Answers that do not mention BigCommerce

For a small handmade-goods shop on a tight budget, I’d pick WooCommerce.

woocommerce.com is open source, self-hostable, free to download, and gives you ownership of the store and its data. You can choose your own hosting and add functionality through its huge ecosystem of extensions. WooCommerce

My shortlist

Why WooCommerce fits you

For handmade goods, you probably need relatively straightforward things: products with variations, inventory, photos, shipping, coupons, payments, customer accounts and perhaps a blog. WooCommerce already handles physical products and has extensive extension/customization capabilities. WooCommerce

The important distinction is software cost vs. operating cost. Self-hosting doesn't make the store completely free—you'll still pay for hosting, your domain, payment processing, backups, and potentially premium extensions. But you aren't paying a platform subscription just to keep your store running.

PrestaShop would be my second choice if you don't want WordPress at all. Its open-source edition is freely downloadable and designed specifically as a standalone ecommerce platform, with catalog, inventory, orders, shipping and payment functionality built in. PrestaShop PrestaShop

I wouldn't start with Medusa or Saleor for this use case. They're excellent if you're a developer who wants to build a custom/headless commerce system, but that flexibility also means substantially more engineering and maintenance. WP Bundle WP Bundle

Bottom line: start with self-hosted WooCommerce + inexpensive managed/shared WordPress hosting. Keep the number of plugins small, use automatic backups, and avoid expensive “all-in-one” plugin bundles until you actually need their features.

If you tell me roughly how many products you have, whether you need Etsy/Instagram integration, and your monthly hosting budget (e.g. $5/$10/$20), I can give you a very lean self-hosted stack and estimated monthly cost.

For a 10-person startup building a high-traffic, headless storefront, I'd shortlist Medusa, Vendure, and Saleor. I would not start with WooCommerce/Magento unless you have a strong reason to inherit their ecosystems.

My pick: Medusa

I'd choose Medusa for most startups in your situation.

Why:

• TypeScript/Node.js — particularly attractive if your storefront is Next.js/React and your team is already JS/TS-heavy.
• Headless-first architecture rather than adapting a traditional monolith into headless.
• Modular/workflow architecture gives you room to customize checkout, pricing, promotions, fulfillment, etc. without immediately forking the platform.
• PostgreSQL + Redis-friendly architecture, so you can scale the stateless application tier horizontally.
• MIT-licensed core, which is a meaningful advantage for a startup that doesn't want licensing surprises.
• Its ecosystem/community is currently larger than Vendure's, which matters when a 10-person team hits an obscure production problem. Current comparisons put Medusa at roughly 36k GitHub stars versus ~8k for Vendure and ~23k for Saleor. Owerczuk Kryeo

The important caveat: don't interpret "high traffic" as "the commerce backend needs to serve every page request." I'd architect it roughly:

                    ┌───────────────┐
                    │    CDN / WAF  │
                    └───────┬───────┘
                            │
                    ┌───────▼───────┐
                    │ Next.js /     │
                    │ Storefront    │
                    └───────┬───────┘
                            │
                ┌───────────▼───────────┐
                │   Commerce API        │
                │   Medusa              │
                └──────┬─────────┬──────┘
                       │         │
                 ┌─────▼───┐ ┌──▼──────┐
                 │Postgres │ │  Redis  │
                 └─────────┘ └─────────┘
                       │
             ┌─────────▼──────────┐
             │ Payments / ERP /   │
             │ PIM / fulfillment  │
             └────────────────────┘

Cache/catalog/search aggressively at the edge. Keep checkout/order mutations strongly consistent. Don't make your commerce database the bottleneck for anonymous product browsing.

The shortlist

Vendure is particularly compelling if your team loves GraphQL and NestJS. Its plugin architecture is clean, and it's a strong choice for sophisticated multi-channel/multi-brand setups. The major thing I'd investigate with your legal team is its GPLv3 licensing model. Owerczuk Vendure

Saleor is the one I'd seriously consider if you have Python expertise or unusually complex enterprise commerce requirements. Its GraphQL-first approach and built-in capabilities around channels, warehouses, currencies, etc. are excellent. But for a 10-person predominantly TypeScript startup, running a Python/Django commerce backend introduces another technology stack to own. Kryeo Grafen

What I'd avoid

I wouldn't pick based primarily on benchmark claims like "33ms product queries." At your scale, your architecture will matter vastly more than raw framework benchmarks. CDN caching, database indexes, search architecture, connection pooling, queueing, observability, and how you handle flash-sale/traffic-spike behavior will determine whether the system survives 10× traffic.

I also wouldn't build the commerce backend yourself. A 10-person startup can burn an enormous amount of engineering capacity recreating:

• carts
• promotions
• tax
• inventory reservations
• order state machines
• payment workflows
• refunds
• customer accounts
• multi-region pricing
• fulfillment
• admin tooling

That's precisely the part I'd outsource to an existing commerce engine.

Bottom line: if you're a modern TS/React startup, I'd start with Medusa, prototype your three hardest commerce workflows, and load-test the architecture before committing. If you're GraphQL/NestJS-heavy, put Vendure through the same exercise. If you're Python-heavy or have complex global/B2B commerce, evaluate Saleor.

If you tell me roughly your expected peak requests/sec, catalog size, countries/currencies, B2C vs B2B, and whether you're using Next.js, I can give you a much more concrete architecture and tell you which of the three I'd choose.

Saleor is the one I'd seriously consider if you have Python expertise or unusually complex enterprise commerce requirements. Its GraphQL-first approach and built-in capabilities around channels, warehouses, currencies, etc. are excellent. But for a 10-person predominantly TypeScript startup, running a Python/Django commerce backend introduces another technology stack to own. Kryeo grafen.tn

What I'd avoid

I wouldn't pick based primarily on benchmark claims like "33ms product queries." At your scale, your architecture will matter vastly more than raw framework benchmarks. CDN caching, database indexes, search architecture, connection pooling, queueing, observability, and how you handle flash-sale/traffic-spike behavior will determine whether the system survives 10× traffic.

If your two non-negotiables are multi-vendor marketplace functionality and full ownership/control of your data and infrastructure, I’d shortlist Bagisto, Vendure, and Medusa—but they fit different situations.

My recommendation

For most companies wanting a true marketplace rather than a commerce engine they must heavily customize, I’d start with Bagisto.

It has a dedicated multi-vendor marketplace layer with seller accounts/panels, product approval, commissions, seller management, and multi-seller carts. Because it is self-hostable, your database and infrastructure can remain under your control. Bagisto Documentation Bagisto

bagisto.com

Why I'd put Bagisto first

Its marketplace functionality is much closer to what you're asking for out of the box. The marketplace documentation specifically covers seller management, seller panels, product approval, commissions, and customers purchasing across multiple sellers in one cart. Bagisto Documentation

It also uses the Laravel ecosystem, which can be a major advantage if your engineering team already knows PHP/Laravel.

When I'd choose Vendure instead

Choose Vendure if you're building a sophisticated, API-first platform and have a strong TypeScript/NestJS team.

Vendure has an explicit multi-vendor marketplace architecture and example plugin covering multiple sellers, seller-specific products, and splitting customer orders between vendors. Vendure

docs.vendure.io

The trade-off is that you'll be doing more engineering around the marketplace model rather than simply configuring a marketplace product.

When I'd choose Medusa

Medusa is particularly attractive if your marketplace is going to be highly differentiated.

Its own documentation shows how to build vendors, vendor-admin accounts, vendor products/orders, and split orders, but explicitly describes marketplace functionality as something you build through customization rather than native functionality. Medusa Docs

docs.medusajs.com

The upside is tremendous architectural flexibility. The downside is that your company owns more of the implementation burden.

The key distinction

If by "total data ownership" you mean:

• Database hosted in your AWS/Azure/GCP account
• Full database access
• Source code available
• Ability to self-host
• No dependency on a SaaS marketplace provider
• Ability to migrate away later
• Vendor/customer/order data under your control

then all three can satisfy the architectural requirement. But open source alone doesn't guarantee data sovereignty—you should also examine third-party payment, analytics, search, email, tax, and shipping services.

My ranking for your stated requirements:

1. 🥇 Bagisto — best if you want multi-vendor functionality relatively ready to use.
2. 🥈 Vendure — best for a sophisticated headless/enterprise architecture.
3. 🥉 Medusa — best if your engineering team wants maximum customization and is willing to build the marketplace layer.

If you tell me your expected number of vendors/products, B2B vs B2C, tech stack (PHP/Node/Python), and whether vendors need their own payouts, I can narrow this down to a specific architecture and recommend the best option.

Why I'd put Bagisto first

Its marketplace functionality is much closer to what you're asking for out of the box. The marketplace documentation specifically covers seller management, seller panels, product approval, commissions, and customers purchasing across multiple sellers in one cart. Bagisto Documentation

It also uses the Laravel ecosystem, which can be a major advantage if your engineering team already knows PHP/Laravel.

When I'd choose Vendure instead

Choose Vendure if you're building a sophisticated, API-first platform and have a strong TypeScript/NestJS team.

Vendure has an explicit multi-vendor marketplace architecture and example plugin covering multiple sellers, seller-specific products, and splitting customer orders between vendors. docs.vendure.io

docs.vendure.io

The trade-off is that you'll be doing more engineering around the marketplace model rather than simply configuring a marketplace product.

When I'd choose Medusa

Medusa is particularly attractive if your marketplace is going to be highly differentiated.

Its own documentation shows how to build vendors, vendor-admin accounts, vendor products/orders, and split orders, but explicitly describes marketplace functionality as something you build through customization rather than native functionality. docs.medusajs.com

For a solo founder with limited development skills, I’d narrow it to Vendure, Medusa, and Saleor—and my pick would be Vendure if modern admin + easy theming are your top priorities.

🥇 I'd choose Vendure

Vendure has a particularly good combination for your situation.

• Modern admin: its current dashboard has a modern theming system built around CSS variables, Tailwind and shadcn-style theming. You can change brand colors and other visual properties without rebuilding the whole admin. Vendure
• TypeScript: if you eventually need to customize things, you're staying in the relatively approachable JS/TS ecosystem.
• Official Next.js storefront starter: it includes product listings, product pages, search, cart, checkout and accounts, with Tailwind styling already set up. Vendure
• Good extension model: you can add functionality through plugins rather than hacking the core.
• You aren't starting with a blank admin: this matters enormously for a solo founder.

The big caveat is that it's still headless commerce. You're assembling the storefront rather than getting a Shopify-like visual site builder.

🥈 Medusa if storefront design is your #1 priority

Medusa is arguably the better choice if you have enough React/Next.js knowledge to build the frontend.

Its architecture is very developer-friendly for modern web stores, and the ecosystem is heavily oriented around Next.js/TypeScript. Current comparisons generally position Medusa as the strongest choice for a JavaScript/TypeScript-native team. ShipGarden OSSAlt

But there's an important distinction for you:

Medusa gives you a great commerce engine; Vendure gives you a great commerce engine plus a more complete admin experience.

For a solo founder, the latter can save a surprising amount of work.

🥉 Saleor

Saleor has an excellent admin and is extremely capable. Its dashboard is extensible, with 45+ extension mount points, and its official stack includes a Next.js/React/Tailwind storefront. Saleor Commerce GitHub

I'd choose it if you anticipate things like:

• multiple sales channels
• sophisticated B2B
• international/multi-region commerce
• a larger engineering team later

But for "I'm one person and I want to get a store live", I'd rather not introduce Saleor's Python/Django + GraphQL stack unless there's a specific reason to.

My recommendation

If your requirements are:

solo founder + limited coding + modern-looking admin + easy brand customization + modern storefront

I'd rank them:

1. Vendure → best overall fit
2. Medusa → best for a React/Next.js developer
3. Saleor → best for sophisticated/enterprise commerce
4. WooCommerce → best if minimizing development altogether is more important than having a modern architecture

One important caveat: none of these is really "Shopify but open source." If by easy theming you mean "I want to change the entire storefront's appearance using templates/CSS without touching much code," WooCommerce may actually be the better choice despite its older architecture.

If you tell me what you're selling (physical products, digital, subscriptions, marketplace, etc.) and whether you want to use Next.js, I can give you a much more definitive choice and a lowest-effort stack for getting to launch.

Yes. If the requirement is “open-source commerce backend + API-first + static/headless frontend”, I’d shortlist Medusa, Saleor, and Vendure. They’re all designed for headless use rather than being traditional storefronts with an API bolted on. Vendure

My recommendation: Medusa

docs.medusajs.com

For a typical static site built with Astro, Next.js, Nuxt, plain React, etc., I’d choose Medusa first.

• REST API — Medusa's Store API is explicitly intended for building custom storefronts and mobile apps. Medusa Docs
• TypeScript/Node.js — particularly nice if your frontend is also JavaScript/TypeScript.
• Headless by design — products, carts, customers, orders, payments, shipping, etc. live in the backend.
• Extensible — you can add your own commerce logic without turning the frontend into a monolith.
• MIT licensed, which is attractive if you're building a commercial product. Kryeo
• You can self-host it and put your static site on a CDN independently.

A simple architecture would be:

                    ┌──────────────────┐
                    │   Static Site    │
                    │ Astro / Next /   │
                    │ React / etc.     │
                    └────────┬─────────┘
                             │
                         REST API
                             │
                    ┌────────▼─────────┐
                    │     Medusa       │
                    │ Commerce Backend │
                    └────┬────┬────┬───┘
                         │    │    │
                    PostgreSQL Redis Payments

The alternatives

Saleor is particularly compelling if you specifically want GraphQL. Its documentation describes it as a GraphQL-native, API-only commerce platform, with strong support for multi-channel commerce, checkout, payments, and internationalization. GitHub Saleor Documentation

docs.saleor.io

Vendure is my other strong recommendation if you're a TypeScript team. Its Shop API is GraphQL and exposes products, collections, customers, carts, checkout, and accounts; its plugin architecture also makes extending the API unusually clean. Vendure Vendure

docs.vendure.io

The important caveat with Vendure is its current GPLv3 core license, versus Medusa's MIT license and Saleor's BSD-3-Clause core. GitHub Kryeo

What I'd pick

For most new static-site projects: Medusa.

Pick Saleor instead if:

• GraphQL is a hard requirement.
• You need sophisticated multi-channel/B2B functionality.
• Your team is comfortable with Python/Django.

Pick Vendure instead if:

• Your team is heavily TypeScript/NestJS oriented.
• You want GraphQL.
• You expect to build substantial custom commerce functionality through plugins.

If you tell me what your static site is built with (Astro, Next.js, Hugo, plain HTML, etc.) and whether you need Stripe, subscriptions, variants, inventory, or multi-vendor, I can narrow this down to the best architecture and API for your specific stack.

docs.saleor.io

Vendure is my other strong recommendation if you're a TypeScript team. Its Shop API is GraphQL and exposes products, collections, customers, carts, checkout, and accounts; its plugin architecture also makes extending the API unusually clean. docs.vendure.io docs.vendure.io

For a subscription-box business, I’d shortlist WooCommerce, Medusa, and Saleor, but my recommendation depends heavily on whether you want a ready-to-run store or a custom application.

My pick: WooCommerce

If your priority is getting recurring physical-product billing working reliably with minimal custom engineering, I’d choose WooCommerce + WooCommerce Subscriptions.

WooCommerce Subscriptions already supports recurring schedules, automatic recurring payments through many gateways, failed-payment retries, customer self-service, upgrades/downgrades, trials, and subscription management. It specifically supports subscription-box/product-of-the-month use cases. WooCommerce

The catch: WooCommerce Subscriptions is a commercial extension, so the ecommerce platform is open source/self-hostable, but the subscription functionality isn't entirely free/open source.

woocommerce.com

If you want everything to be custom: Medusa

I'd choose Medusa if you're building a proper application rather than a conventional WordPress store.

It's a headless TypeScript/Node commerce engine designed to be customized, and it is self-hostable. GitHub

However, there's an important catch for your use case: subscriptions aren't native. Medusa's own documentation describes implementing subscription logic yourself, including renewal jobs, expiration handling, subscription management, and payment integration. Alternatively, you can delegate recurring billing to Stripe. Medusa Docs Medusa Docs

That's actually attractive if your subscription box has unusual rules such as:

• "Skip next box"
• Pause/resume
• Change box contents before renewal
• Seasonal boxes
• Different billing and shipping dates
• Build-your-own box
• Proration
• Multiple shipments per billing cycle
• Inventory-aware renewal
• Failed-payment recovery

You can model those explicitly instead of fighting a plugin's assumptions.

docs.medusajs.com

Saleor

Saleor is my third choice. It's an excellent headless, GraphQL-first commerce platform and is designed for composable/custom commerce. It is also genuinely self-hostable. Saleor Saleor

I'd pick it over Medusa if you already have a Python/Django + GraphQL team or expect complex multi-channel/enterprise commerce.

For a small-to-medium subscription-box startup, though, I think Saleor introduces more architecture than you need.

What I'd actually build

If I were starting your business today: I'd use Medusa + Stripe Billing + PostgreSQL + a Next.js storefront if engineering flexibility is important. Let Stripe handle the recurring-payment machinery, while Medusa remains the source of truth for products, orders, inventory, and fulfillment. Medusa's own documentation explicitly presents Stripe as an option for delegating recurring payment handling. Medusa Docs

If you don't have a strong engineering team, I'd go WooCommerce + Subscriptions instead. It will get you to a working subscription box considerably faster.

If you tell me roughly (1) expected subscribers, (2) whether customers choose the contents of each box, and (3) Stripe vs other payment processors, I can recommend the architecture I'd use and where I'd put subscription state, billing, inventory, and fulfillment.

For an enterprise team prioritizing scalability and containerized deployment, I’d recommend Saleor.

Why Saleor

• Built for scale: Saleor is designed around an API-first, service-oriented architecture and supports multi-channel commerce from a single backend. Its production platform currently reports handling 1B+ API requests/month in its cloud environment. Saleor Commerce Saleor Commerce
• Excellent container fit: Its architecture separates the API, workers, dashboard, database, and supporting services, making it well suited to Docker/Kubernetes-style deployments.
• Enterprise-friendly architecture: GraphQL-native APIs, granular permissions, webhooks, extensibility, and technology-agnostic integrations make it easier to integrate with existing enterprise systems. Saleor Commerce Saleor Commerce
• Multi-store/multi-channel: Particularly useful if you're expecting multiple brands, regions, storefronts, currencies, or sales channels. Saleor Commerce
• Open source: The core is openly available and self-hostable, so you retain control over the infrastructure and data. GitHub

How I'd shortlist the alternatives

Bottom line: If you're starting a new enterprise platform and your architecture is microservices/composable + Docker/Kubernetes + headless, Saleor is probably the strongest default choice. If your organization is already heavily invested in Symfony/PHP, I'd put Sylius ahead of it.

If you tell me your cloud (AWS/Azure/GCP), expected traffic/orders, and backend language, I can narrow this down to a specific architecture and deployment choice.

If you need one self-hosted open-source ecommerce system that handles a mixed catalog (physical goods + digital downloads/licenses) well, I’d shortlist these:

Open Source Groups OpenSourceProjects.cc## My recommendations by scenario

1. You want to run the store yourself with minimal development

Pick WooCommerce.

It is usually the easiest answer for a mixed catalog:

• physical SKUs with stock tracking
• downloadable files
• memberships/subscriptions
• product variations
• shipping rules
• taxes
• a huge ecosystem of extensions

The downside is that a large WooCommerce stack can become plugin-heavy, so careful extension selection matters. Open Source Groups

2. You expect a serious catalog or custom workflows

Pick Spree Commerce.

It is designed more like a commerce engine than a website plugin. It can handle:

• physical products with inventory/shipping workflows
• digital products such as downloads, licenses, and subscriptions
• custom checkout logic
• API-driven storefronts

This is a better foundation if you have developers and expect unusual business rules. Spree Commerce Spree Commerce

3. You want a Shopify-like backend but your own frontend

Look at Medusa or Saleor.

These are good when you want:

• a custom React/Next.js storefront
• mobile apps
• multiple sales channels
• deeper integrations

They are less ideal if you want to install today and start adding products from an admin panel. OpenSourceProjects.cc Toolradar

A practical ranking for your use case

1. WooCommerce — best overall for a merchant selling both physical and digital products
2. Spree Commerce — best long-term architecture if you have developers
3. PrestaShop — best traditional standalone ecommerce option
4. Medusa — best if you want a custom/headless build
5. Saleor — best for larger engineering teams

A key design question is how your digital products work:

• simple files (PDFs, ZIPs, videos) → WooCommerce is usually enough
• license keys/software activation → Spree, Medusa, or a customized WooCommerce setup may fit better
• subscriptions/memberships → compare those workflows specifically, since they often need extra modules

If you share your approximate catalog size (e.g., 50 vs 50,000 products), tech preference (PHP/WordPress vs Node/Python), and whether digital items are downloads, licenses, or subscriptions, the shortlist can be narrowed considerably.

If SEO + fast page loads + structured data are your top priorities, I’d shortlist Saleor, Medusa, and Shopware—but for different reasons.

My ranking

🥇 I'd pick Saleor for a new performance-first store

Saleor is headless/API-native, and its current storefront architecture is explicitly designed around caching: product and category pages can use cached/static shells while dynamic portions stream in separately. Saleor GitHub

It also has strong built-in SEO fundamentals, including XML sitemaps, robots handling, and JSON-LD structured data for products/homepages. Saleor Documentation

That combination makes Saleor + Next.js + CDN/edge caching particularly attractive if you're targeting excellent Core Web Vitals.

🥈 Medusa if your team is JavaScript-heavy

Medusa is a very good alternative if you want TypeScript/Node throughout the stack. Its headless architecture lets you control the frontend completely, so your Next.js implementation—not the commerce backend—can be optimized for LCP, INP, caching, image delivery, etc. OpenReplay Blog WP Bundle

The catch is that you should expect to own more of the SEO layer yourself than with a platform that treats SEO as a first-class storefront feature.

🥉 Shopware if SEO features matter more than having the leanest architecture

Shopware is unusually strong on built-in SEO: SEO-friendly URLs, metadata, XML sitemaps, hreflang, canonicals, redirects, and performance/Core Web Vitals optimizations are all part of its SEO offering. Shopware

So if you don't necessarily need a pure headless architecture, Shopware can be a compelling choice.

One important distinction

Don't choose based solely on which ecommerce backend is "fastest."

For ecommerce, your storefront architecture usually has a bigger impact on page speed than the commerce engine. A well-built Next.js storefront with static generation/ISR, aggressive CDN caching, optimized images, minimal JavaScript, and server-side JSON-LD can dramatically outperform a poorly optimized implementation of the same platform. WP Bundle

If I were starting from scratch today:

Saleor + Next.js + CDN/edge caching + JSON-LD + optimized image pipeline

would be my first choice for a store where organic search and Core Web Vitals are unusually important.

If you tell me your catalog size, expected traffic, and whether you want headless/Next.js, I can narrow this down to Saleor vs Medusa vs Shopware for your specific architecture.

For a low-traffic boutique store run by a non-technical owner, I’d pick WooCommerce—but only if you use managed WordPress hosting rather than maintaining the server yourself.

My ranking

1. WooCommerce — easiest overall

• Huge ecosystem and lots of hosting providers can handle the server side.
• The owner can manage products, orders, discounts, etc. through WordPress.
• Lots of tutorials and third-party help available.
• The downside is plugin/theme maintenance: WooCommerce itself recommends backups, staging, and testing updates because conflicts between plugins, themes, payment gateways, and WooCommerce are a common source of problems. WooCommerce WooCommerce
• PrestaShop — good dedicated-store alternative

• More purpose-built for ecommerce than WooCommerce.
• A small store can run quite comfortably on it.
• Hosting partners offer one-click/automatic installation, which reduces setup complexity. PrestaShop Help Center
• However, I'd put it below WooCommerce for a non-technical owner because version upgrades and server requirements can require technical assistance. PrestaShop itself recommends technical expertise for some major migrations. PrestaShop Help Center
• Shopware — excellent, but overkill here

• Powerful and very capable, but self-hosted Shopware means you're responsible for the infrastructure and software management. Shopware explicitly says its self-hosted version requires a degree of technical know-how. Shopware Documentation
• I'd choose it for a more sophisticated operation, not a tiny boutique.

The important distinction

If by "self-hosted" you mean "I want to own/control the store software and have a hosting company run the server":

→ WooCommerce is my recommendation.

If you mean "I personally want to maintain the Linux server, PHP, database, backups, security, and updates":

→ I wouldn't recommend any of these to a non-technical owner. The server administration becomes more difficult than the ecommerce software itself.

For your particular scenario, I'd aim for:

WooCommerce + managed WordPress hosting + automatic backups + minimal plugins + a lightweight theme.

That gives you the ownership/control benefits of self-hosting without turning the boutique owner into a part-time sysadmin. WooCommerce's own documentation also points users toward hosting providers for server configuration issues. WooCommerce

If you tell me roughly how many products you have and whether you need things like variants, subscriptions, appointments, or local pickup, I can narrow this down to the simplest setup.

If strong inventory and warehouse management is the priority, I would not start with a pure ecommerce platform. Most ecommerce engines are good at products, carts, orders, and checkout, but warehouse operations (bin locations, receiving, picking, packing, replenishment, lot/serial tracking, purchasing, manufacturing) are usually ERP/WMS territory.

My recommendations would be:

1. ERPNext + ecommerce frontend — my top pick for inventory-first businesses

Best fit:

• Multiple warehouses
• Real stock control
• Purchasing and supplier workflows
• Manufacturing or assembly
• Barcode operations
• B2B + B2C
• Companies that want one operational system

Why I like it:

• Inventory is a core module, not an add-on.
• Includes warehouse locations, stock movements, purchasing, sales, accounting, and manufacturing.
• Fully open source (GPL), so you are not forced into a commercial edition for core ERP capabilities. Flectic ERP Research

Typical architecture:

Next.js / React storefront
          |
          |
      ERPNext
          |
 Inventory + WMS + Purchasing + Accounting

Tradeoff:

• The ecommerce experience is not as polished as dedicated commerce platforms.
• You may want a separate storefront.

2. Odoo — strongest all-in-one ecosystem

Best fit:

• You want ERP + ecommerce + warehouse in one product.
• You need many business modules.
• You expect to hire consultants or integrators.

Strengths:

• Very mature inventory workflows.
• Multi-location warehouses.
• Putaway rules.
• Barcode scanning.
• Picking strategies.
• Purchasing and manufacturing integrations. Flectic ERP Research

Tradeoffs:

• The truly open-source Community edition has fewer features than Enterprise.
• Many advanced warehouse features are where Odoo's commercial offering becomes attractive. ERP Research

Good architecture:

Odoo (ERP + Inventory)
        |
        |
WooCommerce / Magento / custom storefront

3. Saleor — if you want a modern ecommerce engine and integrate a WMS

Best fit:

• Large catalogs
• Headless commerce
• Multiple sales channels
• Engineering-heavy teams

Strengths:

• Excellent API-first design.
• Strong multi-channel commerce model.
• Better suited than traditional carts for custom applications. Toolradar OSSAlt

But:

• I would not make Saleor your warehouse system.
• Pair it with ERPNext, Odoo, or a dedicated WMS.

4. Medusa — great developer stack, weaker as a warehouse backbone

Best fit:

• Custom storefronts.
• Subscription businesses.
• Marketplace-like experiences.
• Teams using TypeScript/React.

Strengths:

• Modern modular architecture.
• Good for building custom commerce flows. Pawel Owerczuk WP Bundle

Weakness:

• You will likely end up integrating an ERP/WMS once warehouse complexity grows.

Common stack:

Medusa
   |
ERPNext / Odoo
   |
Warehouse operations

Platforms I would avoid if warehouse depth is your main requirement

Open Source Groups Velocity

My shortlist by scenario

If I were building a new ecommerce operation where warehouse accuracy matters more than flashy storefront features, I would choose:

ERPNext as the system of record + a modern frontend (Next.js) on top.

That gives you a real inventory engine instead of trying to turn a shopping cart into a warehouse system.

Yes. For a SaaS company selling a small number of software licenses, I’d avoid choosing a traditional ecommerce platform based primarily on its checkout features. The more important question is: can it cleanly hand off recurring billing, payment state, and license/entitlement provisioning to your existing SaaS system?

My shortlist

1. I'd probably choose Medusa

Medusa is particularly interesting for your use case because you can treat ecommerce as a thin commerce layer around your SaaS billing system, rather than making ecommerce the system of record for subscriptions.

Medusa is headless and TypeScript/Node-based, and its subscription documentation explicitly shows integrating recurring purchases with Stripe while leaving the underlying subscription logic customizable. Medusa Docs WP Bundle

Even more relevant: Medusa now has a digital-products plugin designed specifically around software licenses, entitlements, activations, expiration, and subscription/trial license policies. Medusa

That makes a pretty natural architecture:

                 ┌─────────────────────┐
                 │   Your SaaS app      │
                 │                     │
                 │ users / orgs         │
                 │ subscriptions       │
                 │ entitlements        │
                 │ license activation   │
                 └──────────┬──────────┘
                            │
                     API / webhooks
                            │
                 ┌──────────▼──────────┐
                 │      Medusa         │
                 │                     │
                 │ products            │
                 │ pricing             │
                 │ checkout             │
                 │ orders              │
                 │ coupons             │
                 └──────────┬──────────┘
                            │
                       Stripe/etc.

For a small store, this is preferable to putting your entire SaaS subscription lifecycle inside the ecommerce system.

2. WooCommerce if you want the least engineering

WooCommerce is probably the safest choice if your priority is "get a store running quickly."

Its huge plugin ecosystem makes subscriptions, payments, digital products, invoicing, tax, etc. relatively easy to assemble. WooCommerce also has APIs specifically designed around SaaS billing/subscriptions. The WooCommerce Developer Blog

The downside for a SaaS company is architectural: you can end up with:

WordPress
   ↓
WooCommerce
   ↓
WooCommerce Subscriptions
   ↓
Payment processor
   ↓
Your SaaS

when what you really want may be:

Your SaaS
   ↕
Billing
   ↕
Payment processor

       +

Small ecommerce frontend

I'd pick WooCommerce if the store is going to become a real marketing/content/ecommerce site, rather than merely a place to purchase licenses.

3. Drupal Commerce is surprisingly good for licensing

If license management is a major concern, Drupal + Drupal Commerce deserves consideration.

Drupal Commerce has a Commerce License module specifically designed around selling access to resources. Licenses can have statuses and expiration dates, and the module explicitly works with Commerce Recurring for subscription licenses. Drupal.org

That's a much closer conceptual match to:

Customer buys product → receives entitlement → entitlement expires/renews → SaaS grants access

than a generic ecommerce platform.

I'd consider it particularly if you need things like complex license rules, organizational accounts, content access, or highly customized entitlement workflows.

4. Saleor is excellent commerce infrastructure, but I'd rank it lower here

Saleor is a very capable self-hosted GraphQL commerce engine. It has a clean API, webhooks, and a service-oriented architecture that makes it straightforward to integrate external systems. Saleor Commerce

But subscriptions aren't really its killer feature. You'd be building more of the billing integration yourself.

So I'd choose Saleor when the requirements are more like:

"We are building a sophisticated commerce platform."

rather than:

"We're a SaaS company and need a small shop for licenses."

What I'd actually do for a SaaS

The key architectural decision I'd make is:

Don't make the ecommerce platform the source of truth for SaaS access.

I'd have:

• Stripe Billing / your billing system → source of truth for payment/subscription status
• Your SaaS backend → source of truth for users, organizations, licenses and entitlements
• Medusa → catalog, product pages, cart, checkout, orders
• Webhooks → synchronize purchase/payment/subscription events
• License service → generates/activates/revokes licenses

For example:

                    ┌───────────────┐
                    │    Medusa     │
                    │               │
                    │ Catalog       │
                    │ Storefront    │
                    │ Cart          │
                    │ Checkout      │
                    └───────┬───────┘
                            │
                            ▼
                    ┌───────────────┐
                    │    Stripe     │
                    │               │
                    │ Payments      │
                    │ Subscriptions │
                    │ Invoices      │
                    └───────┬───────┘
                            │ webhooks
                            ▼
                    ┌───────────────┐
                    │ Your SaaS API  │
                    │               │
                    │ Customer      │
                    │ License       │
                    │ Entitlement   │
                    └───────────────┘

That separation is especially valuable when someone cancels, upgrades, downgrades, refunds, or chargebacks. Your application receives the billing event and changes the entitlement accordingly.

My recommendation

If you're a developer-led SaaS with an existing application and billing infrastructure:

1. Medusa — my first choice
2. Drupal Commerce — if licensing/entitlements are unusually complex
3. WooCommerce — if you want the easiest turnkey store
4. Saleor — if you want a highly customizable GraphQL commerce backend

And I would not introduce a full ecommerce platform at all if the store is literally only 5–20 license products. A lightweight Next.js storefront directly on top of Stripe Billing + your existing SaaS API may be substantially simpler.

If you tell me what you're currently using for billing (Stripe, Paddle, Chargebee, Recurly, custom, etc.) and your stack (Next.js, Django, Rails, Laravel, etc.), I can narrow this down to the best 1–2 architectures.

WooCommerce is probably the safest choice if your priority is "get a store running quickly."

Its huge plugin ecosystem makes subscriptions, payments, digital products, invoicing, tax, etc. relatively easy to assemble. WooCommerce also has APIs specifically designed around SaaS billing/subscriptions. The WooCommerce Developer Blog

The downside for a SaaS company is architectural: you can end up with:

WordPress
   ↓
WooCommerce
   ↓
WooCommerce Subscriptions
   ↓
Payment processor
   ↓
Your SaaS

when what you really want may be: