Choosing a Payment Gateway for a New E-commerce Store

Choosing a payment gateway feels like a pricing decision. It is not. It is a risk decision, and the store owners who treat it as pricing are the ones who wake up to a frozen balance three weeks before Christmas.

Here is how to choose, in the order that actually matters.

First, work out what risk band you are in

Processors classify merchants long before they tell you they have. You are treated as higher risk if any of these apply:

• Long delivery times because goods ship from overseas.
• New business with no processing history.
• Subscription billing or free-trial-to-paid conversion.
• Health, supplements, CBD, adult, weapons, gambling, or ticketing.
• Average order value above about 200 and physical delivery.
• High expected refund or dispute volume in your category.

Two or more of these and mainstream aggregators will onboard you instantly and then review you at exactly the moment volume spikes. Knowing your band up front changes which providers you should even apply to.

The three types of provider

Aggregators (Stripe, PayPal, Square, Mollie). You share a master merchant account. Onboarding takes minutes, pricing is transparent, developer experience is excellent. The trade-off is that they manage risk by acting fast and asking questions later — rolling reserves, holds, or termination with limited negotiation.

Dedicated merchant accounts (via an acquiring bank or ISO). You get your own MID. Onboarding takes one to three weeks and requires financials, but you get a named risk contact, negotiated terms, and far less chance of a sudden freeze. Worth it above roughly 25–50k monthly volume, or immediately if you are in a flagged category.

High-risk specialists. Higher rates (often 3.5–6% plus a rolling reserve), but they underwrite your model deliberately and do not panic. If you are genuinely high-risk, this is cheaper than being shut down.

Read the cost structure properly

Headline rate is the smallest part of the real cost.

Cost item	What to ask
Transaction rate	Domestic vs international vs commercial cards
Fixed per-transaction fee	Matters enormously on low AOV
Currency conversion	Typically 1–2% on top, often overlooked
Payout schedule	T+2, T+7, or monthly?
Rolling reserve	Percentage held and for how many days
Chargeback fee	15–40 per dispute, win or lose
Refund handling	Do you get the original fee back? Usually not
Monthly / gateway / PCI fees	Common with dedicated accounts

Model it on your real numbers. On a 25 average order with 60% international cards, a 2.9% + 0.30 aggregator and a 2.4% + 0.25 dedicated account with 1.5% FX can land within a few basis points of each other — and then the payout schedule becomes the deciding factor because cash flow, not fees, is what kills young stores.

Coverage is a conversion feature

Every market has a payment method that meaningfully outperforms cards:

• Germany and the Netherlands: SEPA direct debit, iDEAL, invoice-after-delivery.
• Nordics: Klarna and local wallets.
• Brazil: Pix and instalments (parcelamento) — offering instalments can double conversion.
• India: UPI.
• Southeast Asia: GrabPay, GCash, local bank transfer.
• Middle East: cash on delivery expectations in several markets.
• Everywhere: Apple Pay and Google Pay, which reliably lift mobile conversion by shortening checkout.

Adding the correct local method usually beats any optimisation you can make to the checkout page design.

Always run two providers

This is the single most valuable operational habit in this article.

Integrate a primary and a secondary provider from day one, even if the secondary handles only 5% of volume. Keep it live and processing real transactions so the account stays in good standing. If the primary freezes you, you flip a setting and keep trading while you resolve it. Stores that only discover this problem after a freeze spend two to three weeks offline.

For redundancy to work: keep both accounts warm, keep your product catalogue and policies identical across both, and make sure your order system can reconcile payments from either source.

Reduce disputes before they happen

Processors judge you on dispute ratio. Above roughly 0.65–0.9% you enter a monitoring programme; above 1% you are usually terminated.

Practical measures that work:

• Billing descriptor that matches the store name shoppers remember, plus a contact number in the descriptor field.
• Honest delivery estimates with proactive updates when something slips. "Item not received" is the largest dispute category for overseas fulfilment.
• Tracking that updates in the destination country, not only at origin.
• Fast, generous refunds. A refund costs a fee. A chargeback costs a fee, the goods, and your ratio.
• 3D Secure / SCA on higher-value or higher-risk orders, which shifts liability for fraud disputes.
• A visible, answered support channel. Most disputes start as an unanswered email.

Application checklist

Whatever provider you approach, having these ready shortens underwriting from weeks to days:

1. Registered company details and director identification.
2. Business bank account in the company name.
3. Live website with all policy pages and a working test checkout.
4. Realistic monthly volume and average order value projections — do not inflate them.
5. Supplier and fulfilment description in plain language, including delivery timeframes.
6. Prior processing statements if you have any.

Underwriters do not expect you to be big. They expect you to be consistent. The fastest approvals come from applications where the website, the policies, the projected volume, and the stated fulfilment model all describe the same business.

A sensible default setup

For a new store in a normal category: Stripe as primary, PayPal as secondary, Apple Pay and Google Pay enabled, plus the dominant local method in your top market. Review at 25k monthly volume and start a dedicated merchant account application then — not after the first hold.