ChatGPT answers, unedited

If low cost + headless flexibility are the priorities, I’d put Medusa at the top of the evaluation list.

My shortlist

Why Medusa: its core is open source, and its current managed hosting starts at $29/month, with 0% GMV/platform fee. The $99/month Launch tier adds things like autoscaling, backups, custom storefront domains, and zero-downtime deployments. medusajs.com

That makes it particularly attractive if your frontend is something like Next.js and you want to keep the commerce backend inexpensive. You can also self-host it if you want to squeeze infrastructure costs further. vendure.io

What I'd evaluate

1. Medusa — first choice
TypeScript/Node, API-first, modular, open source. I'd start here for a small team that wants maximum control without a big SaaS bill. medusajs.com

2. Vendure — strong alternative
Also TypeScript-based and headless, with a particularly good fit if you expect more sophisticated catalog, B2B, or multi-channel requirements. Its core can be self-hosted for free, though its licensing/commercial model deserves review before committing. vendure.io

3. Shopify Hydrogen — evaluate if engineering/ops time is more expensive than platform fees
You get a very capable headless frontend with Shopify remaining the managed commerce backend. It's less attractive if your primary goal is minimizing recurring platform costs and owning the backend. vendure.io

4. Saleor — probably overkill for a tight budget
Technically excellent and very headless/GraphQL-native, but its managed offering currently starts at $1,599/month, making it hard to justify for a cost-sensitive small shop. saleor.io

Bottom line: I'd prototype Medusa + Next.js first. If the requirements turn out to be more enterprise/B2B-heavy, compare it directly with Vendure; if minimizing DevOps is more important than minimizing platform cost, add Shopify Hydrogen to the bake-off.

For a headless shop on a tight budget, I’d put Medusa at the top of the evaluation list.

My recommendation: Medusa

Medusa is particularly attractive if your storefront will be Next.js/React. Its Cloud offering starts at $29/month, with $99/month for the next tier, and it currently advertises 0% GMV/platform fees. You can also self-host the open-source platform if minimizing infrastructure costs is more important than minimizing DevOps work. docs.medusajs.com

That gives you a nice progression:

Prototype → ~$29/mo managed → ~$99/mo production → self-host if/when economics justify it.

It also has the commerce primitives you'd expect—products, variants, pricing, promotions, payments, shipping, returns, etc.—while keeping the storefront decoupled. medusajs.com

Runner-up: Vendure

Vendure is worth evaluating alongside Medusa if you have a strong TypeScript/NestJS team. Its self-hosted runtime has no additional runtime fee, and its commercial platform uses a flat annual subscription rather than GMV/order-based charges. vendure.io

I'd skip Saleor for this particular constraint

Saleor is a very capable headless platform, but its current managed pricing starts at $1,599/month, versus Medusa's $29/month starting point. Saleor's free sandbox is useful for experimentation, but it's not the budget production option you're looking for. saleor.io

Bottom line: I'd run a small proof-of-concept with Medusa + Next.js + your preferred CMS + Stripe/payment provider, and keep Vendure as the second candidate. If the goal is lowest total cost without building the entire commerce backend yourself, Medusa is the strongest starting point.

For a tight budget + headless-friendly shop, I’d evaluate these in roughly this order:

vendure.iohunchbite.com

My shortlist for a budget-conscious team

1. Medusa — probably the first one I’d prototype

• Good match if your frontend team is already using Next.js/TypeScript.
• Open source, API-first, and designed to be customized.
• You can run it cheaply on your own infrastructure instead of paying a large commerce SaaS bill. hunchbite.com
• Good for:
  • DTC brands
  • custom UX
  • marketplaces
  • unusual checkout/business rules

Watch out: it is more of a commerce engine than a complete “set up a store in an afternoon” product. You’ll still need to build the storefront, integrations, and operational workflows. designodin.com

2. Shopify headless — if speed matters more than ownership

• If your goal is “sell products soon,” Shopify is hard to beat.
• Use Shopify as the commerce backend and build your own frontend.
• You avoid running inventory, payments, orders, and security infrastructure yourself.

The downside is that you’re paying for Shopify and accepting platform constraints. ribbsaetersystems.com

3. Vendure — if your team is strongly TypeScript-oriented

• Similar philosophy to Medusa: API-first, customizable, self-hostable.
• Worth evaluating if you want a more structured backend architecture and are comfortable operating it yourself. vendure.io

What I would avoid on a tight budget

• Enterprise composable platforms (commercetools, Elastic Path, etc.) — excellent products, but usually aimed at teams with large implementation budgets. designodin.com
• Building a completely custom commerce backend — headless does not mean you should rebuild carts, inventory, tax, payments, and fulfillment from scratch.

A practical low-cost stack I’d test

Option A: lowest engineering cost

• Shopify backend
• Next.js storefront
• Headless CMS only if needed

Option B: lowest platform cost

• Medusa backend
• Next.js storefront
• PostgreSQL
• Stripe payments
• Managed hosting/VPS

Option C: content-heavy storefront

• Medusa or Shopify
• Sanity / Contentful as CMS
• Next.js frontend

If you tell me your expected scale (e.g., <100 SKUs vs thousands, B2C vs B2B, and whether your team is JS/TS, PHP, or Python), I can narrow this to 1–2 realistic choices.

For a tight-budget, headless-friendly shop, I’d evaluate these in roughly this order:

vendure.io

My shortlist for a budget-conscious build

1. Medusa — probably the first one I’d prototype

• Good match if your frontend team is already using React/Next.js + TypeScript.
• API-first, modular architecture.
• Can be self-hosted to minimize recurring costs, or use managed hosting if you want less DevOps. vendure.io
• Good for: custom UX, marketplaces, unusual checkout flows, content-heavy storefronts.

2. Shopify headless — evaluate if you value launch speed

• A common mistake is assuming headless automatically means open source. Shopify can provide the commerce backend while you build a custom frontend.
• Strong admin experience, payments, apps, and operational tooling.
• Better if your team is small and you want to avoid maintaining carts, inventory, tax, orders, etc. vendure.io

3. Saleor — good if your team is GraphQL/Python-oriented

• Strong API model and good fit for complex catalogs or multi-channel commerce.
• Less attractive if your stack is mostly JavaScript/TypeScript. contracollective.com

What I would avoid on a tight budget

• Enterprise composable platforms like commercetools/Elastic Path: powerful, but usually aimed at organizations with larger budgets and dedicated commerce teams. vendure.io
• Building a fully custom commerce backend from scratch: the storefront is usually the easy part; promotions, inventory, refunds, taxes, and fulfillment become expensive.

A practical low-cost stack to test

• Frontend: Next.js
• Commerce backend: Medusa
• CMS: a lightweight headless CMS (or even Markdown/content files initially)
• Payments: Stripe
• Hosting: managed frontend hosting + low-cost backend hosting

If you have one or two developers, I’d probably start with Medusa vs Shopify headless as a bake-off:

• “Do we need maximum control?” → Medusa
• “Do we need to sell reliably with minimal engineering?” → Shopify headless

The deciding factors are your catalog size, B2B vs DTC, and whether you already have a frontend stack.

For a low-cost, headless-first shop, I’d put Medusa at the top of the evaluation list.

My shortlist

1. Medusa — my first evaluation.
Medusa Cloud currently starts at $29/month, with 0% GMV platform fees, unlimited products/orders/sales channels, and hosting for the backend, admin and storefront. You can also self-host the open-source platform if minimizing recurring costs is more important than minimizing DevOps. docs.medusajs.com

That combination makes it particularly interesting for a tight-budget headless build: Next.js/SvelteKit/etc. on the front, Medusa as the commerce engine, and relatively little platform lock-in. Its Cloud service supports Next.js, SvelteKit and TanStack Start storefronts. docs.medusajs.com

Medusa Cloud pricing

2. Shopify + Hydrogen — evaluate if operational simplicity matters more.
Shopify's Hydrogen storefront framework is free, and Shopify says Hydrogen storefronts can be deployed on Oxygen at no additional hosting cost. You can also use Shopify's Storefront API with your own frontend rather than Hydrogen. apps.shopify.com

It's probably the safer business choice if you want payments, merchandising, apps, admin tooling and operational maturity without building much yourself. The tradeoff is that your commerce backend remains Shopify.

Shopify headless commerce

3. BigCommerce — good SaaS alternative.
Its current Core plan is $29/month when billed annually, with higher tiers as sales grow. It's worth testing if you want a managed commerce backend but don't want to build as much infrastructure around an open-source platform. www.bigcommerce.com

I would skip Saleor for a genuinely tight budget. Its open-source version is attractive, but the current managed Cloud offering starts at $1,599/month, so it's difficult to justify for a small shop unless you specifically need Saleor's architecture. saleor.io

What I'd actually prototype

Medusa + Next.js + inexpensive hosting would be my first POC.

Then build the same small slice—product → cart → checkout → order—on Shopify/Hydrogen. Compare total engineering time, not just monthly SaaS cost. If Medusa saves $100/month but costs you weeks of engineering, Shopify can easily be cheaper overall.

If you tell me roughly monthly orders, number of SKUs, B2C vs B2B, and your preferred frontend (Next.js/Astro/etc.), I can narrow this to a recommended stack and estimated monthly cost.

For a tight-budget, headless-friendly shop, I would evaluate these in roughly this order:

hunchbite.com

My shortlist for a low-cost headless build

1. Medusa — probably the first platform I’d prototype

• Good fit if your frontend is already something like Next.js/React.
• API-first architecture, TypeScript-based, and you avoid a large SaaS platform fee.
• You can run it cheaply on common cloud hosting and add only the services you need.
• The trade-off: you need engineering ownership for integrations, upgrades, and operations. hunchbite.com

2. Shopify + custom frontend — evaluate if speed matters more than cost

• Keep Shopify for products, inventory, payments, checkout, and admin.
• Build a custom frontend using Shopify’s APIs.
• This is attractive if your team is small and you do not want to operate commerce infrastructure. hunchbite.com
• The downside: you may pay Shopify fees while still paying developers to build the headless layer.

3. Saleor — worth a look for more sophisticated needs

• Strong choice if you have Python expertise or need GraphQL-first commerce.
• Better suited when you anticipate multiple storefronts, regions, or complex catalog rules. hunchbite.com

A budget-conscious stack I’d consider

• Frontend: Next.js
• Commerce backend: Medusa
• CMS: Sanity or Payload CMS
• Payments: Stripe
• Hosting: inexpensive managed hosting + managed database

Avoid overbuilding if:

• You have fewer than a few thousand SKUs
• You mostly need standard checkout/cart/inventory
• Your differentiator is marketing/content, not commerce logic

In that case, a headless Shopify frontend may actually be cheaper overall because engineering time dominates infrastructure cost. www.buildwithmatija.com

A useful rule of thumb:

• 1–2 developers + custom UX needs → Medusa
• No dedicated backend engineers → Shopify headless
• Complex catalog/multi-region/B2B → Saleor

If you share your approximate monthly order volume, SKU count, frontend stack (Next.js? mobile app?), and whether you need B2B features, I can narrow it down.

If low cost + headless flexibility are the priorities, I’d evaluate Medusa first, with Vendure as the second option.

Why Medusa: it's MIT-licensed, Node.js/TypeScript, headless-first, and can be self-hosted. That makes the software cost essentially $0; your main expenses become hosting, payments, and engineering time. Current comparisons put it among the strongest open-source choices for JS/TS teams. vendure.io

Vendure is worth a serious bake-off if you're already TypeScript-heavy or expect B2B requirements. Its plugin architecture and GraphQL API are particularly strong. vendure.io

I'd put Saleor behind those for a tight-budget project. It's excellent technically and has a free self-hosted core, but its current managed Cloud production pricing starts at $1,599/month, so it isn't the obvious low-cost route unless you're comfortable operating it yourself. saleor.io

Shopify Hydrogen is the wildcard: probably the cheapest option in engineering time if you want a polished store quickly, but not the cheapest if your goal is to minimize recurring platform costs and own the backend. vendure.io

My shortlist

1. Medusa — best starting point for a genuinely low-cost headless build.
2. Vendure — evaluate alongside it if B2B/complex workflows matter.
3. Shopify + Hydrogen — benchmark against these to see whether paying for managed infrastructure saves enough engineering time.
4. Saleor — consider if GraphQL/Python and more sophisticated commerce requirements justify it.

For a small store with a React/Next.js frontend, I'd probably prototype Medusa + Next.js + Postgres + Stripe first and see how far you can get before committing to a more expensive commerce platform.

For a tight-budget, headless shop, I’d evaluate these in this order:

1. Medusa — my first evaluation

Medusa is probably the one I'd prototype first. It's designed as a headless commerce engine, so you can run your own Next.js/Astro/etc. storefront and keep the commerce backend separate. Its modular architecture gives you catalog, pricing, regions/currencies, promotions, etc., without forcing you into a monolithic frontend. docs.medusajs.com

The big advantage for a budget-conscious team is self-hosting: you can keep infrastructure relatively small and scale it as needed rather than starting with a costly managed-commerce subscription.

Good fit if: you have developer capacity and want maximum control for minimum platform spend.

2. Shopify + Hydrogen — best "cheap to operate"

Shopify is worth serious consideration even though it isn't open source. Hydrogen is free, and Shopify says Oxygen hosting for Hydrogen is included with Shopify plans. apps.shopify.com

That means you can have:

Shopify backend → Storefront API → Hydrogen/React frontend → Oxygen

without having to operate the commerce backend, checkout infrastructure, payments, inventory, etc. Shopify's current plans also support headless storefronts, with the number depending on the plan. www.shopify.com

Good fit if: your real goal is lowest total engineering/operations cost, rather than lowest software license cost.

3. Saleor — technically excellent, but I'd defer it

Saleor is arguably the most compelling pure headless/composable architecture here: open source, GraphQL-native, and explicitly designed to work with arbitrary frontend stacks. saleor.io

However, its current managed Cloud pricing starts at $1,599/month, which makes it hard to justify for a genuinely shoestring launch. saleor.io You can self-host the open-source version, but then you're taking on more infrastructure responsibility.

Good fit if: you anticipate complex catalogs, multi-market commerce, or sophisticated composable architecture and have the engineering team to support it.

My recommendation

If you're developer-heavy and cash-light: Medusa.

If you're developer-light and want the cheapest path to a reliable production store: Shopify + Hydrogen.

If you're building something complex from day one: evaluate Saleor, but don't choose its managed offering unless the budget supports it.

One important caveat: headless itself can be the expensive part. You're taking on storefront development, deployment, caching, search, CMS integration, analytics, and potentially checkout complexity. So I'd compare total cost to launch + maintain, not just monthly platform fees. A conventional Shopify storefront can easily beat a "free" headless platform on total cost. apps.teamzlab.com

If you tell me roughly monthly orders, number of SKUs, B2C vs B2B, and your preferred frontend (Next.js/Astro/etc.), I can narrow this to a specific stack and estimated monthly cost.

If low cost + headless + minimal platform lock-in are the priorities, I’d put Medusa at the top of the evaluation list.

My shortlist

Why Medusa: its core is open source with no license or GMV fee, and its managed cloud currently starts at $29/month. That gives you a relatively inexpensive way to start without committing to a proprietary commerce backend. medusajs.com

If you're comfortable operating infrastructure, self-hosting Medusa can reduce the software bill further—you'd mainly be paying for hosting, database, search, email, payments, etc. The tradeoff is that you become responsible for operations and maintenance.

Shopify is the alternative I'd evaluate alongside it. Its current plans start at $29/month when billed annually, and Shopify supports headless storefronts. It's likely to win on operational simplicity and ecosystem, but you give up more backend control and incur Shopify's commerce/payment economics. www.shopify.com

I would not start with Saleor for a tight-budget project: its current managed production offering starts at $1,599/month, despite the open-source core. saleor.io Vendure is worth testing if your team is strongly TypeScript/NestJS-oriented, but its commercial platform starts at €40,000/year, so its inexpensive path is primarily the self-hosted open-source route. vendure.io

What I'd evaluate first

1. Medusa → best budget/headless candidate
2. Shopify → benchmark for how much you're paying to avoid infrastructure work
3. Vendure → if TypeScript + GraphQL + extensibility are especially important

If you tell me your expected monthly orders/GMV, tech stack, and whether you want to self-host, I can narrow this to the cheapest realistic architecture and estimate the monthly TCO.

For a tight-budget, headless storefront, I’d put Medusa at the top of the evaluation list.

Shortlist

Why Medusa: its current Cloud Develop plan starts at $29/month, has no GMV fee, and includes backend/storefront hosting, GitHub deployment, previews, and its commerce features without additional platform fees. docs.medusajs.com

It also makes sense architecturally if you're building a custom frontend: Medusa is modular, supports multiple regions/currencies, and lets you own the storefront rather than forcing a particular frontend. docs.medusajs.com

Vendure is my second evaluation. Its Core is open source and can be self-hosted, making it potentially extremely cheap if you already have DevOps capability. Its managed Cloud Core currently starts at €199/month, though, so I'd favor self-hosting if cost is the overriding constraint. vendure.io

Saleor is technically compelling—GraphQL-native and genuinely headless—but its managed Cloud starts at $1,599/month, which makes it difficult to justify for a budget-conscious launch. saleor.io

My recommendation

Evaluate Medusa first → Vendure second → Saleor only if you have specific GraphQL/catalog requirements.

If you're comfortable managing a small Node/Postgres deployment yourself, I'd also compare self-hosted Medusa vs self-hosted Vendure. That could get your commerce backend down to infrastructure costs rather than a platform subscription.

If you tell me your expected monthly orders/GMV, frontend stack (Next.js, Nuxt, etc.), and whether you can self-host, I can narrow this to the cheapest realistic architecture.

If headless is a hard requirement and you're trying to keep the platform bill low, I’d put Medusa at the top of the evaluation list.

My shortlist

Why Medusa first: its current managed Cloud entry point is $29/month, with 0% GMV platform fee, unlimited products/orders/sales channels, and the ability to self-host the open-source platform if you want to squeeze infrastructure costs further. docs.medusajs.com

It is also TypeScript/Node-oriented and deliberately modular, which makes it particularly attractive if your storefront/BFF is already in the modern JS ecosystem. docs.medusajs.com

The important caveat

Don't assume "open source = cheapest." Headless shifts cost from licensing to engineering. You'll be responsible for the storefront, deployment, integrations, checkout UX, monitoring, and potentially more custom maintenance.

That's why I'd evaluate Medusa vs. Shopify Hydrogen rather than simply choosing the cheapest software. Shopify is more expensive as a backend, but its mature commerce infrastructure and app ecosystem can make the total cost lower if you don't have much engineering capacity. vendure.io

I would not put Saleor first for a tight-budget project if you're planning to use their managed cloud: its current published plans start at $1,599/month, despite the core being open source and self-hostable. saleor.io

My evaluation order:
1. Medusa → 2. Vendure → 3. Shopify Hydrogen → 4. Saleor

If you tell me your expected monthly orders/GMV, tech stack (Next.js/etc.), and whether you have 1–2 developers or a larger team, I can narrow that to the cheapest stack that is realistically production-worthy.

If low cost + headless + minimal vendor lock-in are the priorities, I’d put Medusa at the top of the evaluation list.

Why Medusa: its core is open-source/MIT, it is built specifically as a headless commerce backend, and its modular TypeScript architecture is a good fit if your storefront is Next.js/React. Current comparisons also put its managed-cloud option at the low end compared with enterprise headless platforms. contracollective.com

Vendure is the other one I'd actually prototype. It's TypeScript/GraphQL and has a particularly strong plugin architecture, but its commercial licensing becomes more relevant if you need the advanced platform features. Its free Core can still be self-hosted. vendure.io

I would not start with Saleor for a tight-budget project unless you specifically need its GraphQL-first architecture, complex catalog, or multi-region capabilities. Its managed cloud currently starts at $1,599/month, which makes it a very different proposition from a shoestring self-hosted build. saleor.io

My shortlist

1. Medusa → prototype this first
2. Vendure → evaluate alongside it if B2B/custom workflows are important
3. Saleor → only if its particular architecture solves a requirement you have
4. Shopify/Hydrogen → consider if minimizing engineering/DevOps is more important than minimizing platform cost

One important caveat: “open source = cheap” isn't automatically true. With Medusa/Vendure/Saleor, you trade license fees for hosting, maintenance, upgrades, monitoring, backups, and engineering time. For a genuinely small team, a managed service can actually be cheaper overall. contracollective.com

If you tell me roughly monthly orders, number of SKUs, B2C vs B2B, and your frontend stack, I can narrow this to the cheapest sensible architecture.

For a tight budget + headless requirement, I’d evaluate these first:

hunchbite.comwww.wpbundle.com

My shortlist for a budget-conscious headless build:

1. Medusa — probably the first platform I’d prototype

Good choice if you have developers comfortable with TypeScript/Node.js and want:

• Next.js/React storefront
• Full API ownership
• Custom checkout, pricing, bundles, workflows
• Low infrastructure cost

It is designed as an API-first commerce engine rather than a traditional storefront platform. powergatesoftware.com

A typical lean stack:

• Frontend: Next.js
• Commerce: Medusa
• CMS: Sanity or Contentful
• Payments: Stripe
• Hosting: inexpensive managed hosting or cloud VM

2. Shopify headless — if your team is small and wants fewer headaches

Use Shopify as the commerce backend and build your own frontend.

Best when:

• Merchants need a polished admin
• You need inventory/orders/payments to "just work"
• The storefront is where you want differentiation

You pay more for the platform, but you save engineering time. Shopify’s headless approach is usually the fastest path to production. www.askantech.com

3. Saleor — if you have Python expertise

Worth evaluating for:

• B2B
• Multi-channel catalogs
• Complex permissions
• GraphQL-first architecture

It is powerful, but I would hesitate for a very small team because the operational burden is higher. hunchbite.comwww.wpbundle.com

My recommendation by scenario

• <$5k/month budget, strong dev team: Medusa
• Small team, need launch quickly: Shopify headless
• B2B or complex catalog: Saleor or Medusa
• Existing WordPress business: WooCommerce headless (carefully)

Before choosing, I’d answer three questions:

1. Do you need a custom checkout/business logic, or just a custom frontend?
2. Who will operate the store day-to-day (developers or non-technical merchandisers)?
3. What is your expected scale: hundreds, thousands, or millions of SKUs/orders?

Those answers usually narrow the choice to one or two platforms quickly.

If headless is a requirement and budget is tight, I’d evaluate Medusa first.

My shortlist

1. Medusa — my first evaluation

Medusa

Medusa is open-source and built around a modular commerce backend, so you can run the commerce API yourself and pair it with whatever frontend you want. Its current stack is particularly attractive if your team is comfortable with TypeScript/Node. docs.medusajs.com

For a tight-budget build, the important advantage is that you're not immediately committing to a pricey managed-commerce tier. You can keep hosting relatively lean and spend money on the parts that actually differentiate the shop.

2. Saleor — strong alternative

Saleor

Saleor is arguably the more polished choice if you need sophisticated catalog, internationalization, multi-channel, or composable-commerce capabilities. It's genuinely headless and open source. saleor.io

However, I wouldn't choose Saleor Cloud for a tight-budget project: its current paid Cloud plans start at $1,599/month, although self-hosting the open-source platform is possible. saleor.io

3. Shopify + Hydrogen — evaluate if engineering time is more expensive than hosting

Shopify

Shopify's Basic plan is currently $29/month when billed annually, and Shopify supports headless storefronts. www.shopify.com The tradeoff is that you're buying into Shopify's ecosystem rather than minimizing platform dependence.

What I'd do

For a low-cost MVP, I'd test:

Next.js storefront → Medusa → PostgreSQL → Stripe → inexpensive managed hosting

Then compare that against:

Next.js/Hydrogen → Shopify

I'd only put Saleor ahead of Medusa if you already know you'll need its more sophisticated commerce model or your team strongly prefers Python/GraphQL.

One important caveat: headless isn't automatically cheaper. You're taking on frontend, deployment, caching, checkout integration, and ongoing maintenance that a conventional Shopify storefront handles for you. If the shop is relatively simple, Shopify's $29/month entry point can actually produce a lower total cost once developer time is included. www.shopify.com

Bottom line: start the evaluation with Medusa, keep Shopify as the "less engineering" benchmark, and consider Saleor if your requirements are more complex.