Stripe vs PayPal vs Shopify Payments, and When to Add BNPL

I once spent an afternoon convinced I'd found a 0.6% pricing edge by switching processors, only to realize the "cheaper" option was about to cost me a 2% platform penalty on every order. The headline rate is the part everyone reads. The fine print is the part that actually empties your bank account. So let me save you that afternoon. I've set up checkout on more stores than I'd like to admit, and the processor question always comes down to the same three contenders plus one newer decision: do you bolt on Buy Now, Pay Later? Here's how I think about all of it.

Start with the problem, not the logo

Your processor does three jobs: take the money, pass it through cleanly, and put it in your account fast. Where founders get burned is treating those as one number. They're not. A processor can have a great rate and a brutal payout delay, or a fair rate that becomes terrible the second your platform adds its own cut. That last bit is the trap. If you're on Shopify and you use anything other than Shopify Payments, Shopify charges an extra transaction fee on top of your processor's fee. On the Basic plan that surcharge is up to 2% per order. So "Stripe is 2.9% and Shopify Payments is 2.9%, they're the same" is wrong on Shopify. With the third-party fee stacked on, Stripe can effectively cost you closer to 4.9% there. Context changes the math completely.

The processors, compared

These are published US rates. Real rates vary by volume, plan, and negotiation, so treat them as the starting line.

The pattern I'd pull out: on Shopify, Shopify Payments almost always wins on total cost because it dodges the third-party surcharge. Off Shopify, it's not an option, and the real fight is Stripe vs PayPal, where Stripe usually wins on rate and developer control, PayPal wins on shopper familiarity. Plenty of stores run PayPal alongside a primary processor rather than instead of one, because some buyers will only check out with PayPal.

When BNPL is worth it (and when it isn't)

Buy Now, Pay Later (Klarna, Afterpay, Affirm) is a different animal. You're not replacing your processor, you're adding a financing option that the BNPL provider pays out in full and collects from the customer in installments. They charge you for that, and it's not cheap. Merchant fees run roughly Klarna ~3.29–5.99% + $0.30 (most US merchants near the top of that band), Afterpay ~6% + $0.30, and Affirm ~3–6% + a flat fee, all per transaction and often negotiable on volume. That's roughly double a card rate. So the only reason to accept that hit is if BNPL grows the order enough to pay for itself. Does it? The vendor-reported numbers are loud, so label them as marketing estimates, not independent proof:

• Affirm claims partners saw up to a 70% lift in average cart size, and reports merchants citing an ~85% AOV increase when shoppers use Affirm vs other methods (Affirm, vendor-reported, estimate).
• RBC Capital Markets (analyst estimate) pegs BNPL at +20–30% conversion and +30–50% average ticket size (estimate).
• A Shopify-cited figure puts BNPL at +27% conversion and +21% order value for merchants (vendor-reported, estimate). Even if the true lift is half what the vendors claim, the math can still work for higher-ticket goods. Here's the back-of-napkin test I use: if BNPL costs you ~3 extra points versus a card, you only need a modest, real AOV lift on BNPL orders to come out ahead. It tends to pay off for furniture, electronics, fitness equipment, anything where the price tag causes hesitation. It rarely pays off for $25 impulse buys, where you're just renting out margin for orders that would've converted anyway.

How I'd decide

Processor first. On Shopify, default to Shopify Payments unless you have a specific reason (subscriptions, custom flows) to bring Stripe. Off Shopify, default to Stripe and add PayPal as a secondary button for the buyers who insist on it. Watch payout timing if cash flow is tight, since a 2-day Stripe rolling payout behaves very differently from a holdback or reserve. BNPL second, and only after you know your AOV and category. High-ticket and hesitation-prone? Pilot one provider, tag those orders, and measure the actual AOV delta against your non-BNPL baseline for 60–90 days. If the lift covers the fee, keep it. If it doesn't, you just gave away margin.

The takeaway

The cheapest processor on paper is rarely the cheapest in your context, because platform surcharges and payout timing quietly rewrite the bill. And BNPL is a margin trade, not free money: worth it when it genuinely grows the basket, wasteful when it just discounts orders you'd have won anyway. So, real question for you: do you actually know your blended effective rate across every order, including platform fees and disputes? Or are you still quoting the headline number off the homepage?

Sources

• Stripe — Pricing (2.9% + $0.30, US). https://stripe.com/pricing
• Stripe — Payout schedules FAQ (2-day rolling US; Instant Payouts). https://support.stripe.com/questions/payout-schedules-faq
• PayPal — Merchant fees (Checkout 3.49% + $0.49). https://www.paypal.com/us/business/paypal-business-fees
• Shopify — Payments & transaction fees. https://www.shopify.com/pricing
• Klarna — How are fees calculated. https://www.klarna.com/us/business/merchant-support/how-are-fees-calculated/
• Afterpay / Affirm / Klarna merchant fee comparison. https://www.chargeblast.com/blog/affirm-vs-afterpay-klarna-which-has-lower-merchant-fees
• Affirm — AOV lift claims (vendor-reported). https://investors.affirm.com/news-releases
• RBC Capital Markets BNPL conversion/ticket estimates (via Shopify). https://www.shopify.com/blog/buy-now-pay-later