Payment Service Provider (Stripe) vs. Merchant of Record (Paddle): The SaaS Founder's Guide

When scaling a SaaS product internationally, founders (including me) usually hit a sharp operational wall around payments, cross-border tax compliance, and local merchant entities.

The choice isn't just about picking Stripe vs. Paddle. It's an architectural and operational decision between two fundamentally different financial models: Payment Service Providers (PSPs) and Merchants of Record (MoRs).

1. Core Structural Difference

The primary distinction comes down to legal ownership of the transaction and where financial liability sits.

PSP Flow

flowchart LR
    C["Customer"] --> P["Payment Processing Tech (PSP)"]
    P --> E["Your Legal Entity
(Seller on Record)"] E --> T["You handle Tax, VAT, Disputes"]

MoR Flow

flowchart LR
    C["Customer"] --> M["Merchant of Record
(Legal Reseller / Seller)"] M --> R["Handles Tax, VAT, Chargebacks, Risk"] M --> Y["Payout to Your Entity"]

2. PSP vs. MoR Comparison

Feature / ResponsibilityPayment Service Provider (PSP)Merchant of Record (MoR)
Legal SellerYour CompanyThe MoR Entity (e.g., Paddle, Lemon Squeezy, FastSpring)
Headline RatesLower upfront (~2.9% + $0.30 in the US, ~1.5% + €0.25 for EEA cards on an EU entity)Higher headline fee (~5% + 50¢ or custom)
Global Indirect Tax (VAT/GST)You calculate, register, file, and remitTransaction tax offloaded to the MoR (your own corporate tax and accounting remain)
Chargebacks & Fraud RiskUltimate dispute liability sits with youMoR manages transaction-level risk and the chargeback workflow, subject to its terms and acceptable-use policies
Payout MechanicsDirect daily/rolling payoutsProvider-dependent and generally less flexible; often batched on a fixed monthly cycle
Merchant Identity on Card StatementYour Brand NameTypically MoR * Your Brand Name, depending on descriptor configuration
Payout & Entity OverheadHigh if opening multi-region bank accountsLow (one payout channel worldwide)

3. Deep-Dive Tradeoffs

Option A: Payment Service Provider (Stripe, Adyen, Braintree)

The Build-Your-Own-Stack Approach

Pros:

Cons & Hidden Costs:

Option B: Merchant of Record (Paddle, Lemon Squeezy, FastSpring, Polar)

The Offloaded Compliance Engine

Pros:

Cons:

4. How to Choose: Decision Framework

flowchart TD
    Q1{"Where are your customers?"}
    Q1 -->|"Single Country / US-only"| Q2{"B2B or B2C SaaS?"}
    Q1 -->|"Global / Multi-Region"| Q3{"B2B or B2C SaaS?"}
    Q2 -->|"B2B"| A1["PSP + Tax Tool"]
    Q2 -->|"B2C"| A2["PSP or MoR"]
    Q3 -->|"Mostly B2C"| A3["Pure MoR"]
    Q3 -->|"Enterprise B2B"| A4["Hybrid / PSP + Reverse Charge"]
      

Choose an MoR if:

  1. You sell B2C software or self-serve B2B globally: Managing micro-registrations for VAT/GST in 30+ countries manually will destroy engineering and accounting velocity.
  2. You have a lean engineering/finance team: You prefer paying a higher percentage to avoid hiring external tax advisors or building dedicated billing infrastructure.
  3. Speed to market matters: You need to start taking cross-border subscription payments instantly.

Choose a PSP if:

  1. You are purely domestic or operate in a single currency union (e.g., US-only or localized EU).
  2. Your customer base is overwhelmingly B2B with valid tax IDs: A predominantly B2B base materially reduces indirect-tax collection complexity where valid VAT IDs and reverse-charge rules apply. It does not eliminate global tax compliance: you still deal with customer location evidence, VAT ID validation, invoicing requirements, domestic B2B sales, US sales tax, and GST regimes in Canada, Australia and elsewhere. The EU rules are also moving, with ViDA (VAT in the Digital Age) changes phasing in from 2027.
  3. You process high volume ($10M+ ARR): The gap between a ~3% PSP cost and a ~5% MoR fee represents substantial capital that easily funds dedicated in-house tax ops and tool integrations.

5. The European Case: Running a PSP Setup from an EU Entity

If you go the PSP route from an EU company (my case), the compliance work you take on is concrete and well-defined. Here is what it looks like in practice.

flowchart TD
    S["Sale"] --> W{"Where is the buyer?"}
    W -->|"EU"| B{"Valid VAT ID?"}
    W -->|"US"| US["No VAT
Assess state sales-tax nexus"] W -->|"Non-EU"| ROW["Determine local indirect-tax rules
(GST/VAT thresholds vary)"] B -->|"Yes (VIES validated)"| RC["No VAT charged by you
Invoice states reverse charge applies"] B -->|"No (consumer)"| T{"Cross-border EU B2C
over EUR 10,000/year?"} T -->|"No"| SK["Charge your home country VAT rate
File domestically as usual"] T -->|"Yes"| OSS["Charge buyer's local VAT rate
File one quarterly Union OSS return"]

How it works under EU tax law:

The part that changes the math: your EEA (European Economic Area) rate does not apply to US cards. On an EU Stripe account a non-EEA card is currently 3.15% + €0.25, with a further 2% where currency conversion is required. Charge in dollars and settle in euros and you are at roughly 5.15% - and note what that number is: pure payment processing, before you have paid for a tax engine, nexus tracking, fraud tooling or dispute handling. Paddle's 5% + 50¢ is its all-in price including all of those. When raw processing alone reaches the MoR's total fee, the PSP margin advantage on that revenue is not thinner, it is gone. So the advantage is real on EEA cards and absent on converted US card volume, which happens to be the same revenue that carries the nexus tracking burden. If you sell mostly into the US from an EU entity, run this calculation with your own card mix before assuming PSP is the cheaper architecture.

Best for: Pure B2B SaaS, where reverse charge already does most of the work, and SaaS whose revenue is concentrated in EEA cards, where the retained margin comfortably covers compliance software and an accountant who knows cross-border VAT. Past roughly $1M ARR that trade is usually worth making; building genuine in-house tax operations rather than buying tools and an advisor is a later problem, closer to the $10M mark in the framework above.

6. It Does Not Have to Be One or the Other

The PSP vs. MoR framing assumes you pick one and live with its tradeoffs everywhere. That assumption is starting to break.

Portable billing layers like Clink, and modern payment routing setups more generally, let a SaaS product run both at once: direct PSP connections for the high-volume home market where you already have the tax registrations and want the margin, with international long-tail sales routed automatically through an MoR that absorbs the compliance work.

The strongest evidence that this is where things are heading is that Stripe now sells the MoR model itself. Managed Payments is Stripe's merchant-of-record product, and it operates at the transaction level: you can apply it to specific markets or products while the rest of your traffic stays on your normal Stripe setup. The either/or is dissolving from inside the incumbent PSP.

On price, do not mistake it for an undercut: the Managed Payments fee is 3.5% on top of standard Stripe processing fees, which on an EEA card lands around 5% + €0.25 - roughly Paddle's all-in rate. What Stripe is selling is not a cheaper MoR but a composable one: MoR economics only on the transactions you choose, native to an integration you already run.

So the useful question is not "which provider is cheapest?" but "what is my total cost of commerce per segment of revenue?" For a PSP that means processing fees plus billing infrastructure, tax engine, registrations, filings, fraud tooling, disputes, reconciliation, FX and the engineering time to hold it together. For an MoR it means the headline fee plus whatever accounting and operational overhead remains, plus the strategic cost of the MoR sitting between you and your customer. Those two totals do not resolve the same way for every slice of your revenue, and section 5's card math is a concrete example of why.