Merchant of Record vs PSP

If a SaaS founder is aware of MoRs like Paddle, the usual decision process for comparing Paddle and Stripe is simply looking at the flat 5% + $0.50 fee vs Stripe’s lower processing rate. This leads them straight into a trap as they are not looking at the costs of compliance with international tax rules and the subsequent administrative headaches versus the hidden cost of a slowed down cash velocity. The comparison here details that, in the 10-market worked example, the fee-only crossover can range from roughly £0.77m to £2.91m of annual card volume depending on card mix and accounting cost. This matters because most developers of SaaS simply integrate Stripe as a default.

A PSP like Stripe acts as the fixed infrastructure for handling money. When a customer clicks buy now or subscribe, the PSP takes their card details (or bank transfer details) and securely sends encrypted data through the card networks to authorise the funds. Although with a PSP the seller doesn’t have to worry about detailed technical and compliance rules like PCI-DSS data security the seller of the record remains with the SaaS company, meaning you retain full and ultimate liability for VAT, sales tax and GST in all jurisdictions where obligations arise.

Conversely, an MoR like Paddle is a reseller. They become the merchant of record and so take on legal ownership and full GST, sales tax and VAT liability for every transaction. This means if they trigger nexus, the taxman comes knocking on their door, not yours.

Just like how if a customer buys Nike shoes in a Shoe Zone shop they are not buying from Nike, those shoes are sold to Shoe Zone and Shoe Zone sells them to the customer. So Nike only has one transaction and one tax bill for that sale even if Shoe Zone had outlets in multiple countries stocked as part of the deal.

The main difference between the two services is the consistency of the fees. Paddle charges a static fee of 5% + 50 cents on every transaction regardless of what country the buyer is in. By contrast Stripe charges variable fees depending on what country the card used by the buyer is issued in. At a high-ticket price point like £100 to the customer, the standard card processing is incredibly cheap compared to Paddle’s flat 5% rate.

When you add in the additional tools to the Stripe stack (Stripe Payments + Stripe Billing at 0.7% + Stripe Tax Basic at 0.5%) the unit economics are as follows:

For a UK seller selling to domestic UK customers, it is a massive £2.48 saving and you are using your own home tax base, which heavily favours direct integration. For a UK seller selling to the EU, they can achieve a decent saving of £1.48 and, if you only have one or perhaps two EU destinations you sell to and you understand their filing requirements (say Ireland), Stripe is worth considering.

However, there is a compliance trap lying in the international landscape. For a US buyer, the saving is a marginal £0.83 and the seller has to deal with 50 states which have varying and shifting threshold rules. While the California threshold of $757,070 seems generous on paper, there is another trap: if the seller has an employee or contractor there, it will trigger an immediate corporate footprint and nexus. Even if the seller uses an Employer of Record like Remote, if that worker engages in revenue-generating activities like sales or marketing, they create physical nexus for the end client, not just for the EoR. As soon as California payroll or property meets the threshold of $75,707, the company is legally required to file a California return even if they have no California sales. Worse, you can be considered doing business in California below those thresholds if you have an employee conducting business towards profit there.

The EU is a much tighter bottleneck for foreign digital services sold by a company to a consumer. The EU enforces a strict €0 threshold for foreign digital services sold to consumers. If a company makes a single £100 sale to any country in the EU to a consumer, this may include a freelancer or unregistered micro-business that does not have a VAT ID for automatic processing (such as a German Kleinunternehmer operating below the local VAT threshold), or one that may be excluded but require additional documentation. In order to gain a £1.48 per transaction saving on Stripe, the company must now monitor 27 Member States for sales and complete the required VAT registrations and returns in each Member State in which they do business, or submit a quarterly Non-Union OSS scheme return.

When should a company cross over from an MoR to a PSP? This depends entirely on the operating methodology of the company. The missing software cost can now be made explicit. Anrok’s current Starter pricing for software companies is US$100 per market per month, with each US state counted as a market and EU OSS counted as one market. In a 10-market worked example that is US$12,000 a year, or about £9,040 at the 30 September 2026 exchange rate above. Adding the existing £10,000-£15,000 annual accounting and filing assumption gives fixed compliance costs of about £19,040-£24,040 a year. Against the corrected per-£100 Stripe savings above, the fee-only crossover is about £0.77m-£0.97m of annual card volume for UK cards, £1.29m-£1.63m for EEA cards, and £2.30m-£2.91m for international cards. If substantially all recurring revenue is billed by card, annual card volume is approximately ARR; otherwise card volume is the more accurate denominator.

However, there is a major operational catch lurking in the timeline. After the initial validation period Stripe’s default European payout speed is generally three business days subject to eligibility and risk, and daily automatic payouts are supported. By contrast, Paddle moves the previous month’s balance into payout on the 1st and sends it by the 15th of the following month. If Paddle sends the payment on the 15th, sales made evenly through a 30-day month wait an average of 29.5 days before the payout is sent. Both Stripe and Paddle can retain funds where risk requires it, but neither publishes a standard reserve rate that can defensibly be inserted into this example, so reserves are excluded from the worked comparison.

On a timing-only model with £100,000 of monthly customer spend, the average gross cash flow awaiting payout is about £98,333 under Paddle if payout is sent on the 15th (£100,000 × 29.5/30). For Stripe on daily automatic payouts, T+3 working days is approximately four calendar days across a seven-day sales week, giving about £13,333 pending (£100,000 × 4/30). These are illustrative timing balances before fees, tax, refunds, reserves and bank posting delays. The point is the order-of-magnitude difference in cash velocity, not a guaranteed account balance.

This realisation changes the calculus completely for a growth-minded SaaS. If the company is actively scaling paid marketing channels or expanding an engineering team where a quick Customer Acquisition Cost payback period is needed, then getting early access to that cash is an absolute game changer. The massive liquidity advantage of being able to invest capital into compound user growth completely outweighs the cost of paying an accountant and software fees to manage international tax portals. If the startup is targeting a single core market where they are confident filing (and especially if they are avoiding the complex multi-state US market), then it makes financial sense to deploy the Stripe architecture much earlier.

The exact fee-only crossover point in the 10-market worked example depends on card mix: £0.77m-£0.97m for UK cards, £1.29m-£1.63m for EEA cards, and £2.30m-£2.91m for international cards.

The Low Growth Bootstrapped Track. For this style of company optimising for fewer administrative headaches, the fee-only crossover is the relevant benchmark.

The High Growth Track. For this style of company optimising for capital deployment and cash flow velocity and scaling via paid channels or rapid team expansion, the cash-flow advantage can justify switching before the fee-only crossover.