SEPA Request-to-Pay: How It Works for Merchants and Billers
The SEPA Request-to-Pay messaging flow, who is actually reachable today, the reconciliation gains for merchants and billers, and the real costs and limits.
Key takeaways
- SEPA Request-to-Pay is a messaging scheme, not a payment instrument, so the money still moves as a SEPA credit transfer, usually instant
- Participation is voluntary, so reachability depends on which banks and communities have actually joined — check the EPC register before you build a business case
- The reconciliation payoff comes from structured ISO 20022 data travelling with the request, not from the request itself
- Budget for homologation, API work or provider contracts, testing waves and the operational handling of refusals, expiries and unsupported options
SEPA Request-to-Pay is a message, not a payment
SEPA Request-to-Pay (SRTP) lets a payee — a merchant, a utility, a telco, a landlord, an invoice issuer — send a structured payment request directly into the payer's banking environment. The payer sees it in their banking app or online banking, and can accept or refuse it.
The scheme's own rulebook is blunt about what it is: a messaging functionality, a way to request a payment initiation, not a payment means or a payment instrument. The money that follows is a separate SEPA credit transfer, in practice usually a SEPA Instant Credit Transfer. This matters for anyone writing a business case, because an accepted request is not a settled payment, and the payee cannot force one.
Version 4.0 of the SRTP scheme rulebook has been effective since 5 October 2025. It simplified and clarified the scheme rather than expanding it: complementary reason codes, a defined set of options, mandatory registration in the EPC Directory Service (EDS), new sealing requirements in the API specifications, and a lighter homologation process. The rules are written and maintained by the European Payments Council (EPC), the same body behind the SEPA credit transfer and direct debit schemes covered in our guide to payment rails.
Two clarifications worth making early. First, SRTP is not a card replacement and does not compete on interchange economics — there is no interchange revenue in the flow, which changes who is motivated to push it (see our breakdown of card interchange fees). Second, "RTP" also happens to be the name of a US instant credit transfer network, which is a different thing entirely.
The four-corner messaging flow
SRTP runs on four roles:
- Payee — the party asking for money.
- Payee's SRTP service provider — the payee's bank or a non-bank provider.
- Payer's SRTP service provider — typically the payer's bank.
- Payer — the party who decides.
Notably, the service provider roles are not reserved for payment institutions. The EPC scheme description states that eligible entities from all SEPA countries may participate on a level-playing-field basis between payment service providers and non-PSPs. That is what allows an invoicing platform, an ERP vendor or a billing software company to sit on the payee side.
The message path is short:
- The payee creates the request, with amount, currency, expiry date/time, requested execution date, remittance information and identifiers, and sends it to its SRTP service provider.
- The payee's provider routes it to the payer's provider. Since 30 November 2023 the rulebook requires, at minimum, API-based message exchange between scheme participants.
- The payer's provider identifies the payer and presents the request in the customer channel — app, online banking, notification.
- The payer accepts, refuses, or lets it expire.
- On acceptance, the payment leg is initiated: a credit transfer, normally instant, with the payer authenticating as required.
- Status messages travel back to the payee, which can match and post the payment automatically.
The scheme allows three timing combinations, and this is where product design decisions get made:
| Variant | What the payee asks for | Typical use |
|---|---|---|
| Accept now / pay now | Decision and payment at once | Checkout, in-store QR, on-delivery collection |
| Accept now / pay later | Decision now, execution on a later date | Recurring collections with per-invoice approval |
| Accept later / pay later | The payer reviews and schedules | B2B and utility invoices with payment terms |
An expiry date/time is mandatory on every request. The payer's provider must always accept an expiry of up to three months from the request timestamp; anything longer requires a bilateral agreement between the two providers. Beyond the expiry, the request simply becomes void — no funds move and no liability transfers.
Who is actually reachable today
The scheme has been in force since June 2021, but participation is voluntary, and that is the single most important practical constraint. Instant euro credit transfers were made mandatory for euro-area providers under the EU's Instant Payments Regulation (Regulation (EU) 2024/886), in force since April 2024 with the main obligations applying to euro-area providers from January 2025. Request-to-Pay was not included in that mandate. A payer's bank is not obliged to offer it.
The EPC's June 2026 status update to the Euro Retail Payments Board reported three successfully homologated SRTP service providers published in the register, and four homologated Referenced Technical Solution Providers (RTSPs) — third parties that can carry the messaging for scheme participants. It also reported that further applicants plan to join by the end of 2026 and into 2027, and that the EDS was scheduled to open to SRTP participants by the end of September 2026, with homologation waves running through 2026 (EPC scheme calendar). The EPC publishes a regularly updated register of participants, which is the right place to check before committing to a build. The scheme manager's stated strategy for closing the reachability gap is a "federated approach" that starts from national or community initiatives and migrates them toward the full scheme, described in an EPC clarification paper.
On the infrastructure side, EBA CLEARING's R2P service is a pan-European messaging layer aligned with the EPC scheme, developed with the support of 27 users from 11 European countries according to its Annual Report 2025. That report also records that the Italian community selected R2P for payment requests from public administration, following a joint announcement in October 2025 by CBI, Nexi and EBA CLEARING, with the first users onboarded in early 2026 and the use case going live that year. In practice, the country where SRTP is most visibly in production is Italy, and it is doing it for public sector bills.
Reconciliation is the real prize for billers
For merchants with a normal card checkout, adding SRTP mostly changes who pays the fee and how disputes work. For billers — utilities, insurers, telecoms, B2B suppliers, landlords, public bodies — the payoff is different and larger, and it sits in reconciliation rather than in the payment itself.
The request carries structured ISO 20022 data: amount, currency, an expiry, requested execution date, remittance information, and the invoice or end-to-end identifiers. Because the same data set travels with the request and the resulting credit transfer, an ERP or accounting system can match incoming cash to the correct open item automatically. That is a step beyond a plain bank transfer with a free-text reference, which is where most "unidentified payments" and manual clean-up work originate.
Typical uses the EPC reports from existing participants include e-commerce, invoicing, payment collection in various contexts, and — worth noting for collectors — the recovery of failed direct debit transactions, where a request replaces a one-off retry. The rulebook also allows optional datasets for attachments, URLs and instalment amounts, though these are options: a provider that does not support them may reject the message with a reason code.
One caveat on the cost side of the argument. Consultancy cost-benefit work, such as the PPI whitepaper on Request-to-Pay, puts end-to-end automation savings in the range of roughly EUR 1.40 to EUR 3.90 per processed invoice when billing, banking and posting are joined up. Those are third-party estimates for a specific process design, not scheme figures, and they depend on invoice volume and on how much manual work exists today.
What implementation actually costs
There is no published retail price list, because SRTP is a business-to-business scheme. The cost lines to budget for are these:
- Scheme participation and homologation, including testing against the API specifications. The rulebook provides for annual participant fees set by the scheme manager.
- Build or buy on the payee side: either direct API integration to your bank, or a contract with a Referenced Technical Solution Provider or an infrastructure service such as R2P.
- Presentation and decisioning on the payer side, if you are the bank or the biller's software provider — this is where most product work sits.
- Back-office integration: ERP, e-invoicing, dunning and cash application logic that consumes the status messages.
- Ongoing messaging and connectivity fees charged by your provider or the infrastructure layer.
- Operational handling: expiries, refusals, unsupported optional datasets, payer opt-outs, and support for payers whose bank does not show the request.
Risks and costs
- Reachability risk. If the payer's bank or community has not joined, the request cannot be delivered. Design the payee journey so SRTP is one option, not the only path.
- No funds guarantee. A request is not a promise to pay. The payer can refuse or ignore it, and there is nothing to enforce. Optional request-for-guarantee datasets exist but depend on both providers supporting them.
- Finality cuts both ways. Because the payment leg is typically an instant credit transfer, it is not reversible in the direct-debit sense once completed. Refunds and disputes need a separate, deliberate process — different from card chargeback machinery.
- Fragmentation of options. Attachments, URLs, instalments and payment-guarantee requests are optional; a recipient may reject a message that uses one, with a reason code rather than a completed payment.
- Fraud and social engineering. The scheme's trust framework governs authentication between service providers, not what a payee does with the message. Fraudulent requests that mimic a legitimate biller are a payee-reputation and payer-education problem, and the presentation layer is where that must be handled.
- Cost of low-volume channels. If only a small share of your customer base is reachable, per-request and integration costs may not be covered by the reconciliation savings for some time.
- Data protection. Requests carry invoice and customer data. Collecting it in a messaging flow brings GDPR obligations that a plain bank transfer with a reference does not.
Fallbacks when the payer's bank is not reachable
Plan these before launch, not after:
- Keep cards and wallets at checkout for consumer reach and dispute handling.
- Use open-banking pay-by-bank initiation where available, which is covered in our open banking APIs guide. It requires the payer to complete a redirect-style flow and carries its own abandonment rates.
- Retain SEPA Direct Debit mandates for recurring collections. It is pull-based, needs a signed mandate, and comes with return and refund rights.
- Fall back to a QR code or a payment link plus a standard credit transfer, so the payer can still pay from a phone.
- Route the request through a payment orchestration layer that decides per transaction which rail to try — see our review of payment orchestration platforms.
- For failures inside SRTP itself — expiry, refusal, an unsupported option or a message rejected by the recipient — treat the status message as a trigger for a dunning or reminder process rather than a dead end.
Bottom line
SEPA Request-to-Pay is a well-specified, infrastructure-agnostic messaging layer with a clear reconciliation argument for high-volume billers, and a genuine convenience argument for consumer checkout. It is not a card killer and not a direct debit substitute, and it is not mandated, so its usefulness tracks reachability rather than its technical quality. For fintech product teams, the sensible sequence is: check the EPC register for the banks and communities your customers actually use, prototype one narrow use case — public sector or utility bills, or direct debit retries — measure the cash-application gain, and keep cards, mandates and pay-by-bank in the journey as fallbacks.
Follow the shortlist
Following payments, banking and market-infrastructure change? Our global ETF and stock research watchlist is tracked daily. Premium members get the full list, the reasoning behind every entry and updates when entries change.
See premium accessInformation and education only, not investment advice.
Related
Stablecoin Regulation Compared: EU MiCA vs the US GENIUS Act
How MiCA and the GENIUS Act differ on who may issue payment stablecoins, what must back them, how redemption works and where depeg risk still sits.
FinTechTokenised Deposits vs Stablecoins: How Bank-Issued Money Differs
Tokenised bank deposits and reserve-backed stablecoins both move value on programmable rails, but they differ in legal claim, deposit insurance, balance-sheet treatment and risk.
FinTechInstant Payments for Small Business: FedNow, RTP, Zelle Compared
How FedNow, RTP and Zelle differ for US small businesses on speed, settlement finality, limits and cost, plus which emerging options fit SME cash flow.