware. If changing acquirers requires the vendor’s cooperation because they hold the merchant account, you’re locked into the vendor. A neutral layer makes all three cheap by design, because none of them are wired into the vendor’s business model.
Where’s the catch?
Fair. It’s the right question to ask anyone who shows up promising freedom, and you should ask it of us too. Here’s the substance, not the reassurance.
First, the layer is built on an open standard you don’t get from us. OCPI is governed by the industry, through the EV Roaming Foundation — not by any single vendor. The connection between a terminal and a charging network rides a public spec. The thing that would normally be a vendor’s proprietary moat is, here, a document anyone can read and anyone can implement.
Second, Optechain helps write that standard — a contributor seat in the community that maintains OCPI. The work to standardize ad-hoc terminal payments — the DirectPayment module, the open answer to the AFIR contactless-reader mandate — happened in that open community: published, versioned, implementable by anyone. That’s the opposite posture from a walled garden. A walled garden wants the integration secret so you can’t leave. A standards contributor wants the integration public so the whole market interoperates — including the parts that aren’t us. You can’t simultaneously be the company writing the open spec and the company holding you hostage to a private one.
Third — and this is the part that should actually convince you — we never become the merchant. The money settles to your acquirer. We don’t touch it. Keeping the operator out of card-data scope is part of the same posture, not a badge we wave. That single architectural fact is what makes the rip-out test pass. A vendor who never holds your cash has no leverage to trap you, because the leverage in payments is the cash. Remove it, and “no lock-in” stops being a promise and becomes a structural fact about where the money lives.
The sharp line
Lock-in isn’t a clause in a contract. It’s a question about where your money settles, whose protocol your stack speaks, and how expensive it is to walk away.
A neutral payments layer answers all three the same way: your bank, an open standard, and a low cost to leave. Anyone can say “no lock-in.” Only an architecture that hands you the rip cord means it.