Early-payment discounts fail for a mechanical reason, not a financial one. A supplier offers 2% for payment within 15 days; the buyer’s accounts payable function cannot get a paper invoice through receipt, coding, matching and approval in fewer than 23 to 25 days. The discount was never refused — it expired before anyone could act on it. That gap between the discount window and the approval cycle is the single variable that decides whether an early-payment programme produces cash or a spreadsheet.
E-invoicing closes it. And because tax mandates are already forcing structured, dated, validated invoices through a controlled channel, the data that makes dynamic discounting work is a byproduct of a compliance project most European businesses are running anyway. The working capital case is the part of that project nobody costed.
Approval latency, not postage, is what kills the discount
The savings usually quoted for e-invoicing — printing, envelopes, postage, storage — are real but small, and they are not where the working capital sits. What matters is elapsed time between the chargeable event and the moment an invoice is approved for payment, because that is the window inside which any early-payment decision has to be made.
Paper destroys that window in three ways at once. Invoices arrive decentrally, so the cash manager has no view of the payable population. Data entry introduces errors, and a significant share of invoices — commonly 20% to 30% in the analyses done during the paper era — end up handled as exceptions rather than straight-through. And approval routing runs on physical circulation, which is why 10% to 15% of invoice volume ends up generating a payment reminder that would never have been needed.
None of this is speculative. The e-invoicing analyst Bruno Koch documented a client that had been missing discounts worth an average of €1.50 per paper invoice, and concluded that “the discount benefits alone more than compensate the project costs” of moving to electronic invoicing. That is a payback argument built entirely on discounts foregone, before counting a single euro of processing saving.
What e-invoicing actually changes about the cash cycle
Three changes matter, in descending order of value.
- Visibility. An e-invoicing programme almost always produces a single inbound gateway. Every payable becomes visible on receipt rather than on entry, which is the precondition for managing the payable population as a portfolio instead of a queue. Historically, 30% to 35% of larger organisations still handled invoices in a decentralised way, across multiple ERP and accounting systems — a structure in which no one can state the total amount and status of open payables on demand.
- Latency. Automated import, validation and line-item matching against the order remove the days that discount windows are lost in. Reported reductions in days sales outstanding cluster in the low single digits — one German survey put the improvement at 5.4 days.
- Accuracy. Structured data validated at the point of issue means fewer disputes, and disputes are where the long tail of unpaid invoices lives.
The scale of the prize depends on how far the automation goes. Analysis at Politecnico di Milano put the net benefit of VAT-compliant e-invoicing at €4–12 per invoice, rising to as much as €65 per cycle where the whole trade process is integrated. Those figures date from research presented in 2010 and are best read as a ratio rather than a current price: full integration was worth several times invoice digitisation alone.
How dynamic discounting works, and what makes it fail
Dynamic discounting replaces a fixed term such as “2/10 net 30” with a sliding scale. The buyer allocates a pool of liquidity, sets limits and an implied rate, and once an invoice is approved the supplier is offered early payment at a discount that decreases the closer the date is to the original due date. The supplier chooses, invoice by invoice. In practice this shows up either as a plug-in inside the ERP or as a “pay me early” action on a buyer portal or a network operator’s portal.
The design is attractive because it is self-funding for the buyer and optional for the supplier. It fails for predictable reasons:
- The approval clock is too slow. If invoices are approved on day 22, a discount curve that starts on day 5 has nothing to sell. Every day removed from approval latency is a day of curve that becomes offerable.
- Approval is confused with acceptance. Only approved invoices can be discounted. A programme built on invoices that have merely been received will offer early payment on amounts that later get disputed.
- Supplier reach is too narrow. The suppliers who most value early cash are the smallest, and they are the hardest to onboard. A programme that only reaches the top 50 suppliers by spend usually reaches the ones with the cheapest access to credit.
- The buyer’s cost of capital is wrong for the offer. An annualised discount that undercuts the buyer’s own return on cash is a subsidy, not a treasury strategy.
Dynamic discounting also sits beside, not inside, supply chain finance. In receivables finance and reverse factoring a third party funds the early payment; in dynamic discounting the buyer funds it from its own balance sheet. The distinction matters for accounting treatment and for the regulatory point in the next section.
The legal ceiling on payment terms
This is the constraint most vendor material omits: you cannot manufacture the working capital benefit by extending terms, because European law caps them.
Under Directive 2011/7/EU, business-to-business payment terms default to 30 days and may be extended to 60 only where expressly agreed; anything longer is valid only if it is not grossly unfair to the creditor. Public authorities pay within 30 days, with a 60-day exception available in limited cases such as public health entities. Late payment triggers statutory interest of at least the European Central Bank reference rate plus eight percentage points, plus a fixed recovery sum of at least €40 per invoice under Article 6 — owed automatically, without a reminder.
Article 3(5) is the provision that bears directly on approval latency: the verification or acceptance procedure itself may not exceed 30 calendar days from receipt of the goods or services unless expressly agreed and not grossly unfair. The law already treats slow approval as a form of late payment.
A replacement instrument is in the legislative pipeline and stuck there. The Commission proposed a Late Payment Regulation in September 2023 with a hard 30-day cap and no contractual extension; Parliament adopted its position on 23 April 2024, softening this to allow 60 days by express agreement and up to 120 days for specified retail categories, raising fixed recovery compensation to between €50 and €150 by transaction value, and adding annual public payment-practice reporting by contracting authorities and a European Observatory of Late Payment. As of the European Parliament’s legislative train entry dated 22 May 2026, the file’s status is blocked, with the Council identified as the blocking institution. The 2011 Directive remains the operative law.
The practical reading: a treasury strategy that works by pushing days payable outstanding further out is aimed squarely at the practice the proposed Regulation exists to stop, and it carries political risk even while the file is stalled. A strategy that works by pulling approval in, and paying earlier in exchange for a discount, moves in the direction regulators are pushing. The Commission’s own reasoning in the proposal is that late payment functions as free finance for the debtor at the creditor’s expense.
Why the mechanics work in 2026 in a way they didn’t a decade ago
The working capital argument for e-invoicing was made a decade ago and largely ignored, because two pieces of plumbing were missing. Both now exist.
Settlement is no longer a constraint. Under the Instant Payments Regulation, Regulation (EU) 2024/886, euro-area payment service providers have been obliged to receive instant euro credit transfers since 9 January 2025 and to send them since 9 October 2025, with funds credited within ten seconds and charges no higher than for a standard credit transfer. Providers in non-euro Member States follow from 9 July 2027. A discount curve priced by the day used to run into batch payment runs; it no longer has to.
The request for payment has a scheme behind it. The European Payments Council’s SEPA Request-to-Pay scheme (EPC014-20), whose version 4.0 took effect on 5 October 2025, names electronic invoice presentment and payment as a use case and is built on ISO 20022 messaging. An early-payment offer can be a scheme message rather than a portal notification.
One consequence deserves attention from anyone maintaining supplier bank details. Since 9 October 2025 the same Regulation requires euro-area providers to offer Verification of Payee, checking the payee name against the account identifier before a transfer is authorised, free of charge. A payer who proceeds after a mismatch warning carries the liability. Supplier bank master data has stopped being a back-office field and become a payment control.
What ViDA adds from 2030
Council Directive (EU) 2025/516 makes two changes that bear directly on the payment side, both from 1 July 2030.
The first is a new Article 226 point (17): the invoice must carry the supplier’s bank account numbers, virtual account numbers, or other identifiers that unambiguously identify the accounts into which the invoice can be paid. Recital 16 states the purpose plainly — administrations want to follow the financial flows and not only the goods. The side effect for business is that remittance data stops being an optional attachment and becomes mandatory structured content, which removes one of the most common causes of unapplied cash.
The second is timing. The deadline for issuing an invoice for a cross-border supply drops to 10 days after the chargeable event, replacing the fifteenth day of the following month. Since payment terms under the Late Payment Directive run from receipt of the invoice, issuing sooner starts the clock sooner. For a supplier billing intra-EU customers monthly in arrears, this alone can pull weeks out of the cash cycle without renegotiating a single contract.
Where the case is weaker than vendors claim
Three honest limits.
The discount economics depend on your cost of capital, and for a buyer with cheap access to credit and a low return on cash, funding early payment from the balance sheet may be a worse use of liquidity than paying on terms. The calculation is company-specific and changes with rates; nobody can tell you the answer from outside.
Adoption, not capability, is usually the binding constraint. E-invoicing mandates deliver structured invoices, but they do not deliver supplier enrolment in a discount programme. Suppliers who do not need cash will not take the discount, and suppliers who need it most may not have the treasury capacity to evaluate an offer invoice by invoice.
And the working capital benefit is asymmetric. The buyer captures the discount; the supplier trades margin for certainty. Framing a dynamic discounting programme as mutual benefit is defensible only where the supplier’s alternative cost of finance genuinely exceeds the discount rate being offered. Where it does not, the programme is a price reduction with extra steps, and suppliers will price it back into the next contract.
Building the business case: the four numbers you need
A credible working capital case for an e-invoicing programme needs four measurements, three of which most organisations do not currently have.
- Current approval latency, measured from invoice receipt to approval-for-payment, at the median and the 90th percentile. The tail matters more than the average, because the tail is where discounts and disputes live.
- Discounts offered versus discounts captured, in euro, for the last twelve months. Most systems record the terms and never record the miss. If you cannot produce this number, that gap is itself the finding.
- Exception rate and its causes. Split into data errors, matching failures and genuine commercial disputes. Only the first two are fixed by structured invoicing.
- Your own hurdle rate for deploying cash early. Without it, the discount curve cannot be priced and the programme is guesswork.
Run those four against the compliance programme you are already funding. In most organisations the mandate is being scoped, staffed and budgeted as a tax project, and the working capital benefit is left unclaimed on the same data.
Frequently asked questions
What is dynamic discounting, in one sentence?
It is an arrangement in which a buyer offers to pay an approved invoice before its due date in exchange for a discount that shrinks as the due date approaches, funded from the buyer’s own liquidity, with the supplier deciding invoice by invoice whether to accept.
How is it different from supply chain finance?
In dynamic discounting the buyer pays early from its own balance sheet. In supply chain finance — reverse factoring, receivables purchase, payables discounting — a third-party funder pays the supplier early and is repaid by the buyer on the original due date. The first is a use of cash; the second is a financing arrangement, with different accounting and disclosure consequences.
Can we extend payment terms to fund an early-payment programme?
Not freely. Directive 2011/7/EU caps B2B terms at 60 days by express agreement, with anything longer valid only if not grossly unfair to the creditor, and holds public authorities to 30 days in most cases. Extending terms to create room for a discount programme is also the specific practice the proposed Late Payment Regulation was drafted to curtail.
Does an e-invoicing mandate automatically improve working capital?
No. A mandate delivers structured invoices reaching a tax authority; it does not by itself change how quickly your approval workflow runs or whether you act on the resulting visibility. The compliance project creates the conditions. Capturing the benefit requires separate decisions about approval routing, tolerance thresholds for automatic matching, and who owns the early-payment offer.
What happens to the late payment rules now that the Regulation is blocked?
Directive 2011/7/EU continues to apply in full across all Member States, as transposed nationally. A blocked file is not a withdrawn one, and Parliament’s position is already adopted, so the proposal can move again if the Council’s position shifts. Contracts written today should assume the current Directive and avoid terms that would need renegotiating under a 30-day cap.