Chapter 1. Breaking down the cost of acceptance.
The “merchant fee” line on your statement, the MSC (merchant service charge), bundles several charges under one heading. It has three layers. Interchange goes to the cardholder's bank, while scheme fees (network fees) are collected by the card network (Visa, Mastercard, or Cartes Bancaires, France's domestic scheme). That leaves the acquirer's or PSP's margin, the only component you can genuinely negotiate line by line. Negotiating without understanding this breakdown is like haggling over a menu without knowing what the ingredients cost.
Where €1.00 in fees goes on a €100 transaction (EEA consumer credit card, online)
Merchant
pays €1.00 in MSC
The single line on the statement, to be broken down before any negotiation
Issuing bank
receives interchange: €0.30
Capped at 0.2% (debit) / 0.3% (credit) for EEA consumer cards by Regulation (EU) 2015/751
Network (Visa, Mastercard, CB)
receives scheme fees: €0.10 to €0.15
License, authorization, and clearing fees: a component that keeps rising and is hard to decipher
Acquirer / PSP
keeps the margin: €0.55 to €0.60
Covers risk, technology, and service. This is where the negotiation happens
0,2 % / 0,3 %
debit/credit interchange caps for consumer cards in the EEA
Regulation (EU) 2015/751, in force since December 2015+30 %
real-terms increase in the scheme fees charged to UK acquirers between 2017 and 2021
UK Payment Systems Regulator (PSR), market review of scheme fees, 20241,15 % / 1,5 %
debit/credit interchange caps on online transactions with cards issued outside the EEA
Visa and Mastercard commitments to the European Commission, 2019The networks whose rate cards shape your costsVisaMastercardCACartes Bancaires (CB)
🔑
What is actually negotiable
Interchange and scheme fees are set by regulation and by the networks. Your acquirer is bound by them just as you are. What you can negotiate is its margin, the fixed and ancillary fees, and the transparency with which it passes the rest on to you. So everything hinges on the pricing model, the subject of the next chapter.
🎯 Quick question
Out of €1.00 in fees paid by the merchant, who receives the interchange?
Chapter 2. Comparing blended and Interchange++ pricing.
Two pricing philosophies compete. Blended pricing charges a single rate (for example, 1.5% + €0.25) regardless of card type. It is simple and predictable but opaque: the acquirer's real margin is invisible and varies with your mix. Interchange++ (IC++) passes interchange and scheme fees through at cost and shows an explicit margin. The model is transparent and usually cheaper at significant volume, but it demands rigorous reconciliation. In between, Interchange+ (IC+) passes interchange through at cost but folds scheme fees into the margin.
| Criterion | Blended | Interchange+ (IC+) | Interchange++ (IC++) |
|---|---|---|---|
| Invoice readability | Excellent (one rate) | Average | Poor without tooling, but complete |
| Margin transparency | None: margin buried in the rate | Partial: scheme fees opaque | Full: margin shown in basis points |
| Who benefits from a favorable mix? | The acquirer (keeps the difference) | Shared | The merchant (pays actual costs) |
| Pass-through of regulatory cuts | Rarely automatic | Interchange only | Automatic and full |
| Typical profile | Small businesses, low volume, simplicity a priority | Mid-range volume | From ~€1M/year in card volume |
| Monitoring effort | Minimal | Moderate | Line-by-line reconciliation essential |
Pricing models compared
Worked comparison: €2M annual volume, €80 average ticket, 25,000 transactions
Card mix: 70% EEA debit, 20% EEA credit, 10% non-EEA / commercial
BLENDED 1.8% + €0.25/tx
Percentage fee : 2,000,000 x 1.8% = €36,000
Fixed fee : 25,000 x €0.25 = €6,250
TOTAL = €42,250 (2.11%)
IC++ margin 0.25% + €0.10/tx
EEA debit interchange : 1,400,000 x 0.2% = €2,800
EEA credit interchange : 400,000 x 0.3% = €1,200
Non-EEA interchange : 200,000 x 1.5% = €3,000
Scheme fees (~0.12%) : 2,000,000 x 0.12% = €2,400
Acquirer margin : 2,000,000 x 0.25% = €5,000
Fixed fee : 25,000 x €0.10 = €2,500
TOTAL = €16,900 (0.85%)
Difference: €25,350/year, more than half the blended cost.⚠️
The favorable mix trap
Under blended pricing, suppose 90% of your cards are domestic debit cards at 0.2% interchange. With that mix, the gap between the real cost and your single rate stays in the acquirer's pocket. Conversely, a mix heavy in commercial or non-EEA cards (hospitality, B2B, travel) can make an aggressive blended rate temporarily attractive, until the “non-qualified surcharge” clause pushes it up. Always demand a simulation based on your actual mix over the last 12 months.
🛍️
Neighborhood retail, < €500K/year
Simple blended pricing or a bank package. The potential savings from IC++ do not justify the reconciliation effort. Focus negotiations on fixed fees and terminal rental.
🛒
Online merchant, €1M to €10M/year
The tipping point. Always ask for both quotes (blended and IC++) based on your actual mix. IC++ almost always wins once most of the mix is domestic.
🌍
International merchant, > €10M/year
IC++ is a must, with periodic audits of scheme fee pass-through and automated transaction-level reconciliation.
🎯 Quick question
Which pricing model shows your acquirer's exact margin?
Chapter 3. Finding the negotiable lines: MSC, minimums, PCI fee, chargeback fee.
A card processing rate card rarely has fewer than 15 fee lines, so the negotiation is not just about the headline rate. Fixed fees, minimums, and ancillary fees weigh heavily on small tickets and irregular volumes. Some, like the PCI fee, are pure margin that many acquirers will drop if you make a reasoned request. Scrutinize every line to check that each fee reflects a real cost to the provider.
| Fee line | Observed range | Negotiability | Negotiating lever |
|---|---|---|---|
| Acquirer margin (MSC under IC++) | 0.05% to 1%+ depending on volume | High | Volume, competitive bidding, accurate mix data |
| Fixed fee per transaction | €0.02 to €0.35 | High | Top lever if the average ticket is low |
| Monthly minimum billing | €0 to €50+ | High | Remove once volume is steady |
| PCI fee (“compliance”) | €0 to €25/month | Very high | Often removed just by asking |
| PCI non-compliance fee | €15 to €40/month | Full | Complete your SAQ → €0 |
| Chargeback fee | €15 to €50 per dispute | Average | Monthly cap, waived if the dispute is won |
| Card terminal rental | €12 to €30/month | Average | Buy vs. rent, contract length, Android terminals |
| Setup / integration fees | €0 to €5,000 | High | Waived under competitive bidding |
| Declined authorization fees | €0 to €0.10 | Average | Cap them: you are paying for routing failures |
| Reporting, API, and payout fees | Varies | Average | Include in the overall package |
Rate card lines and how negotiable they are (observed ranges, mid-2026)
⚠️
The PCI fee does not fund the PCI Council
This “PCI DSS compliance” fee is a commercial line item set by the acquirer, not a regulatory pass-through. The PCI non-compliance fee, for its part, penalizes the lack of an up-to-date self-assessment questionnaire (SAQ). Completing that questionnaire each year takes a few hours and eliminates the line for good. Paying €30/month for non-compliance adds up to €360/year of pure negligence.
🔑
Negotiate line by line, decide on total cost
An acquirer that gives up 0.05% on its margin can make it up elsewhere: on the fixed fee, the monthly minimum, or the chargeback fee. The only honest comparison is the total annual cost simulated on your actual data: volume, number of transactions, card mix, dispute rate. Require this simulation from every bidder, on the same data set.
🎯 Quick question
Which of these fees can most often be removed just by asking?
Chapter 4. Benchmarking by volume and profile.
Before negotiating, you need to know where you stand. The ranges below reflect terms observed in mid-2026 for EEA consumer cards, excluding high-risk sectors (travel, gambling, CBD, and so on) that pay surcharges. A benchmark is never a quote. Two merchants with the same volume can see real costs differ by a factor of two, depending on their debit/credit mix, their share of foreign and commercial cards, their average ticket, and their dispute rate.
| Annual card volume | Dominant model | Observed terms | Negotiating stance |
|---|---|---|---|
| < €100K | Standard blended, published pricing | 1.4% to 2.9% + €0.20–€0.35/tx | Little room to maneuver: play published offers against each other and push on fixed fees |
| €100K to €1M | Negotiated blended | 1.0% to 1.6% all-in | Request an initial IC++ quote to expose the hidden margin |
| €1M to €10M | IC++ common | Margin 0.15% to 0.35% + €0.05–€0.10/tx | Formal competitive bidding, volume commitment in exchange for margin |
| €10M to €100M | IC++ as standard | Margin 0.05% to 0.15% + €0.02–€0.06/tx | Structured RFP, annual review clauses, pass-through audit |
| > €100M | IC++ / custom pricing | Margin < 0.05% possible, fixed fees in cents | Multi-acquiring, optimized routing, dedicated payments team |
Observed ranges by volume band (EEA consumer cards, mid-2026, indicative)
1,5 % + 0,25 €
Stripe's published price in France for a standard EEA card (excluding add-ons), the benchmark for an unnegotiated “list price”
Stripe France public pricing, checked mid-2026≈ 2,9 % + 0,35 €
approximate standard PayPal fee for a domestic payment in France: a wallet's convenience comes at a price
PayPal France public pricing, checked mid-20265x to 10x
cost gap between a capped domestic debit card (0.2%) and a non-EEA commercial card (1.5–2%+): the mix drives the cost
⚠️
A benchmark without the mix is meaningless
Comparing your 1.2% blended rate with a competitor's “0.8%” only makes sense if your mixes are comparable. A B2B site accepting uncapped commercial cards will inevitably pay more than a neighborhood store where 95% of payments are CB debit cards. Before any comparison, export 12 months of transactions and calculate your split between debit and credit, EEA and non-EEA, and consumer and commercial cards. That split is the centerpiece of your negotiating case.
🎯 Quick question
To compare your fees with a market benchmark, what information is essential?
Chapter 5. Running a multi-PSP RFP.
Beyond a certain volume, bilateral negotiation hits a ceiling. A structured competitive process (an RFP, or request for proposal) then unlocks the best terms. A good card acceptance RFP does more than compare rates. It evaluates total cost of ownership (TCO) over two to three years and the payment success rate, which often creates more value than price. It also weighs payment method coverage, integration quality, and ease of switching providers.
S-12
Internal scoping
Export 12 months of data: volumes, card mix, average ticket, success and dispute rates, countries. Define target requirements (payment methods, currencies, marketplace, in-store).
S-10
Shortlist and RFP launch
Four to six bidders at most, each receiving the same package: anonymized data, a mandatory response template, and a required simulation on your actual mix.
S-8
Q&A and quotes
Q&A sessions on equal terms for all bidders; require quotes in both blended AND IC++ pricing to expose the margin.
S-6
Presentations and scoring
Scoring against the weighted matrix defined up front, never adjusted after reading the bids.
S-4
Final negotiation
Two finalists in explicit head-to-head competition. This is when the PCI fee, minimums, and setup fees disappear.
S-2
Contract and due diligence
Legal review of sensitive clauses (see next chapter), migration plan, and exit plan.
Week 0 → week 8
Pilot, then switchover
Route 5% to 10% of traffic to the new PSP, compare payment success rates and actual costs, then migrate gradually.
| Criterion | Weight | What to measure |
|---|---|---|
| Total cost (3-year TCO) | 30 % | Simulation on the actual mix: percentage fees + fixed fees + ancillary fees + migration costs |
| Approval rates | 25 % | Performance measured in the pilot, 3-D Secure handling, smart retries, TRA exemptions |
| Payment method and country coverage | 15 % | Cards, wallets, bank transfers, BNPL, currencies, and required local entities |
| Integration and API | 10 % | Documentation quality, test environment, estimated development effort |
| Reconciliation and reporting | 10 % | Usable statements, IC++ granularity, automated exports to accounting |
| Support and SLA | 5 % | Dedicated contact, escalation times, uptime commitments |
| Financial strength and compliance | 5 % | Authorizations, licenses, financial health, incident history |
Sample scoring matrix for a card acceptance RFP
🔑
Payment success often matters more than price
On €10M of online volume, one point of payment success rate gained is worth about €100,000 in additional collected sales, often more than you would save by shaving a few basis points off the margin. That makes a side-by-side pilot under real conditions the most profitable step in the entire RFP. Promised success rates only count once they have been measured on your own traffic.
Providers and payment methods often evaluated in a European RFPStripeAdyenPayPalApple PayGoogle PayKlarnaWEWero
🎯 Quick question
In an RFP scoring matrix, why does the payment success rate often carry almost as much weight as price?
Chapter 6. Avoiding contract traps and managing the contract over time.
Three carelessly reviewed clauses can ruin the best rate card in the world. A card acceptance contract runs for years and is asymmetric by design. The acquirer carries financial risk (disputes, fraud, merchant default) and protects itself with commitment, revision, and holdback clauses. All of them are legitimate in principle. All of them are negotiable in scope.
⛓️
Long commitment + automatic renewal
An initial commitment of 36 to 48 months, automatically renewed for 12-month periods, with a short termination notice (60 to 90 days) that is easy to miss. Aim for a short initial term, reasonable notice, and a reminder of the renewal date in your management calendar.
📈
Unilateral price changes
A clause allowing the acquirer to change the rate card by simple notice (often 30 days). What to demand in return: a right to terminate without penalty if prices rise, and pricing frozen for the fixed term.
🌫️
Vague scheme fee pass-through
Under IC++, the “scheme fees” line must be passed through at cost. Yet some contracts define it vaguely, leaving room for hidden markups. Demand the list of fees passed through and an annual audit right.
🧊
Rolling reserves and holdbacks
A holdback of 5% to 10% of volume for 90 to 180 days to cover the acquirer's risk. Acceptable at launch or in a high-risk sector, but cap the amount, limit the duration, and set objective release criteria.
🧾
Minimum billing and dormant fees
Monthly minimum fees, inactivity fees, end-of-day batch upload fees, paper statement fees: inconspicuous lines that hurt seasonal businesses. Have them removed or explicitly capped.
🚪
Exit fees and friction
Early termination penalties, “deinstallation” fees, retention of token data. Negotiate at the outset: portability of network tokens and export of transaction data at no extra cost.
| Clause | Risk | Negotiation counter |
|---|---|---|
| Term and renewal | Locked in for 3–4 years, missed exit window | 12–24 months, 90 days' notice, internal calendar alert |
| Price revision | Price increase imposed mid-contract | Penalty-free termination if prices rise, pricing frozen for the fixed term |
| Settlement period | D+3 or later: higher working capital needs | D+1 business day written into the contract, penalties for late payouts |
| Reserves and holdbacks | Cash tied up indefinitely | Cap, maximum duration, objective release criteria |
| Exclusivity | Ban on multi-acquiring | Strike it outright: redundancy is good practice |
| Currency conversion (FX/DCC) | Opaque FX margin on settlements | FX margin set in the contract in basis points, choice of settlement currency |
Clauses to review line by line before signing
⚠️
The contract lives on after signing
Schedule an annual review: check invoices against the rate card, verify pass-through, re-benchmark the market, and flag upcoming renewal dates. Pricing setup errors happen, and they rarely work in your favor. An unmonitored card acceptance contract tends to drift upward through a series of small notices that nobody reads.
✅
The signer's checklist
Before signing: a simulation on your actual mix from every bidder, an IC++ quote, the PCI fee and minimums removed, the chargeback fee capped. In the contract: an exit right if prices rise, no exclusivity, guaranteed token portability. If a provider refuses any of these points without explanation, that refusal already tells you something about them.
🎯 Quick question
Your draft contract provides for a 36-month commitment with 12-month automatic renewal. What should you negotiate first?
Advertisements