Answers that do not mention BigCommerce

For a small handmade-goods shop, tight budget, and maximum control, I’d start with WooCommerce on self-hosted WordPress.

Why it fits:

• No platform subscription or per-order fee — WooCommerce itself is open source; you mainly pay for hosting, your domain, payment processing, and any optional extensions. woocommerce.com
• Easy to self-host — unlike headless platforms such as Medusa or Saleor, you get a complete storefront and admin rather than having to build the frontend yourself. www.opensourcegroups.com
• Huge ecosystem for payments, shipping, taxes, product variations, coupons, email, SEO, etc.
• Excellent for handmade products where you might need variants such as size/color, inventory counts, made-to-order products, and custom shipping rules.
• You retain control of your site, database, code, and hosting, and WooCommerce exposes a REST API for further customization. woocommerce.com

The alternatives

PrestaShop is probably the main alternative I'd investigate. Its current open-source version is fully self-hostable and includes catalog, inventory, orders, shipping, payments, SEO, and customization capabilities. prestashop.com

I wouldn't start with Medusa or Saleor for a tiny shop unless you're already comfortable developing software. They're powerful API/headless systems, but you'll generally take on considerably more development and operational work than with WooCommerce. openbusinessnetwork.online

A very cheap setup

I'd aim for:

Your domain → inexpensive Linux hosting/VPS → WordPress → WooCommerce → free/lightweight theme → Stripe/PayPal → automated backups

The important caveat is that open source doesn't mean zero cost. Hosting, payment processing, backups, security updates, and occasional paid plugins can still cost money. www.velsof.com

If you tell me roughly how many products you have, whether they're made-to-order, and whether you're comfortable with Linux/Docker, I can narrow this down to a specific stack and a realistic $/month self-hosting budget.

For a small handmade-goods shop on a tight budget where you want full control, I’d look first at WooCommerce.

Why WooCommerce fits

• Self-hosted: your WordPress/WooCommerce installation, database, products, customer data, and code live on infrastructure you control.
• No WooCommerce platform fee: WooCommerce says there are no transaction fees beyond whatever your payment processor charges. WooCommerce
• Low hardware/hosting requirements: a modest shared host or inexpensive VPS can handle a small catalog.
• Excellent for handmade products: product variants, inventory, coupons, shipping, photos, blogging/content, and SEO are all readily available.
• Huge ecosystem: you can add payment, shipping, tax, email, accounting, and other functionality without building everything yourself.
• You aren't locked into a hosted service: WooCommerce describes itself as open source with ownership of your store and data. WooCommerce

woocommerce.com

How I'd compare the main choices

PrestaShop is also genuinely self-hostable and open source; its current version 9 includes a redesigned admin, API, and updated Symfony stack. PrestaShop PrestaShop

Medusa and Saleor are more appropriate if you specifically want a headless/custom-built storefront. That gives you considerable control, but you're taking on substantially more development and maintenance than you need for a small handmade shop. OSSAlt Medusa Documentation

A cheap setup I'd use

Domain → inexpensive Linux hosting → WordPress → WooCommerce → payment processor

Start with:

• inexpensive shared hosting rather than a VPS
• a free/lightweight WooCommerce-compatible theme
• WooCommerce's core functionality
• only the plugins you actually need
• automated backups
• HTTPS
• a payment gateway such as Stripe/PayPal, where the processor—not WooCommerce—handles payment processing

The important caveat is that open source isn't literally zero-cost: you'll still pay for hosting, domain registration, payment processing, and potentially premium extensions. You'll also be responsible for updates, backups, and security. WooCommerce ossium

Bottom line: for a small handmade shop, I'd choose WooCommerce unless you already have significant web-development skills and specifically want a headless architecture. It gives you a particularly good combination of ownership, low initial cost, and not having to build your ecommerce system from scratch.

A cheap setup I'd use

For a 10-person engineering team building a high-traffic headless storefront, I’d shortlist Medusa, Saleor, and Vendure. I’d lean toward Medusa if your team is primarily TypeScript/React, but the architecture and API preferences matter more than a generic “best platform” label.

Medusa describes its core as a customizable commerce framework with modules for DTC, B2B, marketplaces, etc. GitHub Saleor is GraphQL-first and currently documents a stack designed for composable/headless deployments; its site also reports production deployments handling very large API volumes. Saleor Saleor Vendure is TypeScript/NestJS/GraphQL and explicitly supports self-hosting on Docker/Kubernetes/cloud infrastructure. GitHub

What I'd choose for your situation

1. Medusa — my default shortlist choice

For a 10-person startup, the big advantage is that your storefront and commerce backend can live comfortably in the same TypeScript ecosystem. That reduces context switching and makes it relatively straightforward to implement custom pricing, promotions, checkout logic, integrations, and asynchronous workflows.

It's also modular rather than forcing you into a giant monolithic commerce suite. GitHub

For high traffic, I'd architect it roughly as:

CDN → Next.js/React storefront → cache → Medusa API → PostgreSQL + Redis → async workers → external services

Don't interpret "high traffic" as requiring the commerce API itself to handle every storefront request. Cache aggressively at the storefront/CDN layer and keep the transactional path—cart, checkout, inventory, order creation—much smaller.

2. Saleor — seriously consider it if GraphQL is central

Saleor is attractive if your team likes Python/Django and wants a GraphQL-native commerce API. Its current open-source stack includes the API, dashboard, extensions/apps, and composable storefront tooling. Saleor

One particularly interesting data point for your use case is Saleor's published infrastructure scale: its site currently reports deployments handling 1B+ API requests/month and hundreds of thousands of orders/month. That doesn't mean your deployment automatically scales that way, but it is evidence that the architecture is intended for substantial traffic. Saleor

I'd favor it when GraphQL + sophisticated multi-channel/B2B commerce are more important than having an all-TypeScript stack.

3. Vendure — strong choice for a structured TypeScript team

Vendure is also a very natural fit for a React/TypeScript startup. It uses NestJS and GraphQL and has a plugin-oriented architecture. GitHub

I'd put it especially high on the list if you expect the commerce domain to become complicated—multiple channels, B2B, marketplace functionality, subscriptions, custom business rules—and you want more architectural structure imposed by the framework.

The decision I'd make

If you told me only:

10 engineers + high-traffic + headless + open source + startup

I'd start with Medusa and Vendure, then add Saleor if your team is comfortable with Python/GraphQL.

The more important question is actually what kind of "high traffic" you mean. A storefront getting 100k requests/minute is a very different engineering problem from a catalog with 10 million SKUs or a checkout system processing 1,000 orders/minute.

If you give me your approximate peak requests/minute, SKU count, orders/minute, B2C vs B2B, and whether your team is TypeScript or Python, I can give you a much more concrete architecture comparison—including where I'd put Redis, Postgres replicas, search, CDN caching, queues, and inventory so you don't accidentally make the commerce backend your bottleneck.

For a 10-person team building a high-traffic headless storefront, I’d narrow it to Saleor, Vendure, and Medusa. I would not start with Magento/Adobe Commerce unless you have a specific reason to accept its operational complexity.

My shortlist

Saleor is explicitly API-only/headless and documents a cloud-native architecture, multichannel commerce, webhooks and extensions. Its current public cloud metrics also show substantial production traffic. Saleor Commerce GitHub Saleor Commerce

Vendure is particularly interesting for a small engineering team: it's TypeScript end-to-end, built on NestJS/GraphQL, and lets you extend or override core behavior through plugins and service overrides rather than maintaining a fork. It supports PostgreSQL/MySQL/MariaDB/SQLite and self-hosting on Docker/Kubernetes. GitHub Vendure

Medusa is another strong TypeScript option. Its core commerce modules are open source and designed specifically as a customizable foundation for DTC, B2B, marketplaces and other commerce applications. GitHub

What I'd pick

If your team is primarily TypeScript: I'd prototype Vendure first, with Medusa as the other serious contender.

The reason isn't that Vendure is universally "better"—it's that a 10-person startup benefits enormously from keeping the commerce backend in the same language/ecosystem as the storefront and from having a fairly opinionated commerce model without locking yourselves into a SaaS platform. Vendure's plugin architecture is also a good fit when you expect unusual pricing, fulfillment, catalog, or checkout logic. GitHub GitHub

If your team is comfortable with Python and you want a very API/composable-oriented architecture: I'd put Saleor at the top of the POC list. Its GraphQL-only architecture and app/webhook model are particularly attractive for a genuinely headless storefront. Saleor Commerce Saleor Documentation

For "high traffic," I'd optimize the architecture—not just the platform

I'd build roughly:

Next.js storefront → CDN/edge cache → commerce API → PostgreSQL + Redis → async workers/events

with:

• aggressive CDN caching for product/category pages
• server-side/API caching for catalog queries
• PostgreSQL as the source of truth
• Redis for appropriate ephemeral/cache workloads
• asynchronous processing for email, search indexing, analytics, feeds, etc.
• a dedicated search engine once catalog/search requirements justify it
• payment provider hosted/tokenized flows rather than putting card data through your infrastructure
• load testing checkout and cart separately from cacheable catalog traffic

The important distinction is that 1M product-page requests and 1M checkout requests are completely different scaling problems. A headless platform can handle enormous read traffic if you keep the storefront/CDN/cache architecture doing its job; checkout, inventory, promotions and order writes are where I'd concentrate the load-testing effort.

One other consideration: license matters for a startup. Vendure Core is GPLv3, while Medusa's core is MIT and Saleor is BSD-3. GitHub GitHub Saleor Commerce I'd have counsel review the exact way you plan to modify/distribute the backend before committing.

If you tell me your expected traffic/orders, catalog size, TypeScript vs Python preference, and whether you need B2B/multi-region/multi-warehouse, I can give you a much more concrete architecture and a Saleor vs Vendure vs Medusa decision matrix for your specific workload.

What I'd pick

If your team is primarily TypeScript: I'd prototype Vendure first, with Medusa as the other serious contender.

The reason isn't that Vendure is universally "better"—it's that a 10-person startup benefits enormously from keeping the commerce backend in the same language/ecosystem as the storefront and from having a fairly opinionated commerce model without locking yourselves into a SaaS platform. Vendure's plugin architecture is also a good fit when you expect unusual pricing, fulfillment, catalog, or checkout logic. GitHub GitHub

If your team is comfortable with Python and you want a very API/composable-oriented architecture: I'd put Saleor at the top of the POC list. Its GraphQL-only architecture and app/webhook model are particularly attractive for a genuinely headless storefront. Saleor Commerce docs.saleor.io

For "high traffic," I'd optimize the architecture—not just the platform

I'd build roughly:

Next.js storefront → CDN/edge cache → commerce API → PostgreSQL + Redis → async workers/events

with:

For a freelance developer doing custom integrations across multiple client stores, I’d put Medusa at the top of the shortlist.

My take

Why Medusa stands out

Medusa's architecture is unusually friendly to the kind of work you're describing. Its backend is organized around modules, workflows, and API routes, and custom modules can live directly inside the application rather than forcing you to maintain a separate integration service. Medusa Docs Medusa Docs

For example, an ERP integration could look conceptually like:

ERP ↔ custom Medusa module ↔ Medusa workflow ↔ orders/products/customers

You can also package recurring work as plugins containing modules, API routes, workflows, subscribers, scheduled jobs, and admin extensions. That is particularly attractive for a freelancer because you can turn solutions you've built for one client into reusable infrastructure for the next client. Medusa Docs Medusa Docs

Medusa explicitly positions its framework as a way to build integrations with systems such as ERPs, CMSs, payment providers and fulfillment services, including workflows that orchestrate operations across systems. Medusa Docs

When I'd choose Saleor instead

Saleor is compelling if your philosophy is "don't modify the commerce core; build everything around a clean API." It's GraphQL-only and its extension model centers on apps, webhooks, metadata and API extensions. Saleor Documentation GitHub

That's excellent for integrations where you want the custom code independently deployable. The tradeoff is that you're generally extending around Saleor rather than having the same degree of in-process access to the commerce application that Medusa gives you.

When Sylius makes more sense

If you're already very comfortable with PHP/Symfony, Sylius deserves serious consideration. Its plugin architecture is designed specifically for extending the platform without modifying core code, and plugins can cover things like payment gateways, CRM/ERP integrations, wishlists, and CMS functionality. Sylius Academy

I'd favor it when the client's application needs deep domain-level customization and Symfony is already part of your stack.

And WooCommerce?

WooCommerce is probably the easiest ecosystem to find existing integrations and developer resources for. Its official developer documentation explicitly supports building extensions and customizing WooCommerce. The WooCommerce Developer Blog

But I'd distinguish "easiest to extend" from "easiest ecosystem to find a plugin for." For a developer building highly custom systems repeatedly, I would generally find Medusa's modular architecture cleaner than accumulating WordPress/WooCommerce hooks and plugins.

If I were choosing a default stack for a freelance custom-commerce practice: I'd start with Medusa + TypeScript/Postgres, build your common integrations as reusable modules/plugins, and use Saleor when an API-first/composable architecture is more appropriate.

If your main business is building custom integrations for clients, I’d look first at Medusa.

Why Medusa fits a freelance integration shop

Medusa is built around a modular architecture rather than treating the store as a giant monolith. Its current architecture gives you:

• Custom modules for client-specific domains and third-party integrations.
• Workflows for multi-step business logic, including rollback/data consistency.
• API routes, subscribers, scheduled jobs, and admin extensions inside plugins.
• Access to the underlying commerce modules—products, orders, customers, inventory, payments, fulfillment, etc.—from your custom code.
• Reusable plugins/modules that you can carry from one client project to another. Medusa Docs Medusa Docs Medusa Docs

That last point is particularly attractive for a freelancer. You could, for example, build your own reusable integration package for an ERP, PIM, shipping provider, or CRM and deploy it across multiple client stores.

How I'd think about the main options

Saleor is also very extensible, but its model is more deliberately API/app oriented: its documentation emphasizes GraphQL, apps, webhooks, and extending the dashboard. Saleor Documentation

Medusa, by contrast, lets your custom code live directly in the application architecture. Its modules can define their own models/services, and workflows can combine those custom modules with Medusa's built-in commerce modules. Medusa Docs Medusa Docs

For a freelance developer, I'd choose based on stack

If you're primarily TypeScript/Node → Medusa.

If you're primarily Python/Django/GraphQL → Saleor.

If you're primarily PHP/Symfony → Sylius.

If you're primarily PHP/Laravel → Bagisto.

If your clients mostly want "WordPress + ecommerce + whatever plugin does it" → WooCommerce.

The interesting distinction is that WooCommerce may be the easiest to extend in terms of sheer ecosystem, whereas Medusa is arguably the cleaner starting point when your actual product is custom engineering. Current comparisons likewise distinguish WooCommerce's enormous extension ecosystem from Medusa's more developer-oriented modular architecture. OSSAlt OpenReplay Blog

For your particular business model, I'd seriously consider standardizing on Medusa + Next.js + PostgreSQL + Redis and building a small internal library of reusable integration modules. Medusa's own docs explicitly support reusable plugins containing modules, workflows, API routes, subscribers, scheduled jobs, and admin extensions. Medusa Docs Medusa Docs

If you tell me what kinds of integrations your clients typically need (ERP, Shopify migration, Stripe, shipping, PIM, marketplaces, subscriptions, B2B pricing, etc.), I can compare Medusa vs Saleor vs Sylius specifically for those workloads.

If your priorities are:

1. Multi-vendor marketplace capability (vendors, commissions, vendor dashboards, vendor fulfilment, payouts)
2. Total data ownership (self-hosting, full database access, no platform lock-in)
3. Commercial use without transaction fees
4. Ability to customize deeply

then you should usually look at open-source commerce engines plus marketplace extensions, rather than traditional hosted marketplace SaaS. OSSAlt

Shortlist

Kryeo Pawel Owerczuk## My recommendation by scenario

1. You are building an Amazon-style product marketplace

Example:

• Multiple sellers
• Sellers upload products
• Platform takes commission
• Seller dashboards
• Vendor order splitting
• Your own payment logic

Best starting point: Medusa + marketplace layer

Why:

• Modern TypeScript stack
• Designed for heavy customization
• Self-hostable
• Easier to build unique marketplace workflows without fighting the core system

Pawel Owerczuk Build With MatijaTypical architecture:

Frontend
(Next.js / React)

        ↓

Marketplace Commerce Engine
(Medusa)

        ↓

PostgreSQL
(your database)

        ↓

Payments
(Stripe Connect / Adyen MarketPay)

        ↓

Your infrastructure
(AWS / Azure / bare metal)

You own:

• Customer data
• Vendor data
• Orders
• Product catalogue
• Analytics
• Business logic

2. You need enterprise B2B marketplace features

Examples:

• Company accounts
• Approval workflows
• Multiple warehouses
• Complex pricing
• International operations

Consider Saleor.

It has a strong commerce foundation with GraphQL, channels, permissions, and enterprise-oriented features. Eryx Innovations Lab Designodin

3. Your team is TypeScript/NestJS focused

Consider Vendure.

It provides:

• Plugin-based architecture
• GraphQL APIs
• Strong TypeScript alignment
• Good foundations for multi-channel commerce

Check licensing carefully if you plan significant modifications to the core. Eryx Innovations Lab Vendure

Solutions I would be cautious about for "total data ownership"

Shopify Marketplace apps

Good operationally, but:

• Your core data lives inside Shopify
• Marketplace logic depends on apps
• You remain dependent on Shopify APIs and policies

SaaS marketplace builders

Good for speed, but:

• Export options vary
• Database ownership is usually not yours
• Deep customization can become difficult

My practical ranking for your stated requirements

For a company saying "we need multi-vendor marketplace + total data ownership":

1. Medusa — best balance of ownership, flexibility, and custom marketplace development
2. Vendure — strong if your team is TypeScript and prefers plugin architecture
3. Saleor — strong if your organization prefers Python/GraphQL and enterprise commerce patterns

Kryeo VendureBefore committing, I would validate these specific requirements against the platform:

• Vendor onboarding workflow
• Vendor admin portal
• Commission calculation rules
• Split payments/payouts
• Vendor-specific fulfilment
• Returns/refunds handling
• Tax/VAT ownership
• Data export strategy
• Multi-region inventory
• SLA/security requirements

If you share your marketplace type (physical products, services, rentals, B2B wholesale, etc.), expected vendor count, and your technology stack, I can narrow this to a more specific architecture.

If multi-vendor marketplace capability + complete control of your data/infrastructure are the two primary requirements, I’d shortlist Mercur, Vendure, and Medusa.

My take

Mercur is the closest match if you want marketplace functionality rather than an ecommerce engine that you turn into a marketplace. Its current project describes vendor onboarding, multi-vendor catalogs, offers, commissions and automated payouts as built-in capabilities, with self-hosting and full source access. GitHub GitHub

Vendure is the one I'd investigate alongside it for a larger, highly customized platform. It is TypeScript/NestJS/GraphQL, explicitly supports B2B and multi-vendor marketplace use cases, and can be self-hosted on your own infrastructure. Its current license is GPLv3, with commercial licensing available for some platform capabilities. GitHub GitHub

Medusa is attractive if your engineering team wants a very flexible commerce framework, and its core is MIT-licensed. However, its current model is open-core: some Enterprise Edition functionality requires a commercial agreement. Also, marketplace functionality is more of a framework/building-block approach than Mercur's marketplace-first model. GitHub

For your requirements, I'd structure the decision this way

• Want the most marketplace functionality out of the box: Mercur
• Want a sophisticated, extensible enterprise architecture: Vendure
• Want maximum freedom to engineer your own commerce model: Medusa
• Want to avoid SaaS/vendor lock-in: all three can be self-hosted, but examine each project's current licensing and any enterprise-only components carefully.

One important distinction: "data ownership" isn't really determined by open source alone. For true ownership, I'd require the solution to let you self-host the application and database, export all marketplace/customer/order/vendor data, control object/file storage, and avoid essential functionality that requires a proprietary hosted service.

If you tell me your approximate vendor count, monthly orders/GMV, B2B vs B2C, and tech stack (e.g. Node/TypeScript, PHP, Python), I can give you a more rigorous architecture-level comparison of Mercur vs Vendure vs Medusa, including vendor onboarding, commissions, payouts, RBAC, catalog ownership, order splitting, search, scalability, and licensing.

Medusa is attractive if your engineering team wants a very flexible commerce framework, and its core is MIT-licensed. However, its current model is open-core: some Enterprise Edition functionality requires a commercial agreement. Also, marketplace functionality is more of a framework/building-block approach than Mercur's marketplace-first model.

For your requirements, I'd structure the decision this way

For complex international tax, currency, language, and regional rules, I’d shortlist Saleor and Medusa. Saleor is particularly designed around multi-channel, multi-currency, tax, warehouse, and permission complexity. Macrostack eryxinnovationslab.com

• Saleor — strongest fit for highly complex localization; Python/GraphQL, channel-based regionalization.
• Medusa — excellent if your team is TypeScript/Node; its Region module supports localized currencies, taxes, shipping, and payments. Medusa
• Shopware — worth considering if you want a more turnkey, traditional ecommerce platform rather than a headless commerce backend. EU Vetted

For your specific description, I’d investigate Saleor first, especially if tax jurisdiction logic and many distinct markets are central requirements.

If complex international tax + currency localization is the core requirement, I’d narrow the self-hosted field to Adobe Commerce (Magento Open Source) and Medusa.

My practical comparison

Adobe Commerce/Magento Open Source is the more established choice when the localization rules themselves are the difficult part. Its tax engine combines product classes, customer classes, tax rates and geographic regions, and it supports multiple websites/stores/store views with separate currencies and localized configuration. Experience League Experience League Experience League

It also has explicit facilities for international VAT, tax-inclusive/exclusive pricing, tax zones, and country-specific configurations. Experience League Experience League

Medusa is particularly attractive if you have a strong engineering team and want to build a bespoke international commerce stack. Its region model associates countries with currencies, payments, fulfillment and tax behavior; its tax module supports regional tax rules and custom tax providers. Medusa Documentation Medusa Documentation Medusa Documentation

Medusa also supports tax-region inheritance/combination and country-specific tax calculation based on the customer's shipping address, which can be useful when your rules become granular. Medusa Documentation Medusa Documentation

What I'd choose based on your situation

• Hundreds of tax/country combinations, B2B + B2C, multiple storefronts, complicated catalog/pricing: Magento Open Source / Adobe Commerce
• Headless architecture, custom checkout, sophisticated engineering team, tax provider integrations: Medusa
• You want the platform to provide as much of the international-commerce machinery as possible: Magento
• You want the commerce engine to be a programmable foundation that you tailor heavily: Medusa

One important distinction: neither platform should be treated as a source of truth for constantly changing tax law. For serious international operations, I'd architect the store so tax calculation can be delegated to a specialized tax provider where necessary, while the ecommerce platform handles countries, currencies, catalog, customers and checkout.

If you tell me roughly how many countries you sell into, B2B vs B2C, whether prices are tax-inclusive, and whether you need VAT/GST/sales-tax automation, I can compare Magento vs Medusa specifically for your architecture.

Medusa is particularly attractive if you have a strong engineering team and want to build a bespoke international commerce stack. Its region model associates countries with currencies, payments, fulfillment and tax behavior; its tax module supports regional tax rules and custom tax providers.

Medusa also supports tax-region inheritance/combination and country-specific tax calculation based on the customer's shipping address, which can be useful when your rules become granular.

For a solo founder with limited development skills, I’d separate “modern admin” from “easy theming,” because the answer changes depending on whether you want a ready-to-run store or a headless commerce engine.

My shortlist

What I'd look at first

1. WooCommerce — if you want to actually launch the store yourself

WooCommerce gives you the biggest advantage for a non-developer: the storefront, admin, themes, plugins, payments, shipping, etc. are already an ecosystem, rather than you having to assemble a headless application. Current comparisons continue to characterize it as the turnkey option with a very large extension ecosystem. OSSAlt Open Business Network

The downside is that it can feel less like a modern SaaS product architecturally, and WordPress/plugin maintenance becomes part of your life.

2. Bagisto — interesting if you want a more purpose-built open-source store

Bagisto is Laravel/PHP-based and includes things like theme management, multi-currency and multi-locale functionality. That makes it considerably more “store platform” than a pure commerce backend. OSSAlt

I'd consider it if you want something closer to “install ecommerce software, customize the storefront, run the business” rather than building a frontend application.

3. Medusa — if “modern” means Next.js/React

Medusa is probably the most compelling choice if you want a modern headless architecture. It's TypeScript/Node-based, has a modern admin, and is designed around building a custom storefront. Its current ecosystem makes it particularly attractive for JavaScript/TypeScript developers. ShipGarden Pawel Owerczuk

But there's an important catch for you: it's not really a Shopify-like turnkey store builder. You generally build/connect the storefront yourself. That makes its theming flexibility excellent, but its ease for a non-developer considerably worse than WooCommerce.

The key distinction

If your requirement is:

“I want to launch a nice-looking ecommerce site myself, change the design easily, and spend my time on the business rather than coding.”

I'd start with WooCommerce or Bagisto.

If it's:

“I want a genuinely modern React/Next.js storefront and I'm willing to learn some development.”

I'd look closely at Medusa.

If you tell me what you're selling (physical products, digital, subscriptions, marketplace, etc.) and whether you're comfortable with WordPress, I can narrow this down to 2–3 options and explain exactly what your first-week setup would look like.

If your priorities are solo-founder friendly + modern admin + easy theming, I’d narrow it to Medusa, Vendure, and Saleor rather than the older all-in-one platforms.

My practical shortlist

Medusa

For your particular situation, Medusa is probably the first one I'd investigate.

Its admin is genuinely modern: it has light/dark themes, search, configurable layouts, and can be extended with custom admin pages/widgets. Medusa Docs Medusa Docs Medusa Docs

The storefront is deliberately separate from the commerce backend, so you can use a React/Next.js frontend and essentially make the customer-facing site look however you want. Medusa also provides a Next.js starter with a fairly polished starting design. Medusa Docs Medusa Docs

The catch for a low-code founder is that headless means you're still responsible for the storefront. You don't get the same "pick a theme, change colors, publish" experience that you get from Shopify or WooCommerce.

Vendure

I'd put Vendure very close to Medusa if you're comfortable with TypeScript.

Its current dashboard has a modern theming system based around CSS variables/Tailwind, and you can customize the dashboard's appearance. Vendure Developer Hub

It's particularly interesting if you expect to eventually need sophisticated commerce logic. The tradeoff is that you're still dealing with a developer-oriented headless system rather than a turnkey merchant platform.

Saleor

Saleor has an excellent modern architecture and very capable dashboard/API, with built-in concepts for things like channels, promotions, multi-region selling, and dashboard extensions. Saleor Documentation

But for a solo founder with limited development experience, I'd be cautious about choosing it solely because it looks polished. Its Python/Django + GraphQL architecture introduces a different technical stack, whereas Medusa keeps you in the JavaScript/TypeScript ecosystem. Eryx Innovations Lab

The important distinction

If by "easy theming" you mean:

"I want to install a theme, change fonts/colors/layout, add products, and mostly stay out of code."

Then I would actually look at WooCommerce or Shopware before these headless platforms.

If you mean:

"I want a modern-looking storefront that I can heavily customize without fighting the platform."

Then Medusa is the one I'd prototype first.

The architecture is essentially:

Medusa → products/orders/payments/commerce logic
Next.js → your actual website/design

That separation is what makes the theming flexible, but it's also what makes Medusa more technical than Shopify/WooCommerce. Medusa Docs

For a solo founder, I'd test one thing before committing: take your desired homepage/product page design and try rebuilding it with Medusa's starter. If you can get ~80% of your desired look without needing custom backend code, you've probably found a good fit.

If you tell me what you're selling (physical products, digital products, subscriptions, marketplace, etc.) and whether you want to use a visual builder, I can narrow this down to 2–3 platforms and explain the actual setup effort.

For a static/headless storefront, I’d shortlist these:

• Medusa — my default for a TypeScript/JavaScript stack. REST API + JS SDK, modular architecture, and MIT-licensed core. Vendure Pawel Owerczuk
• Saleor — excellent if you prefer GraphQL and have Python/Django expertise; particularly strong for multi-channel commerce. Vendure
• Vendure — TypeScript + NestJS + GraphQL, with a structured plugin system; worth considering for complex B2B/multi-channel setups. Vendure Eryx Innovations Lab

For a static site specifically, I’d start with Medusa unless GraphQL is a firm requirement. Your architecture can be:

Static site → Medusa API → PostgreSQL

If you tell me your frontend stack (Astro, Next.js, plain HTML, etc.) and whether you need subscriptions/B2B/multiple stores, I can narrow it down.

If the goal is a static frontend talking to a self-hosted commerce backend entirely through APIs, I'd shortlist these three:

My practical pick: Medusa

docs.medusajs.com

For a static site, Medusa is probably the simplest starting point. Its Store API is explicitly designed for storefronts and mobile apps, and the current API is REST-based with an OpenAPI specification available. Medusa Docs

A typical architecture would be:

Static site
  │
  │ REST / HTTPS
  ▼
Medusa
  ├── Products
  ├── Pricing
  ├── Cart
  ├── Customers
  ├── Orders
  ├── Inventory
  └── Payments
       │
       ├── PostgreSQL
       └── Payment provider

That works particularly nicely if your frontend is Astro, Next.js, React, Vue, or plain HTML/JS. Your static site doesn't need to know anything about the backend implementation.

When I'd choose the others

Saleor is compelling if you specifically want GraphQL. It has a Python/Django core and is designed around a GraphQL-first API. Eryx Innovations Lab Vendure

Vendure is worth considering if your team is heavily TypeScript-oriented but prefers GraphQL + a structured plugin architecture. It's built on NestJS and exposes GraphQL APIs. Eryx Innovations Lab Vendure

One important distinction is licensing: Medusa's core is MIT, Saleor's core API is BSD-3-Clause, while Vendure's core is GPLv3 with a commercial layer, so I'd have your team check the exact components/licenses before committing. Pawel Owerczuk Vendure

So, if you tell me your frontend stack (Astro/Next/plain HTML/etc.), whether you need subscriptions/B2B/multiple stores, and where you plan to deploy it, I can narrow this down to the architecture I'd use.

For a self-hosted subscription-box business, I’d narrow it to WooCommerce or Medusa, depending on how much custom engineering you want.

My practical recommendation: WooCommerce

woocommerce.com

If your priority is getting a subscription box operational without building the billing system yourself, WooCommerce is the safer choice.

• WooCommerce itself is open source and self-hostable.
• WooCommerce Subscriptions adds recurring billing, automatic renewals, trials, signup fees, expiration, billing-date alignment, and subscription management. WooCommerce WooCommerce
• It supports automatic recurring payments through compatible payment gateways and has a documented failed-payment retry system. WooCommerce WooCommerce
• It's particularly well suited to physical subscription boxes because recurring orders can drive your normal inventory and fulfillment workflow. WooCommerce
• The downside: the subscription functionality itself is a paid extension, so it isn't a completely free/open-source subscription stack. WooCommerce

If you're building a custom product: Medusa

docs.medusajs.com

Medusa is more attractive if you're a developer and want a headless, API-first commerce backend with your own frontend and business logic.

The important caveat is that subscriptions aren't native. Medusa's own documentation provides a recipe for implementing subscriptions, including automatic renewal, expiration, customer cancellation, and Stripe-based recurring payments. Medusa Docs

So you'd be building/maintaining more of the subscription infrastructure yourself.

For a subscription box specifically, I'd start with WooCommerce unless you have a strong reason to build a headless/custom stack.

One architectural point I'd strongly recommend: let Stripe/payment infrastructure handle the actual recurring payment authorization rather than trying to store cards and implement recurring charging yourself. WooCommerce's ecosystem supports both site-initiated renewals and, in some configurations, Stripe Billing's off-site billing engine. WooCommerce

If you tell me your stack (PHP/Node/etc.), expected subscribers, Stripe vs. another processor, and whether customers can customize their box, I can give you a more specific architecture recommendation.

For a self-hosted subscription box business, I’d narrow it to WooCommerce vs. Medusa rather than looking at every open-source platform.

My practical recommendation: WooCommerce

woocommerce.com is the safer choice if your priority is getting recurring physical-product billing working reliably without building the subscription system yourself.

• WooCommerce itself is open source and self-hostable. WooCommerce
• Its Subscriptions extension handles recurring billing, renewals, trials, signup fees, expiration, billing-date alignment, and subscription management. WooCommerce WooCommerce
• It supports automatic recurring payments through compatible payment gateways and has a failed-payment retry system. WooCommerce WooCommerce
• It is particularly well suited to physical subscription boxes because each renewal generates an order, which can naturally drive fulfillment/shipping. WooCommerce
• The downside is that WooCommerce Subscriptions is a paid extension, so the overall stack isn't completely free. WooCommerce

When I'd choose Medusa instead

medusajs.com makes more sense if you're a developer and want a custom/headless architecture with your own frontend and business logic.

The important catch: subscriptions aren't native. Medusa's own documentation describes subscriptions as something you implement/customize, including automatic renewal and Stripe integration. Medusa Docs

So you'd be taking responsibility for more of the machinery:

• subscription state
• renewal scheduling
• payment-provider webhooks
• failed-payment handling
• pauses/cancellations
• subscription changes
• fulfillment generation
• customer subscription management

That's attractive if those things are part of your product's differentiation, but unnecessary complexity if you're primarily trying to sell boxes.

Quick comparison

For your particular use case, I'd start with WooCommerce + WooCommerce Subscriptions + Stripe/WooPayments, self-hosted on your own server. WooPayments can use Stripe Billing as an off-site billing engine in the U.S., while automatic renewals are supported through compatible payment methods. WooCommerce

If you tell me roughly how many subscribers you expect (e.g. 100, 10k, 100k), whether customers can customize what's in each box, and whether you're building the frontend yourself, I can narrow down the architecture and hosting stack considerably.

For an enterprise team prioritizing scalability, containerized deployment, and long-term control, I’d shortlist Medusa, Saleor, and Vendure rather than traditional monolithic platforms. Current comparisons consistently identify these as the major open-source headless options. Vendure Pawel Owerczuk

My recommendation: Medusa

For a greenfield enterprise deployment, I'd generally start with Medusa if your engineering organization is already strong in TypeScript/Node.js. Its modular architecture makes it well suited to separating services and scaling workloads horizontally, while its API-first/headless model works naturally with Kubernetes, containers, independent storefronts, and surrounding microservices. Recent production comparisons also report an active ecosystem and substantial contributor base. Pawel Owerczuk

A typical deployment could look like:

                    ┌───────────────┐
                    │ CDN / WAF     │
                    └───────┬───────┘
                            │
                    ┌───────▼───────┐
                    │ Next.js / Web │
                    └───────┬───────┘
                            │
              ┌─────────────▼─────────────┐
              │       Kubernetes          │
              │                           │
              │  Medusa API pods          │
              │  Worker pods              │
              │  Custom commerce modules  │
              └──────┬─────────┬──────────┘
                     │         │
              ┌──────▼───┐ ┌──▼──────┐
              │PostgreSQL│ │ Redis   │
              └──────────┘ └─────────┘

Choose Saleor instead if your organization is predominantly Python/Django and wants a GraphQL-first commerce platform with substantial built-in enterprise/B2B functionality. Pawel Owerczuk Eryx Innovations Lab

Choose Vendure instead if you're a TypeScript/NestJS organization that particularly values its plugin architecture and structured backend. One important consideration is its GPLv3 licensing model, versus the more permissive licensing associated with Medusa's core, so enterprise legal review matters. Pawel Owerczuk

So, assuming TypeScript + Kubernetes + high customization + enterprise scale, my shortlist would start with Medusa, with Saleor as the main alternative for a Python/GraphQL organization. This is a fit-based recommendation rather than a universal ranking—the architecture and team stack should drive the final choice. Vendure

For an enterprise team prioritizing scalability and containerized deployment, I’d put Vendure and Saleor at the top of the shortlist, with Vendure particularly attractive if your organization is TypeScript/Node-oriented.

My practical recommendation

Choose Vendure if your requirements look like:

• Kubernetes/Docker is a first-class deployment target.
• You want horizontal scaling behind load balancers.
• You want commerce workers to scale independently from the API layer.
• Your engineering organization uses TypeScript/Node.js.
• GraphQL is desirable.
• You expect significant customization through plugins rather than modifying the platform core.

Vendure explicitly describes its application layer as stateless and horizontally scalable, with separate worker processes, and supports deployment on Kubernetes, Docker, AWS, Azure, GCP, on-premises, and air-gapped environments. Vendure Vendure

Choose Saleor instead if your enterprise is predominantly Python/Django, wants a GraphQL-first architecture, and prefers extending commerce through external applications/webhooks. Current comparisons consistently distinguish Saleor's Python/Django architecture from the TypeScript-based Vendure and Medusa approaches. Kryeo Eryx Innovations Lab

Choose Medusa if the overriding requirement is maximum flexibility for a TypeScript-native engineering team and you want commerce logic implemented directly as modules/workflows. Kryeo

One important enterprise issue is licensing: current sources report Vendure's core as GPLv3 with a commercial licensing option, while Medusa's core is MIT and Saleor's core is BSD-3-Clause. Have legal review the exact components/licenses you intend to deploy rather than evaluating only the headline license. Pawel Owerczuk

Bottom line: for “enterprise + Kubernetes/Docker + horizontal scalability + TypeScript”, I'd start the technical evaluation with Vendure. For “enterprise + Python + GraphQL”, start with Saleor.

If you want one self-hosted open-source system for a catalog containing both physical and digital products, I’d shortlist these:

My practical recommendation: Medusa

medusajs.com is particularly well aligned with your requirement. Its current product model explicitly supports physical or digital items, including products that don't require shipping or inventory, and lets you configure shipping/inventory requirements separately. Medusa Docs

It's also open source and self-hostable; the core is MIT-licensed, although Medusa uses an open-core model with some Enterprise Edition components under a commercial license. GitHub

That makes a catalog like this natural:

• Physical: T-shirt → inventory + shipping
• Digital: E-book → no inventory + no shipping
• Physical + digital: Printed book + PDF → inventory/shipping + digital fulfillment
• Service: Consultation → no inventory + no shipping

When I'd choose the others

Saleor is worth serious consideration if you expect a large/complex catalog, multiple channels, multiple warehouses, or a highly customized headless storefront. It's API-first/GraphQL, has rich catalog modeling, multi-channel controls, and can be run locally/self-hosted. Saleor Commerce GitHub

saleor.io

WooCommerce is the more conventional choice if you want a ready-to-operate store rather than assembling a headless commerce stack. Its enormous WordPress ecosystem makes physical + downloadable products straightforward, although the extension/plugin model can become more complex over time.

Sylius is excellent when you have PHP/Symfony developers and want the commerce system to be deeply customized. It's explicitly designed as an open-source headless framework rather than a turnkey store. GitHub Sylius

sylius.com

Bottom line: for a new build, I'd investigate Medusa first, then Saleor if your catalog/operations are going to be substantially more complex. If you want something you can administer largely without developers, WooCommerce changes the equation considerably.

If you tell me roughly how many physical products, how many digital products, and whether you need subscriptions/download protection/license keys, I can narrow this to the most suitable architecture.

For a single self-hosted open source system that handles both physical and digital products, I’d look at platforms that treat “product type” and fulfillment separately rather than forcing you into two stores.

Here are the strongest options:

My shortlist by scenario

If you want to launch and manage the store yourself

Choose WooCommerce.

It is usually the easiest path:

• physical products → inventory, shipping, variants
• digital products → downloads, protected files
• one cart and checkout
• lots of payment/shipping integrations

The tradeoff is that very large catalogs or complex workflows can require careful optimization and plugin choices.

If you want one commerce backend that can grow with custom needs

Choose Spree Commerce.

It is a better fit if you expect:

• thousands of products
• custom product types
• multiple storefronts
• ERP/warehouse integrations
• mobile apps or custom frontends

Spree specifically documents combining digital and physical products in one catalog and checkout flow. Spree Commerce

If you are building a developer-controlled platform

Consider Saleor, Medusa, or Spree.

These are less “install a shop and configure it” and more “build your commerce platform on top of an engine.”

Features I would verify before committing

For a mixed catalog, make sure the platform supports:

• One cart containing both product types

• e.g., customer buys a T-shirt + ebook together
• Different fulfillment rules

• shipping required for physical items
• download/license delivery for digital items
• Inventory only for physical goods
• No shipping calculation for digital goods
• Tax handling differences
• Digital file security

• expiring links, download limits, license keys
• Subscriptions/memberships if you sell recurring digital products

A practical recommendation:

• Most merchants: WooCommerce
• Serious catalog/customization: Spree Commerce
• Enterprise-style custom build: Saleor or Medusa

If you share your approximate catalog size (e.g., 50 products vs 50,000), whether you need subscriptions/licenses, and your preferred stack (PHP/Ruby/Node/Python), the choice narrows quite a bit.

If SEO + fast page loads + strong structured data are your priorities, I’d focus on Shopware 6, PrestaShop 9, and WooCommerce.

My take

Shopware 6 is probably the strongest match for your particular priorities. Its current 6.7 releases have added performance improvements, including stateless Store API requests and media-processing improvements. Shopware

It also has built-in SEO URL controls and robots/meta settings. Shopware Documentation Shopware Documentation More importantly for your requirement, Shopware has been improving its native JSON-LD structured-data foundation; 6.7.9 introduced JSON-LD support as an optional feature. Shopware

Its performance architecture also supports HTTP caching, Redis, Elasticsearch/OpenSearch, and other mechanisms useful for larger catalogs. Shopware Documentation

PrestaShop 9 is worth serious consideration if you want something more traditional and lightweight. It moved to Symfony 6.4/PHP 8.1+, reduced legacy code, and introduced improved SEO controls. PrestaShop

WooCommerce has excellent SEO and structured-data capabilities, including automatic JSON-LD product markup. GitHub But its performance ceiling depends heavily on your WordPress theme, plugins, hosting, database, caching, and how much functionality you add. WooCommerce's own documentation emphasizes caching, image optimization, database maintenance, minification, and CDNs for achieving good performance. The WooCommerce Developer Blog

If I were choosing based specifically on your priorities

• Performance-first + SEO: Shopware 6
• Open-source simplicity + solid SEO: PrestaShop 9
• SEO flexibility + huge ecosystem: WooCommerce
• Very large catalog / sophisticated storefront: Shopware 6 becomes particularly interesting

One important caveat: the platform alone doesn't determine Core Web Vitals. A well-configured WooCommerce site can be substantially faster than a badly configured Shopware site. Theme/frontend architecture, third-party scripts, caching, image delivery, hosting, and CDN configuration can matter more than the underlying platform.

If you tell me roughly how many products/variants you expect and whether you're building a headless storefront, I can compare Shopware vs PrestaShop vs WooCommerce specifically for Core Web Vitals, JSON-LD, crawlability, and large-catalog performance.

Its performance architecture also supports HTTP caching, Redis, Elasticsearch/OpenSearch, and other mechanisms useful for larger catalogs. Shopware Documentation

PrestaShop 9 is worth serious consideration if you want something more traditional and lightweight. It moved to Symfony 6.4/PHP 8.1+, reduced legacy code, and introduced improved SEO controls.

If SEO + fast page loads + structured data are your top priorities, I’d narrow the open-source field to Shopware, WooCommerce, and PrestaShop, with an important distinction:

My pick: Shopware 6

For your particular priorities, Shopware 6 is probably the most interesting starting point.

Its newer storefront architecture has dedicated JSON-LD structured-data templates, including WebSite, Organization, WebPage, BreadcrumbList, Product, and ItemList markup. It also lets developers extend that schema through themes/plugins rather than having to bolt the whole system on afterward. GitHub

It also has a headless architecture, so if you eventually want something like a very fast Next.js frontend, you aren't locked into the default storefront. Datafirefly

WooCommerce is the SEO/content alternative

WooCommerce is compelling if organic content is as important as product SEO. Its product pages automatically output Schema.org JSON-LD, including product information and breadcrumbs, and the markup is customizable through WordPress/WooCommerce hooks. GitHub

The catch is that performance depends heavily on your WordPress stack: theme, plugins, database, caching, images, and hosting. WooCommerce gives you enormous SEO flexibility, but it also gives you many more ways to accidentally make a site slow.

If you want maximum page-speed control

I'd also seriously consider Medusa or Saleor + a Next.js storefront. They're headless/open-source approaches where you control the HTML, JavaScript, caching, image delivery, and rendering strategy yourself. Open Business Network OpenReplay Blog

That's potentially the fastest architecture, but it requires substantially more development work. Structured data isn't as turnkey because you're responsible for implementing it correctly in the frontend.

Bottom line

If I were architecting around your priorities:

• Shopware 6 → best balance of ecommerce functionality, performance, SEO and built-in structured data.
• WooCommerce → best if blogging/content marketing and WordPress SEO flexibility are major requirements.
• Medusa/Saleor + Next.js → best if you have developers and want maximum control over Core Web Vitals.
• PrestaShop → solid conventional open-source ecommerce choice, particularly for multilingual/multicurrency stores. Open Source Groups

One caveat: the platform alone doesn't determine Core Web Vitals. A well-built WooCommerce store can outperform a poorly configured Shopware store, and a custom headless store can be slower if it ships too much JavaScript.

If you tell me roughly catalog size (e.g. 1k vs 100k products), expected traffic, and whether you want a traditional storefront or Next.js/headless, I can compare the architectures specifically for LCP, INP, TTFB, crawlability, JSON-LD, and hosting cost.

It also has a headless architecture, so if you eventually want something like a very fast Next.js frontend, you aren't locked into the default storefront. Datafirefly

WooCommerce is the SEO/content alternative

WooCommerce is compelling if organic content is as important as product SEO. Its product pages automatically output Schema.org JSON-LD, including product information and breadcrumbs, and the markup is customizable through WordPress/WooCommerce hooks. GitHub

The catch is that performance depends heavily on your WordPress stack: theme, plugins, database, caching, images, and hosting. WooCommerce gives you enormous SEO flexibility, but it also gives you many more ways to accidentally make a site slow.

If you want maximum page-speed control

I'd also seriously consider Medusa or Saleor + a Next.js storefront. They're headless/open-source approaches where you control the HTML, JavaScript, caching, image delivery, and rendering strategy yourself.

That's potentially the fastest architecture, but it requires substantially more development work. Structured data isn't as turnkey because you're responsible for implementing it correctly in the frontend.

Bottom line

If I were architecting around your priorities:

WooCommerce has the strongest overall plugin ecosystem and community support for most developers. Open Source Groups Toolradar

Quick comparison:

Shopify OpenReplay BlogWhy WooCommerce wins ecosystem-wise:

• Massive number of extensions for payments, shipping, subscriptions, marketplaces, CRM, ERP, and marketing. Shopify
• Huge developer pool because it sits on WordPress/PHP. Toolradar
• Easier to find existing solutions, tutorials, contractors, and community answers. Reddit

Choose Magento instead if you need enterprise-grade customization: multi-store, complex pricing rules, large catalogs, or heavy B2B workflows. Velocity

Choose Medusa or Saleor if you are building a custom headless commerce stack with React/Next.js and want API-first architecture rather than a traditional plugin marketplace. OSSAlt OpenReplay Blog

For a developer prioritizing ecosystem + community + available plugins, the usual ranking is:

1. WooCommerce
2. Magento Open Source
3. PrestaShop
4. Medusa
5. Saleor Open Source Groups OpenReplay Blog

If your priority is specifically the largest plugin/extension ecosystem plus an active developer community, I’d put WooCommerce at the top.

Why WooCommerce

WooCommerce benefits from the enormous WordPress ecosystem: there are thousands of extensions, themes, payment integrations, shipping integrations, and developer resources. Current 2026 comparisons continue to identify it as having the broadest open-source ecommerce ecosystem. Toolradar OSSAlt

Its development is also quite transparent: the WooCommerce GitHub monorepo contains the core plugins, packages, and tools, with public issues and contributions. GitHub

The main downside is that the ecosystem's size can also mean plugin conflicts, inconsistent quality, security/maintenance burdens, and upgrade headaches. A recent 2026 security incident involving a WooCommerce extension illustrates why extension vetting and timely updates matter. TechRadar

The important alternative: Magento Open Source

If by "plugin ecosystem" you mean deep, sophisticated extension architecture for developers, rather than simply the largest number of plugins, Magento Open Source deserves serious consideration.

Adobe says its marketplace has thousands of extensions, and its extension ecosystem includes formal quality checks. Experience League Adobe Commerce Developer Portal Magento also has a substantial developer community, documentation, GitHub activity, forums, and Stack Exchange resources. Adobe Business Adobe Developer

The tradeoff is substantially greater complexity. Magento makes more sense when you're building a large catalog, complex pricing/catalog rules, multiple stores, B2B functionality, or highly customized commerce workflows.

If you're a modern JS/TS developer

I'd look at Medusa or Vendure instead. They're much more developer-oriented and API/headless by design, but their extension ecosystems are nowhere near WooCommerce's maturity and breadth. Current 2026 comparisons describe Medusa and Vendure as particularly strong options for TypeScript/Node teams. Toolradar OpenReplay Blog

Bottom line:

• Maximum ecosystem + community: WooCommerce
• Deep enterprise extension architecture: Magento Open Source
• Modern TypeScript/headless development: Medusa or Vendure
• Python/GraphQL: Saleor
• Traditional PHP ecommerce: PrestaShop

If you tell me your stack (PHP/Node/Python), store size, and whether you're building a monolith or headless, I can narrow this down to the platform that fits your development workflow.

WooCommerce benefits from the enormous WordPress ecosystem: there are thousands of extensions, themes, payment integrations, shipping integrations, and developer resources. Current 2026 comparisons continue to identify it as having the broadest open-source ecommerce ecosystem. Toolradar OSSAlt

Its development is also quite transparent: the WooCommerce GitHub monorepo contains the core plugins, packages, and tools, with public issues and contributions. GitHub

The main downside is that the ecosystem's size can also mean plugin conflicts, inconsistent quality, security/maintenance burdens, and upgrade headaches. A recent 2026 security incident involving a WooCommerce extension illustrates why extension vetting and timely updates matter. TechRadar

The important alternative: Magento Open Source

If by "plugin ecosystem" you mean deep, sophisticated extension architecture for developers, rather than simply the largest number of plugins, Magento Open Source deserves serious consideration.

Adobe says its marketplace has thousands of extensions, and its extension ecosystem includes formal quality checks. experienceleague.adobe.com commercedeveloper.adobe.com Magento also has a substantial developer community, documentation, GitHub activity, forums, and Stack Exchange resources. business.adobe.com developer.adobe.com

The tradeoff is substantially greater complexity. Magento makes more sense when you're building a large catalog, complex pricing/catalog rules, multiple stores, B2B functionality, or highly customized commerce workflows.

If you're a modern JS/TS developer

I'd look at Medusa or Vendure instead. They're much more developer-oriented and API/headless by design, but their extension ecosystems are nowhere near WooCommerce's maturity and breadth. Current 2026 comparisons describe Medusa and Vendure as particularly strong options for TypeScript/Node teams.

For a low-traffic boutique store with a non-technical owner, I’d generally look at WooCommerce first among self-hosted options—but the hosting setup matters almost as much as the software.

Practical comparison

Why WooCommerce: the day-to-day interface is familiar to many small-business owners, and there is a large ecosystem of themes, payment integrations, and hosting providers. Its downside is that you're maintaining WordPress + WooCommerce + themes + plugins, so updates and compatibility can become the main maintenance issue. WooCommerce itself recommends regular updates, backups, and ideally testing updates on staging. WooCommerce WooCommerce

PrestaShop is more purpose-built for ecommerce, but its current requirements are somewhat more involved—for example, PrestaShop 9 recommends PHP 8.4 and at least 512 MB of PHP memory. PrestaShop Help Center PrestaShop

The important distinction

If by self-hosted you mean "I want control over the server and software, but I don't want to be a Linux administrator," I'd use:

WooCommerce + managed WordPress hosting + automatic backups + a very small plugin set.

That gives the owner a relatively simple admin experience while the host handles much of the server-level work. WooCommerce's own documentation explicitly recommends contacting the hosting provider for things such as PHP upgrades. WooCommerce

If you mean "I will personally administer the VPS/server," then I would not call any of these particularly easy for a non-technical owner. The server, PHP, database, HTTPS, backups, security updates, and application updates all become your responsibility.

For a boutique with, say, 20–500 products and tens to a few hundred orders per month, I would keep the stack deliberately boring: WooCommerce, one lightweight theme, one payment gateway, shipping integration, automated backups, and as few plugins as possible. That minimizes the most common WooCommerce troubleshooting source: conflicts between outdated software, themes, and plugins. WooCommerce

If you tell me your budget and whether you're comfortable managing a VPS, I can narrow this down to a specific self-hosted stack.

For a low-traffic boutique store with a non-technical owner, I’d narrow the self-hosted choices to WooCommerce and PrestaShop.

My practical take

WooCommerce is probably the easiest choice. It's built on WordPress, has a large ecosystem of themes/plugins, and routine theme updates can be performed from the WordPress interface. WooCommerce Open Business Network

The important caveat is that self-hosted WooCommerce isn't maintenance-free. You still need hosting, backups, WordPress/WooCommerce/plugin updates, and compatibility management. WooCommerce is also currently considering raising its minimum PHP version to 8.1 for version 11.5, targeted for January 2027, illustrating why keeping the underlying hosting environment current matters. The WooCommerce Developer Blog

PrestaShop is a more dedicated ecommerce application, but that doesn't necessarily make it easier for a non-technical owner. Its own maintenance guidance recommends checking module/theme compatibility, maintaining the database, monitoring logs, and testing updates in a pre-production environment. PrestaShop Help Center

For your specific situation

I'd use managed WordPress hosting + WooCommerce, rather than renting a bare VPS and administering everything yourself.

That gives you roughly:

Hosting provider handles: server, PHP/database environment, backups, SSL, infrastructure
You handle: products, orders, customers, promotions, and occasional software updates

For a boutique with relatively few products and low traffic, there's little reason to introduce the operational complexity of Medusa, Saleor, Magento, or a manually administered VPS. Current comparisons similarly distinguish WooCommerce/PrestaShop as turnkey self-hosted systems, while headless platforms such as Medusa and Saleor are much more developer-oriented. Open Business Network

If you tell me roughly how many products you have (e.g. 20, 200, or 2,000) and whether you need things like subscriptions, appointments, or in-store POS, I can narrow this down further.

If inventory and warehouse management are the priority, I’d narrow the field considerably. My first choice would be ERPNext, especially if you need the system to handle more than just ecommerce inventory.

My shortlist

1. ERPNext — my recommendation if operations matter most

Frappe's ERPNext is unusual because ecommerce is part of a much larger operational system rather than being the entire product.

Its warehouse model can go down to a hierarchy such as:

Warehouse → Room → Row → Shelf → Bin

and ERPNext maintains stock balances by item and warehouse. Frappe Documentation

It also covers purchasing, accounting, sales orders, shipments, returns, manufacturing, and other ERP functions. Its ecommerce functionality can publish ERPNext items to a web store, and its REST API can make ERPNext the inventory/fulfillment backend for a separate ecommerce frontend. Frappe Documentation Frappe Documentation

That makes an architecture like this particularly attractive:

Next.js / custom storefront
↓
ERPNext
├── Inventory
├── Warehouses / bins
├── Purchasing
├── Receiving
├── Fulfillment
├── Returns
├── Accounting
└── Reporting

I'd seriously consider this if you're dealing with multiple warehouses, purchasing/replenishment, receiving, stock transfers, returns, or physical warehouse operations.

2. Saleor — if ecommerce is the center of gravity

Saleor is the one I'd investigate next if you want a modern headless ecommerce platform rather than an ERP.

Its architecture supports warehouse-level inventory, multiple channels, reservations, and fulfillment, making it considerably more interesting for sophisticated ecommerce than a typical plugin-based store. Current comparisons also identify its multi-warehouse capabilities as more mature than Medusa's. Eryx Innovations Lab OSSAlt

The tradeoff is that Saleor is primarily a commerce engine, not a complete warehouse/ERP system.

3. Medusa — excellent if you're building custom

Medusa is another strong option, particularly if your engineering team is TypeScript/Node oriented.

Its inventory module supports inventory items, inventory across multiple locations/warehouses, and inventory reservations. Medusa Docs

I'd choose Medusa when the requirement is essentially:

"We want to build a highly customized ecommerce application and need solid inventory primitives."

I'd lean toward ERPNext or Saleor when the requirement is:

"We need inventory and warehouse operations to be a major part of the business system."

The important distinction

I'd make the decision based on what "warehouse management" means for you.

If you mean:

• Multiple warehouses
• Bin/location management
• Receiving
• Putaway
• Picking
• Packing
• Transfers
• Cycle counts
• Purchase orders
• Replenishment
• Returns
• Inventory valuation
• Barcode scanning
• Warehouse staff workflows

then ERPNext deserves to be at the top of your evaluation. Its warehouse abstraction and broader inventory/ERP functionality go substantially beyond simply maintaining an available_quantity field. Frappe Documentation Frappe Documentation

If you mainly mean:

• Multiple fulfillment locations
• Inventory reservations
• Allocating stock to orders
• Ecommerce fulfillment
• Multiple storefronts/channels

then Saleor is very compelling.

If you tell me your approximate SKU count, number of warehouses, B2B vs B2C, whether you need barcode scanning, and whether you want a headless/custom storefront, I can give you a much more specific architecture recommendation between ERPNext vs Saleor vs Medusa.

and ERPNext maintains stock balances by item and warehouse. Frappe Documentation

It also covers purchasing, accounting, sales orders, shipments, returns, manufacturing, and other ERP functions. Its ecommerce functionality can publish ERPNext items to a web store, and its REST API can make ERPNext the inventory/fulfillment backend for a separate ecommerce frontend. Frappe Documentation Frappe Documentation

That makes an architecture like this particularly attractive:

Next.js / custom storefront
↓
ERPNext
├── Inventory
├── Warehouses / bins
├── Purchasing
├── Receiving
├── Fulfillment
├── Returns
├── Accounting
└── Reporting

I'd seriously consider this if you're dealing with multiple warehouses, purchasing/replenishment, receiving, stock transfers, returns, or physical warehouse operations.

2. Saleor — if ecommerce is the center of gravity

Saleor is the one I'd investigate next if you want a modern headless ecommerce platform rather than an ERP.

Its architecture supports warehouse-level inventory, multiple channels, reservations, and fulfillment, making it considerably more interesting for sophisticated ecommerce than a typical plugin-based store. Current comparisons also identify its multi-warehouse capabilities as more mature than Medusa's. eryxinnovationslab.com ossalt.com

The tradeoff is that Saleor is primarily a commerce engine, not a complete warehouse/ERP system.

3. Medusa — excellent if you're building custom

Medusa is another strong option, particularly if your engineering team is TypeScript/Node oriented.

Its inventory module supports inventory items, inventory across multiple locations/warehouses, and inventory reservations.

I'd choose Medusa when the requirement is essentially:

"We want to build a highly customized ecommerce application and need solid inventory primitives."

I'd lean toward ERPNext or Saleor when the requirement is:

"We need inventory and warehouse operations to be a major part of the business system."

The important distinction

I'd make the decision based on what "warehouse management" means for you.

If you mean:

For strong inventory + warehouse management, I’d prioritize an ERP-backed commerce stack over a pure ecommerce engine.

My shortlist:

My recommendation:

• If warehouse operations are mission-critical: choose Odoo as the operational backbone and connect it to a modern storefront (Odoo website, headless frontend, or another commerce layer). You get purchasing, receiving, put-away workflows, stock moves, replenishment, manufacturing, and accounting in one system.
• If you are building a custom headless commerce product: choose Saleor + an ERP/WMS integration. Saleor has stronger built-in multi-channel and warehouse concepts than many headless alternatives. OSSAlt
• If you want maximum developer flexibility: choose Medusa + Odoo/ERP/WMS rather than trying to make Medusa become your warehouse system. Born Digital

Key warehouse features to verify before committing:

• multiple warehouses and bins
• batch/lot and serial tracking
• barcode scanning
• cycle counts
• pick/pack/ship workflows
• purchase order receiving
• dropship and 3PL support
• inventory reservations and allocation
• returns/RMA handling

For a company selling physical goods at meaningful volume, I’d generally start by evaluating Odoo + a headless storefront versus Saleor + ERP integration.

For a small SaaS store selling software licenses, I’d look at headless/self-hosted commerce platforms rather than traditional ecommerce packages. The key architectural question is whether the ecommerce system should own billing/subscriptions, or whether something like Stripe Billing/Chargebee/Paddle should remain the source of truth.

The main options

Medusa is probably the first one I'd investigate for your particular use case. It has a modular architecture and a native Stripe payment provider, while its documentation explicitly provides a recipe for implementing recurring subscriptions with Stripe, including renewals, expiration, cancellation, and subscription management. docs.medusajs.comdocs.medusajs.com

The important caveat is that Medusa does not natively provide subscription billing as a first-class commerce primitive. Its own documentation describes subscriptions as something you implement on top of the platform. docs.medusajs.comdocs.vendure.ioapps.saleor.io

The architecture I'd consider

For SaaS licenses, I'd separate commerce from billing:

                 ┌──────────────────┐
                 │   Your SaaS      │
                 │   application    │
                 └────────┬─────────┘
                          │
                    license / entitlements
                          │
                 ┌────────▼─────────┐
                 │ License Service  │
                 └────────┬─────────┘
                          │
        ┌─────────────────┴─────────────────┐
        │                                   │
┌───────▼────────┐                  ┌───────▼────────┐
│    Medusa      │                  │ Stripe Billing │
│  product/cart  │                  │ subscriptions  │
│  checkout/order│                  │ invoices       │
└───────┬────────┘                  └───────┬────────┘
        │                                   │
        └───────────────┬───────────────────┘
                        │ webhooks
                 ┌──────▼───────┐
                 │ Your SaaS DB │
                 └──────────────┘

That gives you a useful separation:

• Medusa: products, license SKUs, checkout, orders, coupons, customer/storefront experience.
• Stripe Billing: recurring charges, invoices, payment methods, failed-payment handling, renewals.
• Your application: license issuance, activation, seats, entitlements, expiration, upgrades/downgrades.
• Webhooks: synchronize Stripe → your SaaS and Medusa → your SaaS.

Medusa already has a payment-module abstraction and Stripe provider, so this isn't fighting its architecture. docs.medusajs.com

Why I'd lean away from making ecommerce own subscriptions

For a SaaS company, the actual commercial object isn't really an "order." It's something like:

Customer → subscription → entitlement → license/seats

For example:

Acme Corp
   │
   └── Pro Annual
         │
         ├── Stripe subscription
         ├── 25 seats
         ├── License: ABCD-...
         └── expires: 2027-09-18

That model maps much more naturally onto Stripe Billing + your own entitlement system than trying to turn an ecommerce database into your SaaS billing database.

Where Vendure fits

I'd seriously consider Vendure if your engineering team is TypeScript/NestJS and you want a highly customizable commerce backend. Its plugin architecture is particularly interesting for license provisioning and SaaS-specific workflows. There are already subscription-oriented plugins/integrations in its ecosystem. docs.vendure.io

Where Saleor fits

Saleor is attractive if you're comfortable with Python/GraphQL and want a more platform-oriented commerce system. Its Stripe integration is mature and available for self-hosted deployments. apps.saleor.io

For a small store attached to an existing SaaS, though, I'd be cautious about adopting a larger commerce platform than you actually need.

My shortlist would therefore be:

1. Medusa + Stripe Billing + your license/entitlement service
2. Vendure + Stripe Billing + your license/entitlement service
3. Saleor + Stripe Billing + your license/entitlement service

The deciding factor between the first two is largely Medusa's workflow/module model vs. Vendure's plugin/GraphQL model, rather than basic ecommerce capability. kryeo.com

If you tell me what billing system you already use (Stripe, Chargebee, Paddle, Recurly, etc.) and your stack (Node/Python/PHP/etc.), I can narrow this down considerably and sketch the integration architecture.

For a SaaS company with a small license store, I’d look at Medusa first if you specifically want the ecommerce layer self-hosted. It is headless, API-driven, and can integrate with Stripe for subscription billing, although subscriptions require an integration/plugin rather than being core functionality. Medusa Docs Medusa

• Medusa — best fit for a developer-controlled SaaS storefront; self-hostable and extensible.
• Saleor — another strong headless option, but subscription billing likewise sits in an integration layer. Designodin
• WooCommerce — easier turnkey store, but potentially overkill if your SaaS already has its own account/billing system.
• Lemon Squeezy — worth considering if “self-hosted ecommerce” isn't an absolute requirement: it natively handles subscriptions, software licenses, payments and tax as a merchant of record. Lemon Squeezy Docs LemonSqueezy

Architecture I'd favor: your SaaS remains the source of truth for entitlements/licenses → Stripe Billing (or another billing system) handles recurring charges → Medusa handles the small product/store/catalog experience.

If you tell me whether you use Stripe, Paddle, Chargebee, or something else for billing, I can narrow this to the most natural integration.

If strong security, PCI compliance, and protection of customer data are the primary requirements, I’d narrow the open-source field to Magento Open Source, Shopware, and WooCommerce—but they put different burdens on your security team.

What I'd design for

For a security-sensitive deployment, don't let the ecommerce platform store raw card numbers. Use a PCI-compliant payment provider with hosted checkout/fields or tokenization. That reduces your PCI scope, although it doesn't eliminate your PCI responsibilities. WooCommerce's documentation explicitly makes this distinction. WooCommerce

For customer data, I'd require:

• MFA/strong authentication for administrators
• Least-privilege RBAC
• Encryption in transit and at rest
• Centralized audit logs and monitoring
• Automated vulnerability/dependency scanning
• Rapid security-patch process
• WAF and DDoS protection
• Encrypted, tested backups
• Strict separation of production, staging, and development data
• Data-retention/deletion controls
• A documented PCI DSS responsibility matrix
• Minimal third-party extensions/plugins

The important distinction

If you're specifically asking "which open-source platform gives us the strongest foundation for a serious security/compliance program?", I'd focus the evaluation on Magento Open Source vs. Shopware, rather than choosing based on ecommerce features.

Magento has particularly extensive documented security controls and enterprise-oriented security tooling; its documentation also explicitly covers PCI, GDPR/CCPA, security scanning, authentication and admin security. Experience League Experience League

Shopware is worth serious consideration if you want a somewhat more modern commerce architecture and are comfortable with its ecosystem. Its hosted offering has particularly clear security/compliance attestations, although those shouldn't automatically be treated as certifications of a self-hosted Shopware installation. Shopware

WooCommerce can absolutely be secured, but I'd be more cautious for a highly regulated environment because the security boundary extends across WordPress, WooCommerce, every plugin/theme, and your hosting infrastructure. WooCommerce itself explicitly says the store owner remains responsible for PCI compliance. WooCommerce

One caveat: no open-source ecommerce platform makes your business PCI/GDPR/CCPA compliant by itself. Compliance depends on the complete architecture, payment provider, hosting, configuration, custom code, extensions, processes, and organizational controls. Adobe makes the same point in its compliance documentation. Experience League Experience League

If you tell me your company size, expected transaction volume, cloud/on-prem preference, and whether you need GDPR, SOC 2, HIPAA, or PCI DSS, I can give you a security/compliance architecture comparison of Magento vs. Shopware vs. WooCommerce.

If security, PCI scope, and customer-data protection are primary requirements, I’d shortlist Magento Open Source, Shopware, and WooCommerce—but the right choice depends heavily on how much security engineering you want to own.

My practical recommendation

For a security/compliance-first organization, I'd start with Magento Open Source or Shopware rather than WooCommerce.

Magento Open Source has explicit security capabilities around admin sessions/credentials, CAPTCHA, website restrictions and PCI-oriented payment architecture. Adobe also provides payment services designed to keep PCI-regulated information out of the commerce environment. Experience League Experience League

Shopware is particularly interesting if you want a more managed security program around the platform. Shopware reports an ISO/IEC 27001:2022-certified information-security management system and SOC 2 certification for its organization/services. Its documentation also describes built-in rate limiting against attacks such as brute-force login attempts. Shopware Shopware Documentation

WooCommerce can absolutely be secured, but you inherit WordPress's security model and a potentially large plugin/extension attack surface. WooCommerce itself explicitly notes that the security of a WooCommerce site is effectively dependent on the underlying WordPress installation and its plugins. WooCommerce

One important point about PCI

Don't select a platform simply because it says "PCI compliant." PCI DSS compliance is an attribute of your whole payment environment, not merely the ecommerce software.

For example, WooCommerce recommends using hosted payment fields/gateways so that your infrastructure never handles raw card data, substantially reducing PCI scope. Even then, the merchant retains security and compliance responsibilities. WooCommerce WooCommerce

I'd architect whichever platform you choose roughly like this:

Customer → CDN/WAF → ecommerce application → hosted/tokenized payment provider

with:

• No raw card numbers stored in your database
• MFA and least-privilege administrative access
• Centralized audit logging
• Encryption in transit and at rest
• Automated vulnerability/dependency scanning
• Rapid security-patch process
• WAF/bot/rate-limit protection
• Backups with tested restoration
• Secrets kept outside application configuration
• Data-retention/deletion procedures
• Vendor/plugin security review
• Formal PCI DSS scope assessment
• GDPR/CCPA/privacy controls where applicable

If you tell me your approximate scale (e.g. <$1M, $1–10M, $10M+ annual GMV), cloud/on-prem preference, and whether you need B2B features, I can give you a more specific Magento vs. Shopware vs. WooCommerce comparison focused specifically on PCI DSS, SOC 2, GDPR, IAM, audit logging, vulnerability management, and total security effort.

For a security/compliance-first organization, I'd start with Magento Open Source or Shopware rather than WooCommerce.

Magento Open Source has explicit security capabilities around admin sessions/credentials, CAPTCHA, website restrictions and PCI-oriented payment architecture. Adobe also provides payment services designed to keep PCI-regulated information out of the commerce environment. Experience League Experience League

Shopware is particularly interesting if you want a more managed security program around the platform. Shopware reports an ISO/IEC 27001:2022-certified information-security management system and SOC 2 certification for its organization/services. Its documentation also describes built-in rate limiting against attacks such as brute-force login attempts. Shopware Shopware Documentation

WooCommerce can absolutely be secured, but you inherit WordPress's security model and a potentially large plugin/extension attack surface. WooCommerce itself explicitly notes that the security of a WooCommerce site is effectively dependent on the underlying WordPress installation and its plugins. woocommerce.com

One important point about PCI

Don't select a platform simply because it says "PCI compliant." PCI DSS compliance is an attribute of your whole payment environment, not merely the ecommerce software.

For example, WooCommerce recommends using hosted payment fields/gateways so that your infrastructure never handles raw card data, substantially reducing PCI scope. Even then, the merchant retains security and compliance responsibilities.

I'd architect whichever platform you choose roughly like this:

Customer → CDN/WAF → ecommerce application → hosted/tokenized payment provider

with:

If integration-friendliness is the main criterion—especially connecting an ERP, shipping/fulfillment providers, PIM/WMS, marketplaces, and custom internal systems—I’d shortlist Shopware, Medusa, and Saleor.

My practical take

For a growing conventional store: Shopware is probably the first one I'd investigate.

Its current architecture is explicitly API-first, with APIs, webhooks, apps/plugins, bulk synchronization, and connectors for ERP/PIM/WMS/CRM systems. Shopware also has a substantial marketplace of ERP and shipping integrations. Shopware Shopware Store

For example, Shopware documents ready-made ERP connections for systems such as SAP, Microsoft Dynamics, Xentral, weclapp, and Odoo, rather than requiring every ERP integration to be built from scratch. Shopware Shopware Store

Medusa becomes particularly attractive if you have developers and expect significant custom integration work. Its architecture treats third-party services as modules/providers, and its documentation specifically covers custom ERP integrations and fulfillment providers. It has an existing ShipStation integration, for example, which can expose carriers such as DHL and FedEx through the fulfillment layer. Medusa Docs Medusa Docs Medusa Docs

Saleor is another strong choice if you're building a highly composable/headless stack. Its GraphQL API, apps, and 140+ webhook events are designed around external integrations, and its documentation includes custom shipping and other integration patterns. Saleor Documentation Saleor Commerce

The key distinction

I'd think about it this way:

• "I want lots of existing connectors and a mature merchant platform." → Shopware
• "I have developers and want maximum control over how ERP/shipping/business logic works." → Medusa
• "We're building a sophisticated API-first/composable commerce stack." → Saleor
• "We're already heavily invested in WordPress." → WooCommerce

One interesting development for 2026 is that Shopware has also introduced Nexus, an event-based integration/orchestration layer intended specifically to reduce the amount of point-to-point integration work. Shopware

If you tell me which ERP (e.g. SAP, NetSuite, Odoo, Dynamics, Acumatica) and which shipping providers (UPS/FedEx/DHL/ShipStation/Shippo/etc.), I can compare Shopware vs Medusa vs Saleor specifically for your stack, including where you'd need custom development versus an existing connector.

For example, Shopware documents ready-made ERP connections for systems such as SAP, Microsoft Dynamics, Xentral, weclapp, and Odoo, rather than requiring every ERP integration to be built from scratch. Shopware store.shopware.com

Medusa becomes particularly attractive if you have developers and expect significant custom integration work. Its architecture treats third-party services as modules/providers, and its documentation specifically covers custom ERP integrations and fulfillment providers. It has an existing ShipStation integration, for example, which can expose carriers such as DHL and FedEx through the fulfillment layer.

If ERP + shipping integrations are the priority, I’d shortlist Medusa, Shopware, and Saleor. For a growing store, the choice depends heavily on whether you want a developer-oriented commerce engine or a more complete out-of-the-box platform.

My take

Medusa would be my first platform to investigate if your primary concern is how easily your developers can connect the store to arbitrary external systems. Its architecture explicitly treats third-party integrations as modules, and its documentation provides an ERP integration pattern—including product/order synchronization and custom business logic. It also has fulfillment integrations such as ShipStation. Medusa Docs Medusa Docs Medusa Docs

The important distinction is that Medusa is more of a commerce engine than a turnkey ecommerce package. That's a benefit if you expect your ERP, WMS, 3PL, shipping platform, PIM, etc. to form a larger system around the store.

Shopware is particularly compelling if you want a more complete ecommerce platform while retaining strong integration capabilities. Its current architecture exposes APIs, webhooks, plugins/apps, and bulk synchronization through its Sync API; Shopware also advertises connectors for ERP, PIM, CRM, and shipping systems. Shopware Shopware

Saleor is another strong choice if your engineering team likes GraphQL. Its architecture is explicitly API-first, with apps and webhooks used to connect external services; its current documentation describes 160+ webhooks and independently versioned apps. Saleor Commerce GitHub

What I'd choose

For a store that expects something like:

Store → ERP → WMS/3PL → shipping carriers → marketplaces/PIM/CRM

I'd lean toward:

1. Medusa — if you're building a custom/headless system and have a strong development team.
2. Shopware — if you want more ecommerce functionality out of the box alongside sophisticated integrations.
3. Saleor — if GraphQL and a highly composable architecture are particularly attractive to your team.

One particularly useful Medusa characteristic is that it can act as an integration/middleware layer, rather than simply being the system that owns the storefront. Its framework is designed to orchestrate operations across external systems and provides durable workflows for those operations. Medusa Docs

If you tell me which ERP (e.g. SAP, NetSuite, Odoo, Microsoft Dynamics), shipping providers/3PLs, monthly order volume, and whether you need B2B, I can narrow this down substantially and compare the actual integration architectures.