THE GUIDE · THE BUILDER SIDE · 12 MIN

The other half of the swipe.

Issuing hands out cards; acquiring lets a business accept them. This is the getting-paid side: onboarding and underwriting a merchant, moving the money in, and carrying the risk that a shop disappears before its refunds clear.

SCROLL ↓
IN PLAIN WORDS — READ THIS FIRST

Accepting cards means someone — the acquirer — pays you now for sales that can be un-paid later: a customer can dispute a charge for up to 120 days. So the acquirer treats every merchant like a small loan. It checks the business before boarding it (underwriting), holds back a slice of payouts (a reserve), and can pull money back out (a clawback).

Read the flow below with one question in mind: if this shop vanished tomorrow, who would be left holding its refunds? The answer explains every rule on this page.

PART 01

How a business gets paid — and who carries the risk.

Step through onboarding, settlement, a chargeback, and the exact reason acquirers hold reserves.

PART 02

The acquiring side, up close.

The mirror image of issuing — who onboards the merchant, moves the money, and eats the risk.

THE ACQUIRING SIDE

"Issuing, in a mirror."

Acquiring is the merchant's side of the four-party model: the acquirer signs up sellers, connects them to the networks, moves the settlement money in, and carries the merchant risk. Everything issuing does for cardholders, acquiring does for shops.

MERCHANT UNDERWRITING

"Underwrite the seller, not the buyer."

Before boarding a merchant: KYB (Know Your Business), beneficial ownership, credit and fraud checks, and a read on the business model. High-risk verticals (travel, supplements, adult, crypto, anything with delivery lag) get reserves or declines — because the risk is the merchant failing to deliver.

MoR vs PayFac vs ISO

"Who is legally the merchant?"

Three models. A Merchant of Record takes on the sale (and its tax and compliance) as principal. A PayFac aggregates many sub-merchants under its own master account. An ISO (Independent Sales Organization) just refers merchants to an acquirer. The difference is who holds the licence, the funds and the liability.

THE MERCHANT DISCOUNT

"Three fees in a trench coat."

What a merchant 'pays to accept cards' is three stacked layers: interchange (to the issuer), scheme fees (to the network), and the acquirer's markup. Pricing can be transparent (interchange-plus) or opaque (blended or tiered) — the receipt below unstacks it.

RESERVES & CLAWBACKS

"Money held against tomorrow's refunds."

A rolling reserve holds back a slice of each payout to cover future chargebacks. When a dispute lands, the acquirer claws it from the merchant's balance. Reserves are the acquirer's insurance against a merchant that takes money and can't — or won't — make good.

MARKETPLACES & SPLIT

"One checkout, many sellers."

Ride-shares and marketplaces must pay many sub-merchants from one transaction. Split payments and payout APIs (the Stripe Connect model) handle the fan-out, the sub-merchant KYB, and the tax reporting — turning the platform itself into a mini-acquirer.

The words, one at a time.

Six terms every merchant meets in the first year of taking cards — usually the hard way.

KYB
the background check on a business
"Know Your Business": verifying who owns the company, what it actually sells, and how risky the model is — before it can accept a single card.
A café boards in a day. A travel agency selling tickets months ahead gets extra questions and a reserve.
Why it matters: underwriting the seller is the acquirer's core risk decision, and it never really ends.
Rolling reserve
payouts, partly held back
A fixed share of each payout — often 5–10% — held for a period (commonly 90–180 days) as a cushion against future chargebacks.
A $10,000 payout arrives as $9,500 now. The $500 follows months later, if no disputes eat it.
Why it matters: the reserve is the acquirer's insurance against a merchant that vanishes.
Clawback
money pulled back out
When a dispute lands, the acquirer debits the merchant's balance or next payout for the disputed amount, plus a fee.
A $100 chargeback arrives; the next payout is short about $125.
Why it matters: chargebacks are the merchant's liability. The acquirer is just the one collecting.
Payout hold
the settlement that didn't come
A temporary pause on payouts while risk reviews something the model flagged: a volume spike, a dispute jump, a product change.
A viral day sends sales to 10× normal. Payouts pause pending review, exactly when the shop needs cash most.
Why it matters: from the acquirer's chair, sudden success looks identical to bust-out fraud.
MATCH list
the industry's shared blacklist
A database of terminated merchants, run by Mastercard and checked across the industry during onboarding. Entries persist for years.
Terminated for excessive chargebacks at one acquirer; every new application elsewhere now starts with a red flag.
Why it matters: termination follows you. Preventing it is far cheaper than recovering from it.
Smart retries
failed payments, retried on a schedule
Recovering soft declines (like insufficient funds) by retrying at smarter times — and never retrying hard declines on dead cards.
A code-51 renewal retried on the 1st, after payday, often recovers. A code-43 stolen card retried daily draws network fines.
Why it matters: retry logic is revenue and compliance at the same time. The decline decoder maps the codes.
PART 03

Unstacking the merchant discount.

The same $100 sale, written the way the acquirer's statement records it.

MERCHANT STATEMENT
ONE $100 CARD SALE · REWARDS CREDIT
Gross sale$100.00
− Interchange (to issuer)−$1.80
− Scheme / assessment (to network)−$0.13
− Acquirer markup−$0.30
Net to merchant$97.77
effective rate 2.23%
Illustrative US rewards-credit sale, interchange-plus pricing. Regulated debit would net ~$99.50; a blended or tiered plan hides these three layers behind one number.
WHEN IT BREAKS

When getting paid stops.

The acquiring relationship fails in slow motion, and the warning signs are usually on the merchant's own dashboard first. Three real failure patterns, then a tree for the morning the payout doesn't arrive.

FAILURE 01 · THE PAYOUT HOLD
Sales normal, settlement missing
WHAT YOU SEECards approve all day, customers leave happy — and the payout that funds tomorrow's inventory says "under review."
WHYSomething moved the risk model: a volume spike, a jump in disputes, or a product drift into a riskier category. Underwriting is continuous. The approval you got at boarding was for the business you described then.
THE FIXTell your acquirer before big promotions or launches, watch your dispute ratio like a vital sign, and read your reserve terms before you need them. Holds resolve fastest when the paperwork is ready.
FAILURE 02 · THE RETRY STORM
Your own billing code attacks you
WHAT YOU SEESubscription approval rates sink month after month. Then a letter arrives: fines for excessive retries.
WHYNaive billing logic retries dead cards — stolen, closed, lost — every day forever. The networks fine repeated attempts on hard declines, and issuer risk models start scoring the whole merchant ID down, so even good customers' auths suffer.
THE FIXClassify every decline before retrying: hard codes stop, soft codes get a capped, payday-aware schedule. Enroll in account updater or network tokens so vaulted cards heal themselves when banks reissue them.
FAILURE 03 · TERMINATED, THEN MATCHED
Fired by one, refused by all
WHAT YOU SEEYour account is closed for excessive disputes. Every new acquirer you apply to declines you at the underwriting stage.
WHYTermination landed you on the MATCH list, which the industry checks at every onboarding for roughly five years. The killer was never one bad month. It was ignoring the ratio until the acquirer acted.
THE FIXTreat the network's ~1.5% dispute-ratio line as a red line with a wide margin. Use prevention alerts, fix the root causes early, and if your acquirer warns you, answer with a remediation plan — they'd rather keep a fixed merchant than fire one.
YOUR PAYOUT DIDN'T ARRIVE. WHY?
1 · Did sales volume or ticket size spike recently?
RISK REVIEWSpikes trigger holds because sudden growth mimics bust-out fraud. Send invoices, delivery proof, and an explanation proactively — reviews with documents close in days, reviews without them drag.
VOLUME IS NORMAL — KEEP GOINGGo to step 2.
2 · Is your dispute ratio climbing toward 1%?
RESERVE OR PROGRAM ACTIONExpect a raised reserve or a monitoring-program letter. The fix is the ratio itself: descriptors, refund speed, delivery proof — see how disputes work. Money held this way returns only when the ratio does.
RATIO IS HEALTHY — KEEP GOINGGo to step 3.
3 · Are other merchants on your processor reporting the same thing?
PLUMBING, NOT YOUA processor payout delay, a settlement-file failure, or a bank-holiday calendar quirk. Check the status page and the settlement calendar before assuming the worst.
JUST YOU, NO FLAG YOU CAN SEERead the hold notice precisely and call your acquirer. Holds name their reason, and the reason names the fix.
COMMON QUESTIONS — ASKED PLAINLY

The things merchants actually ask.

Five questions from the first year of taking cards.

WHY DOES MY MONEY TAKE TWO DAYS TO ARRIVE?
Because settlement runs in batches through several hands: the networks clear overnight, the acquirer receives funds, applies its fees and risk checks, and pays out — typically T+1 or T+2. The "instant payout" button some providers offer isn't faster settlement; it's the acquirer advancing its own money early for a fee, then keeping your actual settlement when it lands. Useful in a cash crunch, expensive as a habit.
CAN I PASS THE CARD FEE ON TO MY CUSTOMERS?
It depends where you are. Surcharging is legal in much of the US (with notice requirements and caps), regulated or banned in others — and the 2025 network settlement, if approved, expands US merchants' rights to surcharge and steer. Cash discounts — posting a lower price for cash — are allowed almost everywhere and feel better to customers than a fee. Whatever you do, check your state or country rules and your card-acceptance agreement first; the rules genuinely differ by geography.
I WAS APPROVED — WHY WAS I FROZEN ANYWAY?
Because approval was a snapshot and underwriting is a movie. The acquirer approved the business you described on the application: your products, your volume, your refund pattern. When reality drifts — new product line, sudden spike, disputes creeping up — the risk model re-evaluates you against the new facts. The freeze notice names a trigger; that trigger is your to-do list. Merchants who keep their acquirer informed of changes mostly never meet the freeze at all.
WHAT MAKES A BUSINESS "HIGH-RISK"?
Mostly one variable: the gap between when the customer pays and when they get what they paid for. Travel, event tickets, furniture, custom goods — anything with a long delivery lag can generate months of refund liability if the seller fails. Add categories with high dispute rates (subscriptions with free trials, supplements), heavy regulation (gambling, adult, crypto), and fraud magnets. High-risk doesn't mean unservable; it means reserves, higher pricing, and specialist acquirers who price that risk for a living.
DO I REALLY NEED ALL THESE MIDDLEMEN?
At small scale, yes — and they're cheaper than the alternative. Getting your own merchant account means passing bank underwriting, maintaining PCI compliance, and integrating with an acquirer directly; a PayFac absorbs all of that and boards you in minutes, which is exactly what you're paying the blended rate for. The math flips at scale: once volume is large enough, direct acquiring with interchange-plus pricing saves real money. Most businesses graduate from aggregator to direct acquiring somewhere in the millions per year — the calculator on this site shows where the lines cross.
FIELD NOTES — THE PRO LAYER

For the professionals.

The acquiring side up close — the three merchant models, the vanishing-merchant risk, pricing, marketplaces, and the 2025 interchange settlement.

MoR vs PayFac vs ISO — THE LIABILITY DECOMPOSITION
The three models differ in who holds what. A Merchant of Record (many SaaS and marketplace 'seller of record' setups) becomes principal in the sale — owning sales-tax/VAT, refunds and compliance, and letting the real seller ignore all of it. A PayFac holds a master merchant account and onboards sub-merchants under it, taking on their underwriting and funding (Square, Stripe, Toast). An ISO merely resells an acquirer's service and refers merchants, holding little risk. Choosing among them is really choosing how much licence, funds-flow and liability to carry.
UNDERWRITING THE 'MERCHANT VANISHES' RISK
Card acquiring's defining risk isn't the buyer. It's the seller failing to deliver. A travel agency or furniture shop takes payment now for goods weeks away; if it folds, every customer files a chargeback and the acquirer, not the vanished merchant, owes the networks. So acquirers underwrite the business (KYB, financials, delivery lag, refund history), price high-risk verticals higher, and hold reserves. It's the same discipline the Underwriting Desk game makes you feel — approve growth without boarding a time bomb.
PRICING — INTERCHANGE-PLUS vs BLENDED vs TIERED
Interchange-plus passes through the real interchange + scheme fee and adds a transparent markup — the honest, auditable model. Blended charges one flat rate regardless of card type (simple, but the processor keeps the spread on cheap cards). Tiered ('qualified / mid / non-qualified') is the murkiest — the processor decides which bucket each transaction lands in. Add least-cost routing on debit (a Durbin gift) and 'what do you pay to accept cards' becomes a genuinely hard question — exactly what the fee calculator and benchmark tools exist to answer.
MARKETPLACES, SPLIT PAYMENTS & PAYOUTS
When one checkout must pay many sellers, the platform needs split payments and payouts. The Stripe Connect model lets a platform onboard sub-merchants (with their KYB), take its cut, disburse the rest, and handle 1099-K/tax reporting. Economically the platform becomes a mini-PayFac, which is why 'payfac-as-a-service' exists — the acquiring stack, rented, so a software company can embed payments without becoming a payments company.
THE 2025 US INTERCHANGE SETTLEMENT
After roughly 20 years of litigation, Visa and Mastercard announced a ~$200B settlement with US merchants on 10 Nov 2025: modest posted-rate cuts, a temporary cap on standard consumer-credit interchange, and — the bigger deal — expanded rights to surcharge and to steer customers toward cheaper cards. It is not yet court-approved (final approval expected in 2026, and merchant groups are split), so treat the numbers as provisional. If it holds, it nudges the merchant discount down and hands merchants new routing and surcharging leverage — reshaping the economics this whole chapter describes.
PART 04

Remember three things.

1
Issuing and acquiring are mirror images — one side hands out cards, the other lets businesses accept them and moves the money in.
2
The acquirer's real risk is a merchant that vanishes before its chargebacks clear — which is why underwriting, rolling reserves and high-risk pricing exist at all.
3
The merchant discount is three stacked fees — interchange, scheme, markup — and the choice of MoR vs PayFac vs ISO decides who owns the licence, the funds and the tax.
← PREVIOUS
How anyone launches a card