A B2B software vendor signs a three-year agreement worth $1.2m. On paper the cash flow looks beautiful: same amount, same day, every month. Eighteen months in, the finance team notices something uncomfortable. Failed renewals and late collections have quietly eaten a slice of recurring revenue — not because customers left, but because a card expired, a direct debit bounced, or the buyer's procurement team simply never moved a payment method they promised to change.
Consumer subscription billing has an entire industry of tooling behind it. B2B subscription and recurring payments have the same recurring revenue model with almost none of the same assumptions: larger tickets, longer terms, a buyer who pays from a business bank account, approval flows, purchase orders, and often a currency that is not yours. The machinery does not map cleanly, and that gap is where revenue leaks.
Why recurring B2B payments are harder than they look
Three differences matter more than the rest.
First, ticket size. A $40,000 annual licence sitting on a card invites a card-limit decline and a costly interchange bill; the same amount by bank transfer is cheap but offers no automation at all. Second, who pays: your user is not your payer. The person who renews the contract is rarely the person who releases the funds, so "update your card" emails land in the wrong inbox. Third, rails and currencies: a US vendor collecting from buyers in Europe, India and Brazil is juggling card networks, SEPA, local bank debits and cross-border wires — each with its own mandate, timing and failure mode.
Four ways to collect recurring B2B revenue
Most B2B companies end up using more than one, usually by accident rather than design. Here is the honest comparison.
| Method | Best for | Cost | Main weakness |
|---|---|---|---|
| Card on file | Smaller recurring tickets where speed matters | Highest — percentage-based, plus cross-border | Card limits, expiry, and corporate card policy |
| Direct debit (SEPA, ACH, Bacs) | Domestic and intra-region recurring collections | Low, often a fixed fee per item | Coverage is regional; mandates must be managed |
| Bank transfer + invoice | Large tickets, procurement-driven buyers | Low transfer fee, high manual effort | Buyer has to act every cycle; no automation |
| Virtual cards & open banking | Controlled spend and variable recurring debit | Varies by rail and scheme | Coverage still uneven across corridors |
Figure: recurring B2B collection methods. The trade-off is rarely price alone — it is automation versus control versus coverage.
The trap is not choosing the wrong method. It is letting each customer choose a different one and never consolidating the resulting data. When half your book is on cards, a quarter on SEPA, and the rest on bank transfer, the question "which subscriptions are actually at risk this month?" has no single source of truth.
Mandates: the plumbing nobody wants to own
Every automated recurring debit rests on a mandate — the customer's standing authority for you to pull funds. SEPA Direct Debit, ACH debit authorisations, Bacs Direct Debit and UK variable recurring payments each have their own rules, formats and dispute timetable. Mandates have references, effective dates, and scope (fixed amount or variable); they can be amended, and they can be cancelled at the worst possible moment. Managing them badly shows up as "unauthorised" returns, which carry fees and, at volume, threaten your ability to collect by debit at all. If you collect across borders, mandate management is a product feature, not an admin task.
Involuntary churn: the revenue leak you can actually fix
Failed payments are not the same as lost customers. A large share of failed renewals are involuntary churn: the customer still wants the service, but the transaction failed. The usual causes are mundane — an expired or replaced card, a soft decline from the issuer, insufficient funds on the debit date, or a change of bank. Seen across a portfolio, these failures quietly erode net revenue retention.
The good news is that involuntary churn responds to process, not to pricing. Four levers do most of the work:
- Account updater services that refresh expired or replaced cards before the renewal date, so the first attempt actually lands.
- Smart retry timing. Re-presenting a failed debit at the wrong moment — a Friday evening, or the day before a payroll run — fails again and can trigger issuer blocks. Timing the retry to when funds are likely to be present recovers more.
- Pre-dunning reminders to the payer (not the end user), a few days ahead of a large renewal.
- A fallback rail. When a card declines on a $40,000 renewal, the fix is often an invoice and a bank transfer — not a third retry.
A dunning sequence that keeps the customer
Dunning has a reputation for being hostile. Done well, it is a service recovery flow. A sequence that works for B2B renewals looks roughly like this:
T−7 days: a heads-up email to the billing contact, with the amount, date, and a one-click link to update the payment method. T−0: first presentment. T+1: if it fails, a soft, low-friction retry on an alternative day. T+3: a second retry, plus an in-product banner for the account admin. T+7: a direct email to the billing and procurement contacts offering a bank-transfer alternative. T+14: route to the account manager, because a renewal this size is worth a human. T+21: formal notice, with a short grace period before service is suspended — and a plan to reinstate without penalty.
The mistake to avoid is uniformity: the same three aggressive retries, at the same intervals, for every customer. A $200 seat and a $40,000 renewal deserve different treatment, and so does a first missed payment versus a chronic one.
Multi-currency pricing and the billing-versus-settlement gap
Pricing a subscription in the buyer's currency removes friction — the invoice matches their budget and their bank's reporting — but it moves FX risk onto you, and unlike a one-off payment, that risk compounds every cycle. It also creates a subtle trap: the currency you bill in, the currency you settle in, and the currency your books report in are often three different things. Teams that price in local currencies without a settlement plan discover a steady FX leak that is invisible on any single invoice but obvious across a year of renewals. The mechanics are worth understanding in depth; we cover them in our guide to multi-currency B2B payments.
Reconciling recurring revenue
Recurring collections generate a specific reconciliation headache: hundreds of debits, each mapped to a subscription, a contract period, and an invoice, with some failing, some partly paid, and some arriving with the wrong reference. If reconciliation is manual, the month-end close becomes a detective exercise and failed payments stay hidden until someone notices a customer still using the product without paying. Automating the match between incoming funds and subscription records is the difference between managing churn proactively and discovering it a quarter late — the approach is set out in our guide to automated B2B payment reconciliation.
Tax, mandates and recurring: the layer people forget
Recurring cross-border collections add a tax dimension that one-off invoices can usually ignore. Licensed software can trigger withholding tax on royalties; services delivered into another jurisdiction may fall under reverse charge; and some countries require local registration once recurring revenue crosses a threshold. None of this is exotic, but it has to be designed into the billing flow rather than bolted on later. We unpack it in our guide to VAT, GST and withholding tax on cross-border B2B payments. And when a plan changes mid-term — an upgrade, a downgrade, a disputed charge — the recovery playbook in B2B payment disputes and chargebacks is the one to keep to hand.
Designing the stack without over-building
The instinct is to reach for a consumer subscription-billing platform and hope it stretches to B2B. Sometimes it does — for low-ticket, card-first products. Beyond that, the requirements diverge: multi-rail collection, mandate management, invoice-grade documents, and settlement across currencies. The practical middle ground is a billing layer that owns the subscription logic, paired with a payments layer that owns the rails and the recurring collection mechanics. How to draw that line — and where integrations usually break — is covered in how to choose and build a B2B payment API.
Frequently asked questions
Should a B2B SaaS business take cards by default?
For low-ticket, self-serve plans, yes — cards convert best and the cost is tolerable. As tickets grow, bank transfer or direct debit becomes cheaper and more robust, even though it is less automated. Many vendors offer both and steer by ticket size and region.
Our customer insists on paying subscriptions by bank transfer. Can we still automate?
Partly. You can automate the invoice, the reminder cadence, the reconciliation and the dunning — but the buyer still initiates each payment. Automation there is about reducing manual matching and chasing, not about pulling funds.
What is a "good" involuntary churn rate?
It varies enormously by method, geography and ticket size, so benchmarks are less useful than your own trend. What matters is whether failed payments are measured at all — and whether a retry-and-recover process exists. Teams that instrument it usually find recoverable revenue in the first quarter.
Do we need a local entity to collect by direct debit?
Not always, but it depends on the scheme and the corridor. Some debit schemes are accessible through a partner or sponsor; others effectively require local presence. Card and bank transfer have far fewer such constraints, which is one reason they remain the fallback for cross-border recurring revenue.