Ruby on Rails ecommerce: Spree vs Solidus compared

pexels-photo-230544

Two Ruby on Rails ecommerce frameworks are actively maintained today: Spree Commerce 6.0 and Solidus 4.8, both under the BSD 3-Clause license. They share a common ancestor and now diverge sharply.

Spree 6 is API-first with a React admin and a Next.js storefront, while Solidus remains a Rails-first engine that you extend inside your own app. The right pick depends on whether you want a headless platform, a customized Rails monolith, or a marketplace or B2B build.

Which Ruby on Rails ecommerce frameworks are still maintained?

Two open-source Ruby on Rails ecommerce frameworks are actively maintained: Spree Commerce 6.0 and Solidus 4.8, both under the BSD 3-Clause license. Spree 6.0, released 27 October 2026, is a headless platform with a React admin and a Next.js storefront.

Solidus 4.8, released 9 October 2026, stays Rails-first.

Release status was checked on 13 November 2026. The choice depends on whether you want a headless storefront or server-rendered Rails views, and on how much custom code you already carry.

Netguru builds and maintains Ruby on Rails applications, including Spree-based stores. We recommend Spree 6.0 for new headless builds and Solidus for stores that keep Rails views, because each matches how that codebase will be upgraded.

Is Ruby on Rails good for ecommerce, and when is it not?

Ruby on Rails is a good base for ecommerce when your pricing, catalog or order logic is custom, and a poor one when a standard catalog needs no custom code.

Rails fits stores where the business rules are the product. Customer-specific price lists, bundles with conditional stock, split shipments, quote-to-order approvals and marketplace commissions all end up as domain code. Rails lets you keep that code in your own repository, organized in engines and decorators, under your own test suite. Hosted platforms often push the same rules into app extensions or workarounds.

Ruby on Rails 8 also trims the operations side. Solid Queue is the default Active Job backend for new apps, so order emails, inventory syncs and webhook deliveries can run from the database without a separate Redis service.

A store with very high job volume may still prefer a dedicated queue, and that is a judgment call to make from your own load profile.

Rails also suits headless commerce, where the back end owns catalog, cart and checkout logic behind an API and a separate JavaScript storefront renders it. We suggest this split only when the storefront needs its own team or release cycle, because it adds a second codebase to deploy and monitor.

Rails loses when the catalog is standard, checkout works as shipped, and nobody on the team writes Rails. The Ruby and Rails upgrades, gem updates and security patches then fall on people who cannot do them. In that case a hosted or composable commerce platform is the cheaper path.

Is Shopify built on Rails?

Yes, Shopify's core is a Ruby on Rails application, which Shopify's engineering team has described as a large modular monolith. That does not make Shopify a Rails framework you can build a store on.

Shopify is a hosted product. You cannot add engines, write decorators or change order models in its codebase. You extend it through apps, themes and its APIs. A headless commerce setup is possible through its Storefront API, but the backend stays Shopify's.

Spree and Solidus are code you own. They run inside your own Rails app, on your infrastructure, and you decide when that app moves to Ruby on Rails 8. With Shopify, Shopify decides.

Choose Shopify when you want hosted commerce and can work within its extension points. Choose Spree or Solidus when pricing, checkout or order logic has to live in your own domain code. For other well-known Rails applications, see our list of companies using Ruby on Rails.

Spree vs Solidus comparison table

Spree Commerce 6.0 is an API-first headless platform, and Solidus 4.8 is a Rails-first framework. The table compares them on nine criteria as of 13 November 2026.

Every cell comes from the Spree releases and Solidus releases pages on GitHub, plus each project's docs.

Criterion Spree Commerce 6.0 Solidus 4.8
Architecture Headless platform. The core is still Ruby gems (api, core, dashboard, emails, mcp). Rails-first framework, a 2015 fork of Spree, extended inside your own Rails app.
Storefront and admin Spree Next.js storefront (Next.js 16) and Spree React admin dashboard, built on Vite and TanStack. Storefront is code in your Rails app. solidus_admin v0.5.0 (released 2026-10-14) provides the admin.
API REST API with typed TypeScript clients, webhooks and events, OpenTelemetry. Customization is mainly Ruby-level in your app; no typed client confirmed in the release notes.
Marketplace and B2B features Built in: sellers, commissions, Stripe Connect payouts; companies, catalogs, price lists. Not confirmed as built in for 4.8. Plan custom work or extensions.
Extensions Gem-based. Gateways come through extensions, and Stripe is built in. Gem-based. Gateways come through extensions.
Upgrade path Official 5.6 to 6.0 upgrade guide, which requires a store to be on Spree 5.6 first. Per-release changelog on GitHub. We recommend upgrading one minor version at a time.
Release cadence v6.0.0 on 2026-10-27, v6.0.4 on 2026-11-12. v4.8.0 on 2026-10-09, on a more conservative schedule.
License BSD 3-Clause. Copyright Vendo since 2024, Spark Solutions 2015-2024. BSD 3-Clause.
Community and maintainers Company-led project with docs on spreecommerce.org. Independent open-source project with docs on solidus.io.

The two differ most in who owns the frontend and the admin. Spree 6 ships both as separate JavaScript applications talking to the REST API, which suits a team with React and Next.js skills. Solidus keeps everything in one Rails codebase, which suits a team that wants a single deploy and Ruby-only customization.

The marketplace and B2B row is the one most likely to decide a build. A marketplace on Solidus means writing seller, commission and payout logic yourself.

Which framework wins on each criterion?

Spree 6.0 wins on headless architecture, marketplace and B2B features. Solidus 4.8 wins on in-app Rails control and release predictability, and the license is a tie.

The verdicts below are our judgment, based on each project's release notes and docs as of 2026-11-13. The table gives the winner and the condition that decides it:

Criterion Winner Deciding condition
Headless architecture Spree A separate Next.js or mobile client consumes the REST API and typed TypeScript clients.
In-app Rails control Solidus Storefront and business logic should stay inside the Rails app your team already owns.
Admin Spree You want the React admin dashboard. Solidus wins if you want to stay in Rails with solidus_admin, which is still on a 0.x version.
Marketplace and B2B Spree You need sellers, commissions, Stripe Connect payouts, companies, catalogs or price lists without writing them.
Payments Spree Stripe is your gateway, since Spree 6 includes it. Other gateways come through extensions on both.
Extensions No winner Check the specific extension against each project's supported version before committing.
Upgrade path Solidus The store carries heavy Rails customizations. Spree wins if you are close to the version the Spree 5.6 to 6.0 upgrade guide starts from.
Release cadence Solidus You want predictable, smaller upgrade steps.
License Tie Both use the BSD 3-Clause license.

The upgrade verdict is the least obvious. Spree 6 changes the architecture, so a store with custom admin screens or storefront views has to port them to the new dashboard and storefront model. Read the upgrade guide before you estimate. A store without those customizations has far less to port.

Solidus has shipped 4.4 in November 2024, 4.5 in February 2025, 4.6 in September 2025, 4.7 in April 2026 and 4.8 in October 2026. That is a release every 3-7 months, with the last three gaps at 7, 7 and 6 months. A team that budgets one upgrade slot per release can plan around that rhythm.

The verdicts can point in different directions for the same team. A B2B store with a heavily customized Rails codebase splits between Spree on features and Solidus on upgrade risk. In that case, we recommend deciding on architecture first, because headless or in-app Rails determines most of the other rows.

How to upgrade Spree 5 to Spree 6 with heavy customizations

Upgrading a heavily customized Spree 5 store to Spree 6 means rebuilding every customization that renders HTML in the Next.js storefront or the React admin. Ruby model and job code carries over more easily.

The Spree 5.6 to 6.0 upgrade guide is the authority on which APIs changed. It requires the store to be on Spree 5.6 first, and it states that the Rails admin and Rails storefront gems are not part of Spree 6.0. Read it before estimating, because customization debt, not the version bump itself, sets the cost of the move. The table sorts the usual customization types by what happens to them:

Customization What happens in Spree 6 Action
ERB storefront views and helpers The Next.js storefront renders the frontend, so Rails views have nothing to render into Rebuild as React components against the REST API and TypeScript SDK
Admin overrides The React admin is a separate Vite and TanStack app Rebuild the screens in React; the ERB partials are dropped
Model and controller decorators Core is still Ruby gems, but decorators patch internals that can change between major versions Check each one against the upgrade guide
Third-party extension gems Compatibility depends on each maintainer Audit, then pin, fork, or replace

Decorator debt is the failure mode that hides longest. A decorator that adds a callback to order completion for an ERP sync often still loads, then misbehaves when the method it wraps changes. The early sign is a decorator with no test covering the behavior it adds.

We recommend moving those side effects to webhooks and events first. A subscriber that reacts to an order event or a webhook survives internal refactors, and it works the same before and after the upgrade.

Extension gems lag a major release more often than core does. Count the extensions that touch checkout, promotions or admin screens, and check each one's release history before committing to a date.

This cost estimate changes if most of your customization is in the admin. A store with dozens of custom admin screens may find Solidus 4.8 and solidus_admin cheaper than a rebuild.

How do payments and integrations work in Rails stores?

Rails stores take payments through gateway extensions in Solidus 4.8 and through built-in Stripe in Spree 6. Webhooks and events carry data to everything else, and the choice decides who maintains your checkout code after each upgrade.

Hub and spoke diagram with the store's checkout and framework at the center. Spokes connect to gateway extensions, built-in Stripe, outbound event webhooks, inbound webhooks and REST API pulls.

Payment gateway extensions wrap a provider's API behind the framework's payment method abstraction. You can connect any provider, including regional ones, but each extension has its own maintainer and compatibility range. We recommend checking an extension's supported framework and Rails versions before scheduling a major upgrade. An extension that lags a major release blocks checkout until someone patches it.

Spree 6 builds Stripe into the platform, including Stripe Connect payouts for marketplace sellers and commissions. For a marketplace, that removes the split-payment and payout code you would otherwise write and keep compliant. A single-brand store on another provider gains nothing and still needs an extension or a custom integration.

For everything beyond payments, three patterns cover most needs:

  • Outbound events: subscribe to webhooks and events for orders, shipments and inventory, and push them to an ERP or warehouse system.
  • Inbound webhooks: verify the signature, store the payload, and hand processing to Active Job so retries are safe and idempotent.
  • API pulls: read orders and products through the REST API for analytics or a search index.

For a provider-by-provider comparison, see our guide to online payment service providers.

What makes a Rails store slow, and how do you fix it?

A Rails store usually slows down in three places: cache keys that ignore pricing context, job queues that back up, and N+1 queries on catalog pages. Ruby on Rails 8 makes Solid Cache and Solid Queue the defaults for new apps, which lowers the cost of running the first two but does not fix how you design them (Rails 8 release notes).

Grid diagram with three rows, each pairing a Rails store slowdown with its fix: cache keys that ignore pricing context, job queues that back up, and N+1 queries on catalog pages.

Cache keys that ignore pricing context. Fragment caching keyed on a product's updated_at serves stale prices when a price list changes without touching the product. Keying every fragment per customer is correct but collapses the hit rate.

Cache the shared parts (description, images, options) and key the price fragment on the price list or catalog version. Solid Cache stores entries on disk, so a larger cache is affordable, but key design still sets the hit rate.

Job backlogs on orders and webhooks: order finalization, confirmation emails, and webhook handling often share one Active Job queue. A retry storm from a payment provider then delays customer emails.

Give each type its own queue and a dedicated Solid Queue worker pool, make handlers idempotent, and cap retries. New Rails 8 apps give Solid Queue its own database, and we recommend keeping that separation so job polling never competes with checkout writes.

N+1 queries on catalog pages. A listing page that loops over products and touches variants, prices, images, and taxons issues a query per product for each association. Preload those associations in the listing query and turn on strict_loading in tests so regressions fail the build. Serve images and static assets from a CDN so the app only renders HTML.

OpenTelemetry shows which of the three you have. Trace query counts per request, and alert on queue latency (time from enqueue to start) because queue depth alone hides slow jobs. Spree 6 ships with OpenTelemetry support; on Solidus you add the Ruby instrumentation yourself.

A low-traffic store can run on the Rails 8 defaults for a long time. These fixes pay off when you have price lists, per-company catalogs, or peak-day order volume, and the scaling guide covers the next steps.

What are the security risks of a Rails store?

The main security risks of a Rails store are unpatched dependencies, an unsupported Ruby or Rails version, and payment gateway extensions that lag behind the framework.

Pipeline diagram of the recommended security cadence for a Rails store in three ordered stages: automated audits in CI on every merge and weekly, an extension and payment gateway review before each major upgrade, then verifying patch levels on running servers.

Tie the audit cadence to the support windows. Ruby on Rails 8 gets fixes under the Rails maintenance policy, and Ruby branches leave security maintenance on the ruby-lang.org schedule.

Put both end dates in the team calendar and start the upgrade a full release cycle before the earlier one. Spree Commerce 6.0 and Solidus declare the Rails versions they support, so the framework upgrade often gates the Rails upgrade.

A cadence we recommend:

  1. Run bundler-audit and Brakeman in CI on every merge, plus a weekly scheduled run so idle branches still surface new advisories.
  2. Before each major upgrade, list every extension and payment gateway gem. Check its last release, its supported Rails range and its open security issues, then replace or fork any that lag. Gateway extensions sit closest to payment data, so they get the strictest review.
  3. Verify the patch level on running servers as well as in the lockfile. Enforce TLS, keep secrets in encrypted credentials, and use Active Record encryption for sensitive attributes.

Our Rails security guide covers the application-level checks, and the OWASP Top Ten is a sound baseline for the review.

Which should you pick: Spree, Solidus or something else?

Pick Spree Commerce 6.0 for a new DTC store, a marketplace or a B2B build. Pick Solidus 4.8 for a heavily customized Rails store that needs a conservative release cadence.

The table below maps each situation to a default. The reasoning follows.

Situation Default pick Why
New DTC store, headless commerce Spree 6.0 REST API, TypeScript SDKs, Next.js storefront and React admin dashboard ship together
Marketplace Spree 6.0 Sellers, commissions and Stripe Connect payouts are built in
B2B Spree 6.0 Companies, catalogs and price lists are built in
Existing store on Solidus Stay on Solidus Rails-first upgrades, with solidus_admin as the admin path
Simple catalog, no Rails team Hosted platform No Ruby or Rails upgrades to own

The marketplace and B2B rows carry the most weight. On Spree 6.0, the marketplace and B2B features are part of the platform, so seller onboarding, commission rules and Stripe Connect payouts need configuration, not a custom build. We found no equivalent seller or commission models in the Solidus docs. Expect to build or source them as extensions.

An existing customized store rarely justifies a replatform. Keep a Solidus store on Solidus, because upgrades stay inside the Rails conventions your team already follows. A Spree store should read the 5.6 to 6.0 upgrade guide before committing, then move in stages.

Admin and storefront become separate React and Next.js applications, so customizations written against Rails-rendered views will probably need reimplementing. Extending the current platform beats replacing it when the store already encodes years of business rules.

Netguru took that route for Ledbury, a US luxury shirt retailer running an in-house Spree store. A new Spree module tied online sales to its physical shops, so a customer measured in store could order online, and it extended the existing platform instead of replacing it.

Headless commerce has a cost. It adds a TypeScript application beside the Rails core, so a team with only Rails developers should budget for that skill or stay with server-rendered Solidus.

A hosted platform wins when the catalog is simple, checkout needs no custom pricing or fulfillment logic, and nobody wants to own Ruby and Rails upgrades. Teams that prefer assembling separate services around a thin core should look at composable commerce.

If you are unsure which side of that line your store sits on, ecommerce development scoping will settle it faster than a framework comparison.

Ruby on Rails ecommerce FAQ

Does anybody still use Ruby on Rails?

Yes. Ruby on Rails is still in active use for ecommerce, and both maintained frameworks shipped releases in October 2026: Solidus 4.8 on 9 October and Spree Commerce 6.0 on 27 October. The Ruby on Rails 8 series remains maintained. Check each project's supported Rails version before planning an upgrade.

What are the pros and cons of Solidus?

Solidus is a Rails-first framework under the BSD 3-Clause license, with minor releases from 4.4 in November 2024 to 4.8 in October 2026. The cost is that your team owns most storefront code, and solidus_admin was still at version 0.5.0 in October 2026. It suits customized stores better than fast greenfield launches.

Best practices for customizing Solidus

Customize Solidus through its configuration, event subscribers and extensions, and keep changes outside the gem code. Cover every override with tests, because each one adds upgrade cost, which matters most at minor-version upgrades where the changelog shows which methods you overrode have changed.

Choosing your Rails ecommerce framework

Choose Spree Commerce 6.0 for new headless, B2B or marketplace builds. Choose Solidus 4.8 when you want a Rails-first framework with a conservative release cadence.

Both are BSD 3-Clause, so licensing will not settle the choice. Prototype your most customized checkout or pricing flow on each, and confirm the supported Ruby and Rails versions on the project release pages before committing.

If you want a second opinion on that shortlist or an upgrade plan for an existing store, tell us about your store and Netguru's Rails and commerce engineers will review it with you.

We're Netguru

At Netguru we specialize in designing, building, shipping and scaling beautiful, usable products with blazing-fast efficiency.

Let's talk business