I get this question weekly from EU finance teams that now sell a lot into the UK, and UK platforms that collect in the EU: “How do we move money across the Channel fast, cheap, and with fewer mysteries?” Post-Brexit, the tools didn’t vanish—they just punish sloppy design. The short answer is to treat the corridor as two local systems stitched together: SEPA (ideally SEPA Instant) on the euro side, Faster Payments (FPS) on the sterling side, and a treasury layer that does FX on your terms instead of on your customers’.
The symptoms you’re probably living with
- Slow cross-border wires. Traditional SWIFT EUR→GBP or GBP→EUR hops land T+1 or T+2 with intermediary fees and unpredictable value dates.
- Short-pays and broken references. Intermediary deductions and truncated remittance data make reconciliation a monthly sport.
- Refund drag. Returning money cross-border takes days; customers escalate before finance can even see the return credit.
- Authorization fall-off. If you rely on cross-border card acquiring alone, you’ll feel issuer risk filters and higher costs on both sides.
None of these are laws of nature; they’re design choices. The fix is to collect and pay out locally, then let treasury move liquidity in bulk.
The north star in one sentence
Let buyers and suppliers experience a domestic payment in their currency; do FX and liquidity behind the curtain with clear rules. That’s it.
Rails that actually work on this corridor
- SEPA Credit Transfer (SCT) and SEPA Instant (SCT Inst) for EUR. Most EU payers can push EUR within seconds on SCT Inst. Even plain SCT is predictable and cheap with good remittance fields.
- UK Faster Payments for GBP. Real-time for most banks, cheap, and reliable for both B2B and payout use cases.
- Cards as a precision tool. Keep domestic acquiring on each side for retail flows and long-tail payers, but don’t make cards carry the whole corridor.
- CHAPS and TARGET2/RTGS exist for high values and cut-off-sensitive moves, but your day-to-day e-commerce and payouts should live on SCT/SCT Inst and FPS.
The trick is to hold real EU-side EUR accounts and real UK-side GBP accounts (directly or via a licensed partner), and to issue virtual IBANs or payer-bound references so reconciliation is automatic.
Operating model that turns two systems into one experience
- Collections
- EU buyers pay EUR to a virtual IBAN on your EU account. Funds arrive via SCT Inst when available; otherwise SCT same-day/next-day.
- UK buyers pay GBP over FPS to your UK account.
- Refunds
- Refund in the same currency and rail the payer used: EUR→SCT Inst/SCT, GBP→FPS. State a value-date promise agents can quote.
- Payouts
- EU suppliers and contractors receive EUR via SCT/SCT Inst.
- UK sellers, riders, creators receive GBP via FPS on a fixed calendar with visible cut-offs.
- FX and liquidity
- Treasury runs time/threshold sweeps between the EUR and GBP pools. Convert in bulk (not per transaction), publish achieved-vs-benchmark rates, and book the realized spread.
- Reconciliation
- Webhooks or bank notifications post “funds received/refund executed” into your ERP with the same virtual reference that appeared on the invoice or checkout.
Customers feel local. Finance feels predictable. Support stops guessing.
Why SEPA Instant + FPS beats “international wires”
- Speed. SCT Inst and FPS are near-instant during banking hours. Even classic SCT tends to beat SWIFT chains for end-to-end clarity.
- Cost. Domestic rails are low-fee and transparent; you stop bleeding to correspondents.
- Data. SEPA remittance fields and FPS references survive end-to-end, so auto-match rates exceed 95% if you use virtual IBANs.
- Refund UX. Same-rail refunds are hours, not days. Your ticket queue shrinks the moment agents can quote a date confidently.
Compliance and “small but sharp” policy decisions
- KYB/KYC at the edge. Onboard payees with name/IBAN checks, sanctions/PEP screening, and basic documentation. Don’t push heavy checks to buyers unless risk flags tell you to.
- Name/account lock. First payout must match the verified beneficiary name; account changes get an out-of-band confirmation and a cooling-off period.
- Country fences and MCC hygiene. Some categories need stricter SCA or longer payout tails. Encode this in policy so ops isn’t improvising.
- IBAN acceptance. Keep customer comms and forms tolerant of GB IBANs on the EUR side where supported; avoid unnecessary “IBAN discrimination” that creates support friction.
FX that your CFO can defend and your customers can’t feel
- Pools and thresholds. Keep 7–10 days of outflows in each currency; sweep above a ceiling, top up below a floor.
- Benchmarking. Record every conversion’s achieved rate vs a neutral benchmark so spread is measured, not argued.
- Rate visibility where it matters. Finance and auditors see rates and timestamps; buyers and sellers do not see mid-transaction FX unless your product explicitly offers it.
Your prices stay in shopper currency; your suppliers get paid in their own. FX lives in treasurer space, not in the UI.
Refunds with clocks, not vibes
Publish promises you can keep. For example:
- EUR refunds: “Approved today → credited via SEPA Instant within hours during banking windows; otherwise next business day.”
- GBP refunds: “Approved today → FPS within minutes; after cut-off, early next business day.”
Back it with a pre-funded refund bucket on each side so agents never see “insufficient balance to refund.”
Calendars beat courage: settlement and cut-offs
- EU side: Daily SCT Inst for exceptions; main batch SCT files before bank cut-off to guarantee same-day value where possible.
- UK side: Daily FPS payouts with a hard seller dashboard cut-off (e.g., 16:00 local) for same-day value; late approvals slide to the next banking day.
- Holidays and half-days: publish calendars in both time zones so ops doesn’t discover them the hard way.
A calendar eliminates a hundred “where is my money” emails.
Reconciliation you can actually close on day two
- Virtual IBANs / payer-bound references for all bank-transfer methods.
- Evented accounting: “received,” “settled,” “refunded,” “swept/converted” land as immutable events with timestamps.
- Exception lanes for partials, duplicates, and late credits after invoice expiry with scripted outcomes.
95% auto-match is normal once you stop relying on free-text.
Risk and dispute hygiene that lower cost without hurting approvals
- Risk-scored SCA/3DS. Use TRA exemptions on low-risk EU card traffic; insist on 3DS for high-risk patterns and cross-border cards.
- Descriptors that buyers recognize. Include your brand and, if allowed, a seller/campaign slug. This is the cheapest chargeback reducer you’ll ever ship.
- Evidence packs. When a dispute lands, assemble proof automatically from the ledger: order details, delivery evidence, comms history, refund attempts.
Fewer disputes start; the rest close faster.
What changed for teams that adopted this model
- Collection and payout speed shifted from days to minutes/hours on both sides for most tickets.
- Unit cost improved by triple-digit basis points after moving volume off SWIFT chains.
- Auto-reconciliation crossed 95% with virtual references; month-end stopped being a forensic exercise.
- Refund tickets fell sharply because agents quoted value dates and hit them.
- FX realized spread became a measured metric, not a feeling—leading to better vendor pricing.
Different businesses, same shape of results.
Build order you can deliver this quarter
Week 1–2: Stand up EUR SEPA (preferably Instant) and GBP FPS accounts with virtual IBANs/references. Publish a one-page “How to pay” for both sides.
Week 3–4: Wire webhooks into your ERP; auto-close invoices on receipt; add refund flows with value-date calculators for agents.
Week 5–6: Implement FX sweeps with thresholds and achieved-vs-benchmark reporting; pre-fund refund buckets.
Week 7–8: Ship seller/biller dashboards with cut-offs and expected value dates; encode SCA/3DS policy by risk band.
Week 9–10: Harden exception lanes, holiday calendars, and name/account-lock on payouts; publish the corridor policy so nobody improvises.
Ship the corridor first. Fancy routing can wait.