Paddle Review

A London merchant of record for software, SaaS and AI companies, founded in 2012 and used by more than 6,000 digital product businesses by its own July 2025 count. It publishes one flat price — 5% + 50¢ — and in exchange becomes the legal seller of your product, taking on global sales tax and VAT registration, filing and liability. That is a real transfer of risk, and it is expensive: roughly double a card processor's rate, paid out once a month rather than daily.

Estimated annual cost at three realistic processing volumes, using Paddle’s published online rate plus monthly fees. Real costs vary with average transaction size, chargeback rate, and any negotiated terms.

Not a payment processor

Paddle is easiest to understand by what it is not. Stripe, Braintree and Adyen move your money and leave you holding every other obligation the sale creates — most importantly the tax one. Paddle instead becomes the legal seller of your product. Your customer buys from Paddle; Paddle issues the invoice in its own name, collects the money, and registers, files and remits the sales tax, VAT or GST that the sale triggers, carrying the liability if any of it is wrong.

For a four-person software company selling to customers in thirty countries, that is not a convenience feature. Digital services VAT rules mean a company with no office, no staff and no bank account in a country can still owe that country a return, and the penalty for not knowing lands on the seller. Paddle takes that problem off the table entirely. That is what the price is for.

The price, and everything just outside it

Paddle publishes one number: 5% + 50¢ per checkout transaction, no monthly fee, no migration fee, no setup cost. It is refreshingly legible next to the industry norm, and it includes tax compliance, subscription billing, churn prevention tooling from the ProfitWell business Paddle bought for $200 million in 2022, fraud protection and support. Products under $10 and merchants needing invoicing are directed to custom pricing, and larger merchants are reported to negotiate below 5%.

The costs sitting just outside that headline are all documented in Paddle's help centre, and you should add them up before comparing. A payout taken in a currency other than your balance currency can carry a conversion margin of up to 1.5%. SWIFT wires to certain countries carry a $/£/€15 fee, though where your payout currency matches your bank's local currency the transfer is typically free. A card chargeback costs $15 and a PayPal chargeback $20; and Paddle's help centre states the same fee applies when a pre-chargeback alert is received and it refunds proactively. Against that, Paddle returns both the transaction amount and the chargeback fee to your balance when it wins a dispute — even where the fee is not refunded to Paddle — which is more generous than most.

You get paid on the 15th

This deserves its own heading because it catches people out. Paddle does not settle daily and there is no on-demand withdrawal. It processes payouts on the 1st of each month, sends them by the 15th, and the money takes up to three working days after that to land, by wire or Payoneer. You set a payout threshold anywhere between $100 and $100,000; anything below it rolls into the following month.

For a subscription business with predictable revenue and a cash buffer this is a non-event. For a company used to next-day settlement, or one whose ad spend is funded by yesterday's sales, it is a structural change to working capital that has to be modelled before migration, not discovered afterwards.

Where it stops paying for itself

The honest way to evaluate Paddle is as arbitrage: you are paying roughly two extra percentage points to avoid a compliance workload. At $50,000 a month that premium is about $1,000 — less than a part-time bookkeeper, never mind multi-jurisdiction VAT filings and the risk of getting one wrong. At $500,000 a month it is about $10,000, which buys a finance hire, a tax automation platform and an accountant, with change. Somewhere between those two the arithmetic flips, and it flips sooner for a low-margin business than a high-margin one.

Paddle knows this, which is why custom rates below 5% are available at volume. If you are approaching the crossover point, the productive conversation is about the rate rather than about leaving — because leaving is the expensive part.

The exit is the real lock-in

There is no contract term, no monthly minimum and no exit fee. But the merchant-of-record model creates its own gravity. The tax registrations covering your sales are Paddle's. The invoices your customers hold say Paddle. The stored card credentials behind your live subscriptions sit in Paddle's vault, and moving them is a card-network process rather than a database export. Migrating off means registering for VAT and sales tax yourself in every jurisdiction where you have nexus, moving the subscriptions, and in some cases asking customers to re-authorise payment.

None of that is a reason not to use Paddle. It is a reason to know, on the day you sign up, roughly how you would leave.

What the reviews are actually saying

Paddle's Trustpilot page reads 3.9 out of 5 across 11,048 reviews with a hollow middle — 73% at five stars, 21% at one, almost nothing between. The shape is explained by there being two populations on one page. Most of the one-star reviews are from consumers who bought somebody else's software, found an unfamiliar "Paddle" charge on their statement, and could not work out how to cancel it. That is a direct consequence of the merchant-of-record model rather than a verdict on Paddle's service, and every MoR has the same problem.

The seller-side complaints are rarer and more worth your attention: accounts blocked with payouts frozen and no explanation given, and account or domain validation checks left in progress for over a month with support unresponsive. That is the same failure mode that dogs every aggregator — Paddle underwrites you into its own merchant account, and when its risk team pauses an account the money stops with it. It is the risk you accept in exchange for not carrying a chargeback ratio of your own. Keep a second route to taking payment configured, and keep enough runway that a frozen month is an inconvenience rather than an emergency.