Who we serve › E-commerce
Payment infrastructure for e-commerce merchants processing $100K+ a month
TL;DR. At $100K+ a month, the payments stack most e-commerce merchants run was chosen when they were small: one platform, a blended rate, a single approval number, and no second path. It works until it is expensive or until the platform's risk model changes its mind. The job is not to replace it. It is to read the price on your real card mix, measure approval by issuer and market, use the authentication exemptions you are entitled to, add methods where checkout actually drops, and build a second acquirer that carries live volume before you need it. We do this for mainstream retail as often as for anything else; our largest consolidated engagement is a sportswear retailer.
Who this is for. E-commerce merchants processing $100K+ a month in cards who want to know what the stack really costs, why good customers get declined, and what happens if the one provider says no.
Five things an acquirer owes a merchant at this volume.
A price you can read
IC++ on your real card mix: interchange and scheme fees at cost, a disclosed margin. A blended rate is the acquirer's assumptions about your mix, priced in its favour. IC++ vs blended →
Approval you can see
Approval rate by issuer, by market, by card type, every month. One number is not a measurement. The gap between a 93% and a 96% approval rate is usually worth more than the whole fee negotiation. Run the numbers →
Exemptions actually used
In Europe and the UK, SCA exemptions (transaction risk analysis up to €500 by acquirer fraud rate, low value, merchant-initiated) are where conversion lives. Ask what share was exempted last month, by type.
Methods where checkout drops
Open banking, BNPL, wallets, added where the data shows abandonment or card declines by market, not where a catalogue lists them. Each one costs reconciliation; earn it.
A second path
A second acquirer carrying a bounded share of traffic, routed by market or card type, before the incumbent's risk model or pricing changes. Leverage on the day you negotiate; continuity on the day you need it. Multi-rail →
Terms that move with history
Reserve, cap, settlement timing and change rights set when you were small should not survive unchanged at ten times the volume. A second quote is what moves them.
Four patterns we see in e-commerce at this volume.
01
The cheaper offer that costs more.
A lower headline rate, accepted; approval drops a point and a half on the new acquirer's issuer relationships. Net revenue falls while the fee line looks better. Compare offers on net approved revenue, never on rate.
02
Everything challenged.
The provider sends every European transaction to 3-D Secure because it is simpler for them. Conversion loss at the authentication step is never reported as a decline, so nobody owns it. Exemption strategy is a line in the contract, not a setting.
03
The platform decides.
A growth month, a dispute uptick, a category reclassification: the single platform adds a reserve or closes the account with a form email. Same model that onboarded you in a day. The first 72 hours →
04
Terms from year one.
Pricing, reserve and settlement timing set at $20K a month, still in place at $500K because nothing broke. The acquirer will not volunteer a repricing; the second quote is what starts it.
Diagnose. Improve. Prove.
Diagnose
All-in cost per transaction on your real card mix. Approval by issuer, market and card type. Authentication and exemption rates. Checkout abandonment by market and method. Dispute ratios per scheme. Current terms: reserve, cap, settlement, change rights.
Improve
Reprice on IC++ with offers from acquirers that fit your footprint. Fix the exemption strategy. Add methods where the data says. Build the second acquirer and the routing between them. Renegotiate the incumbent with the second quote in hand.
Prove
Move a bounded share of traffic, by market or card type. Measure net approved revenue, cost per transaction, dispute ratio and settlement timing against the baseline. Expand as the data earns it; keep the incumbent until the second path has carried real volume.
Mainstream retail is not a side business for us.
Our largest consolidated engagement is a sportswear retailer selling in 26 countries across three continents: several regional providers replaced by one acquirer with local presence in every market, ~20% cheaper. In 97% of ~200 pricing reviews across verticals, the offer we brought was cheaper than the incumbent.
We are paid by the provider you end up using, as a commission on volume, so the review costs you nothing and we have no reason to push one provider over another. How we work →
We are on a single global platform and it works. Why look at this?
Because at $100K+ a month three things are usually true and invisible: you are paying a blended rate that hides what your card mix actually costs, your approval rate is a single number nobody has broken down by issuer or by market, and your entire revenue runs through one risk model that can add a reserve or close the account without a conversation. None of that is a reason to leave. It is a reason to know the numbers, and to have a second path that already carries live volume before you need it.
What is the real cost of card acceptance?
Fee rate times volume is the small part. The larger part is the approval rate: every percentage point of good traffic declined is revenue gone, and a cheaper acquirer with a worse approval rate costs more. Then come the pieces hiding in structure: interchange and scheme fees passed through or bundled, cross-border uplifts on cards you did not know were foreign, FX on settlement, chargeback fees. We put all of it on one line per transaction and compare offers on that. The true-cost calculator shows the fee-versus-approval trade-off.
IC++ or blended?
At $100K+ a month, almost always IC++: interchange and scheme fees passed through at cost, plus a disclosed acquirer margin. Blended is simpler to read and usually more expensive on a mixed card portfolio, because the blend is set on the acquirer's assumptions, not your mix. The exception is a very uniform low-ticket portfolio where the blend happens to be fair. Either way, compare on all-in cost per transaction on your real mix, not on the headline rate.
How much do 3-D Secure and SCA cost us in conversion?
In Europe and the UK, strong customer authentication is required by regulation, and the exemptions are where the conversion lives: transaction risk analysis exemptions up to €500 depending on the acquirer's fraud rate, low-value exemptions, merchant-initiated and recurring transactions out of scope. An acquirer that challenges everything is cheaper to run and more expensive for you. Ask what share of your transactions were exempted last month, by exemption type. If nobody can answer, that is the answer.
Which local payment methods should we add?
The ones where your checkout drops off by market, not the ones a provider's catalogue lists. Bank-transfer and open-banking methods in parts of Europe, BNPL where customers expect it, wallets where they already hold them, cards nearly everywhere. Each method adds reconciliation, dispute handling and a settlement flow; add them where the data says the card decline rate or the abandonment rate justifies it.
What does a second acquirer give an e-commerce merchant?
Two things. Leverage: pricing and reserve terms move when the incumbent knows another provider already carries part of your volume. And continuity: if the incumbent's risk model changes its mind about your category or your growth, the deposits keep flowing. Built properly, it is a bounded share of traffic routed by market or card type, with one reconciliation, not two checkouts.
We sell in many countries. Is that a different conversation?
Yes, and it has its own guide. Selling across borders through a provider per region costs more and converts worse than local acquiring under one stack; the consolidation we did for a 26-country retailer came out ~20% cheaper. See /payments/cross-border-retail.
We are not high-risk. Is Milestone for us?
Yes. Pricing, approvals and redundancy work the same way in retail as in any other vertical, and our largest consolidated engagement is a sportswear retailer. In 97% of ~200 pricing reviews the offer we brought was cheaper than the incumbent, across verticals.
How is Milestone paid for this?
By the acquirer or PSP you end up using, as a commission on volume. No fee to you for the review. We place with whichever of ~30 providers fits your footprint and mix, so we have no reason to push one.
Cross-border retail
Selling in many countries: local acquiring under one stack.
Marketplaces & platforms
Who is the merchant of record, payouts, and the licensing line.
Subscriptions
Decline taxonomy, retries and account updater.
Send us three months of statements.
Acquirer statements and an approval report by issuer. We tell you within one business day what the stack costs per transaction, where approval is being lost, and whether a second quote would move your terms. No fee for the review.