Open-source ecommerce platforms: which one fits your store
If you are running a self-hosted store and must pick a platform, the right choice depends on three things: how large your catalog and order volume will get, whether your team is strongest in PHP or JavaScript, and whether you want a coupled storefront that the platform renders or a headless setup where a separate frontend calls the commerce API. A small catalog with a marketing site already on WordPress points toward WooCommerce. A large catalog with a dedicated operations team points toward Magento Open Source, which is licensed under the OSL-3.0 and sits beneath Adobe Commerce, Adobe's commercial edition. PrestaShop, also OSL-3.0 for core, suits merchants who want a self-hosted PHP shop with a maintained 8.x and 9.x line. Shopware is MIT-licensed with a free Community Edition, but its paid Rise, Evolve and Beyond plans are priced on gross merchandise value, and its pricing page says shops above 1 million euros in annual GMV must move to a paid plan under its Fair Usage Policy. Medusa and Saleor are the options covered here for API-first stores that plan a custom storefront on a commerce API.
Open source status and a free product are not the same thing. The license governs what you may run, modify and redistribute, but your costs come from hosting, the search, cache and queue services each platform needs, extensions and themes, agency time for implementation, and the upgrades that follow. Ask each vendor or partner for a written list of the paid editions, plugins and services your projected volume would require, and confirm the current license terms in the repository before you commit. For the platform you shortlist, check the PHP, database and search versions it supports against your hosting plan, because those requirements drive your upgrade calendar. Build a short proof of concept with your real catalog import, one payment provider and one shipping rule, and time the first upgrade in a staging copy before you sign.
Which ecommerce platform fits which scenario
This table matches common store scenarios to the platform that fits best and names the condition under which each fit holds. Read each row against your catalog size, team skills and whether you need a storefront you build yourself. The main takeaway is that no single platform wins across all scenarios, so the deciding factors are usually existing skills and how much custom code you expect to own.
| Scenario | Best fit | Why, and the condition it depends on |
|---|---|---|
| WordPress-based content store | WooCommerce | Fits when content and marketing already run on WordPress and the catalog stays moderate. It holds only if your team can maintain the WordPress and WooCommerce update cadence and the extensions you rely on. |
| Large catalog needing a heavy stack | Magento Open Source | Fits when you have a team that can run PHP, MariaDB or MySQL 8.4, OpenSearch 3, RabbitMQ and Valkey and plan upgrades against the 2.4.x support windows. Confirm your hosting capacity and cache design before committing, since self-managed installs use Varnish. |
| PrestaShop-experienced team | PrestaShop | Fits when your staff already knows PrestaShop modules and themes and you want the Admin API or webservice for integrations. It holds while you stay on a maintained line (8.x or 9.x) and test modules before each upgrade. |
| Shopware-style API store | Shopware | Fits when you want an API-first store with a free Community Edition for small volumes. Shops above 1 million euros in annual GMV must move to a paid plan under the Fair Usage Policy, so model that cost against your GMV before you start. |
| Custom headless storefront on Medusa or Saleor | Medusa or Saleor | Fits when your team will build and own a separate storefront in a JavaScript or similar framework and wants a commerce API as the back end. It depends on engineering capacity to run the storefront, the commerce service and their supporting services, and on confirming which features sit behind paid tiers. |
| Team that wants to avoid paid tiers | Magento Open Source or PrestaShop | Both carry OSL-3.0 licenses with no GMV-based tier in the supplied facts. Fit holds only if you budget for hosting, implementation and paid or custom extensions, which are the real recurring costs. Confirm each extension's license before purchase. |
What running Magento Open Source 2.4.9 requires
Running Magento Open Source 2.4.9 means operating a full supporting stack, and the version pins matter as much as the application code. Adobe's 2.4.9 system requirements list PHP 8.5, MySQL 8.4 or MariaDB 12.3 (recommended) or 11.8, OpenSearch 3, Redis or Valkey 9, RabbitMQ 4.3, Composer 2.10 and nginx 1.30 on Linux x86-64. Adobe supports only the listed combinations, so a platform team should build the environment from that list, not from whatever versions its base image ships. Confirm each component against the Adobe Commerce system requirements before provisioning, and treat any upgrade of PHP, the database or the search service as a planned change with its own test cycle.
Caching and search are where self-hosted stores most often run into trouble. For self-managed installations, Adobe documents Varnish as the full-page cache accelerator, so your team owns its configuration, cache purging and the interaction with Content Security Policy headers, which run in restrict mode by default on storefront and admin payment pages from 2.4.7. Cloud-hosted Adobe Commerce uses Fastly instead, so a store that moves between self-hosting and Adobe's cloud will need its cache layer rebuilt, not simply redeployed. Budget for recurring staff time on OpenSearch, RabbitMQ and Valkey as well as hosting, because these services are part of every release test. Ask any implementation partner to price cache tuning, search reindexing and queue monitoring as separate line items in the written scope.
Each platform in brief, with its license
Six platforms sit on the shortlist for self-hosted ecommerce, and each one carries a different license model. Read the license first, then the fit, because an OSI-approved license does not tell you what a paid tier, a plugin dependency or a usage threshold will cost you later.
Magento Open Source
Magento Open Source is licensed under OSL-3.0 and is the open edition of the platform that Adobe sells as Adobe Commerce. It suits teams that want a full self-hosted store with their own hosting stack and developers who will maintain custom modules. The caveat is the stack: PHP 8.5, MariaDB or MySQL 8.4, OpenSearch 3, RabbitMQ 4.3 and Valkey 9 must all be run and patched. Ask which features your store needs that only Adobe Commerce provides.
WooCommerce
WooCommerce is a GPL-2.0-or-later ecommerce plugin for WordPress, so it inherits the WordPress hosting model and its plugin ecosystem. It fits small and mid-sized catalogs where the content team already runs WordPress. The caveat is that store scale depends on the plugins, themes and hosting you add, and the license of a paid extension is separate from WooCommerce itself. Confirm which extensions you need and their support terms before you commit.
PrestaShop
PrestaShop Core is licensed under OSL-3.0 and its modules under AFL-3.0, so you need to track two license terms in one shop. It runs on PHP with two maintained lines, 9.x and 8.x, which lets you choose between a newer major version and a more conservative upgrade path. It suits merchants who want a self-hosted shop with a built-in admin API. The caveat is that PrestaShop does not support Windows in production, so confirm the Unix-based hosting model before you plan.
Shopware
Shopware is MIT-licensed and offers a free Community Edition alongside paid Rise, Evolve and Beyond plans priced on gross merchandise value. The free code does not remove the commercial decision: once annual GMV passes the threshold in the Fair Usage Policy, you must move to a paid plan. It suits merchants who expect growth and want to model that cost early. Ask for the GMV band and the features each plan includes before you design the store.
Medusa
Medusa is an open-core commerce framework built on Node.js and TypeScript, aimed at teams that want to own the commerce logic and build a custom storefront. It fits developer-led teams with a headless architecture and JavaScript skills. The caveat is that open core means some capabilities sit in commercial or managed offerings, so check the license of each module you plan to use. Confirm the exact license terms of each component with the vendor before you rely on them.
Saleor
Saleor is a BSD-3-Clause commerce platform with a GraphQL API, built on Python and Django. It suits teams building a headless storefront who want a permissive license and a backend in Python. The caveat is that the API-first model shifts storefront work, integrations and hosting onto your team, and Saleor is also offered as a managed cloud product. Confirm the GraphQL schema coverage and the supported deployment model for your use case before you commit.
How hosting stack and extension costs differ by platform
WooCommerce runs as a plugin inside a WordPress site, so the hosting stack is the WordPress stack plus the store data. The published baseline is PHP 8.3 or greater, MySQL 8.0 or MariaDB 10.6 or greater, WordPress 6.9 or greater, a WordPress memory limit of 256 MB or more, and HTTPS support. Legacy PHP 7.4 and MySQL 5.6 may still run, but the WooCommerce documentation notes that those versions are end of life and pose security risks.
- Runtime baseline: Confirm that your hosting provider supports PHP 8.3, the MySQL or MariaDB floor, and WordPress 6.9 together before you commit to a hosting plan.
- Memory and TLS: A 256 MB memory limit and HTTPS are minimums. Only load testing with your own catalog size and checkout traffic shows whether those settings are enough.
- Order storage: High-Performance Order Storage has been stable and enabled by default for new installations since WooCommerce 8.2. Older stores may need a migration project, which should appear as a separate line in the quote.
- Premium extensions and themes: Paid extensions and themes bought from WooCommerce.com receive support through the WooCommerce.com premium support portal. Each license has its own term and renewal, and each needs compatibility checks on every core upgrade.
- Coupled upgrades: WordPress, WooCommerce and each extension must move in a tested sequence. WooCommerce 11.2.1 was published on October 9, 2026, so plan for frequent release cycles.
Before signing, ask for a line item for each extension with its license term, renewal date and who handles compatibility testing after each core upgrade.
PrestaShop is a PHP application that runs on a conventional web server and database stack. The published minimums for PrestaShop 9 are MySQL 5.7 or MariaDB 10.2, Apache 2.4 or later (Nginx also works), and a memory_limit of at least 512M. The project says it works best on Unix-based systems and does not actively support Windows, so the operating system choice affects both hosting cost and the pool of administrators who can run it.
- Database floor: MySQL 5.7 and MariaDB 10.2 are the stated minimums. Both are older lines, so check each database vendor's support status before using them for a new build.
- Web server: Apache 2.4 is the reference server and Nginx also works. Your rewrite rules, TLS settings and security headers become part of your change management.
- Memory: A memory_limit of at least 512M is higher than the WordPress baseline. Size PHP-FPM or Apache PHP pools with that figure in mind, because undersized pools cause errors that look like application bugs.
- Modules and integrations: PrestaShop modules are licensed under AFL-3.0. The classic webservice gives third-party tools CRUD access to products, orders and customers using access keys, and PrestaShop 9 adds an Admin API with OAuth2 authentication. Each module adds its own purchase, support and upgrade cost.
- Version lines: The 8.x line still receives releases, including 8.2.8 on August 18, 2026, alongside 9.x. Decide which line you run and how long you will maintain it.
Before signing, ask for the full module list, the authentication method each integration uses, and whether the quote covers module compatibility testing for each PrestaShop release.
A headless store separates the storefront, the commerce backend, the content service, search, cache and database into parts you deploy and maintain separately. Hosting and extension cost follow the number of those parts, so the software license is only one line in the bill. Saleor, a BSD-3-Clause GraphQL commerce platform on Python and Django, is offered as Saleor Cloud or self-hosted. Shopware Community is MIT-licensed, but shops above 1 million euros in annual GMV must move to a paid plan under its Fair Usage Policy.
- More services to run: The storefront, commerce API, database, cache, search and any CMS are billed and patched separately. Count each one in the quote alongside the license.
- Managed or self-hosted backend: Choosing between a managed cloud offering and your own infrastructure changes who pays for upgrades, backups and incident response. Settle that decision before you size the team.
- Revenue-linked licensing: Shopware's pricing page ties paid plans to gross merchandise value. Model the license against projected sales as well as current sales, because the tier can change as the store grows.
- Custom code over plugins: Many headless features are built into your storefront code, so ongoing cost is mostly developer time. Ask which features are custom and who maintains them after launch.
- Runtime policies: Node.js-based components follow their own supported runtime lines. Strapi 5, for example, runs only on Active LTS or Maintenance LTS Node.js versions, so confirm the equivalent policy for any headless component you choose.
Before signing, ask for an architecture diagram with each service, its hosting owner, its upgrade cadence and the sales thresholds assumed in any paid tier.
How to choose and pilot an ecommerce platform
This sequence turns a shortlist of open-source ecommerce platforms into a decision you can defend, by testing licensing, integrations and upgrade effort against your own catalog, revenue and team before you commit.
Define catalog, integrations and team skills
Start by writing down your catalog size, the number of product variants and orders you expect, the payment providers, tax, shipping, ERP and CRM systems you must connect, and the languages and currencies you sell in. Ask your developers and operations staff which PHP, Node.js or Python skills and hosting experience they already have. This step is done when the team agrees on a one-page requirements document that a vendor or implementation partner could quote against. Illustrative example: a store with a few thousand SKUs, two payment providers and an existing ERP has a very different integration load from a store with a headless storefront feeding several channels.
Shortlist two or three platforms
Pick two or three candidates whose architecture matches the requirements from the first step. A monolithic platform with a built-in storefront, such as Magento Open Source, WooCommerce or PrestaShop, fits a different team than an API-first framework such as Medusa, Saleor or Shopware with a custom storefront. Document why each candidate made the list and why the others were dropped. This step is done when the shortlist includes a written reason for each platform and a named owner for the evaluation.
Check license and paid tiers against revenue
For each candidate, confirm the license of the code you will deploy, the license of extensions or modules you plan to use, and any paid edition or usage threshold that applies to your business. Shopware's Community Edition, for example, has a published Fair Usage Policy that moves shops above 1 million euros in annual GMV to a paid plan, so compare that threshold with your forecast. Ask each vendor in writing which features require a paid edition and which support windows apply. This step is done when you have a table of license, paid tiers, triggers and support dates, verified against the vendor's own pages rather than third-party summaries.
Build a pilot with your payment provider
Install the shortlisted platforms in a non-production environment and connect your real payment provider in sandbox mode. Add the three or four extensions you cannot operate without, such as tax, shipping, or ERP connectors, and confirm that each one works on the platform version you intend to run. Include checkout, refunds, order export and a stock update in the test script. This step is done when every item on the script passes for each candidate, and the failures are logged with the extension, version and fix required.
Test the upgrade path on staging
Copy the pilot database and files to a staging environment, then perform a minor and a major upgrade on each candidate using the vendor's documented steps. Record the time spent, the extensions that broke, the manual database or configuration changes needed, and whether your custom code required rewrites. Check the upgrade notes for the version you plan to run and confirm the required PHP, Node.js or database versions with the vendor. This step is done when you can state, for each platform, how many hours the upgrade took and which components needed rework.
Document support windows before committing
Before signing anything, record the end-of-support date for the version you will run, the date of the next major release, and the security patch policy for each candidate. Confirm which hosting services, managed options and third-party agencies the vendor recognizes, and what a partner or retainer would cover. Ask for these dates in writing and file them with the requirements document. This step is done when every candidate has a dated support statement that the decision owner has reviewed, and any gap between your planned upgrade cycle and the vendor's support window is either closed or accepted in writing.
How upgrade effort and support windows shape long-term cost
Upgrade effort depends less on how often a platform releases than on how far your installed version falls behind its support window. For Magento Open Source, Adobe's lifecycle policy gives each 2.4.x release three years of standard support from general availability. Regular support for 2.4.6 ended on August 11, 2026, 2.4.7 runs until May 31, 2027, 2.4.8 until May 31, 2028, and 2.4.9 until May 31, 2029. A store still on 2.4.6 is already outside regular support, so every later security fix becomes a catch-up project. Ask Adobe to confirm the dates for your exact version, and plan the next move from the Adobe Commerce lifecycle policy before the window closes, not after.
PrestaShop and WooCommerce ask for different habits. PrestaShop 9.0 reached its first stable release on June 10, 2025, and 9.2.0 followed on September 30, 2026. The 8.x line still receives releases, with 8.2.8 and 9.1.5 both published on August 18, 2026, so a shop on 8.x must decide between staying on a maintained branch and migrating to 9.x. WooCommerce ships more often: 11.2.0 and 11.2.1 were published two days apart in October 2026. Frequent point releases are cheap to apply only if the store is not carrying custom database changes. WooCommerce stores orders in dedicated tables (wc_orders and wc_orders_meta) under High-Performance Order Storage, which has been stable and enabled by default for new installations since WooCommerce 8.2. Older stores that still keep orders in the WordPress posts tables need a confirmed migration path before any upgrade that touches order data.
Custom extensions raise the cost of every upgrade, because each one must be retested against the new core, the new PHP version and the new database or cache stack, and someone must own the fix when it breaks. Extensions from the marketplace carry the vendor's upgrade schedule, while in-house modules carry yours. Before you buy or build, ask for the upgrade path of each extension, the vendor's support window for each core line, and who pays for patches when a core release changes an API. In a quote, separate the one-time build cost from the recurring cost of keeping each extension current. Fixed-scope upgrade work and ongoing retainers can both be written in advance, but the scope must name the extensions and the target versions, or the effort will be estimated again at every release.
What buyers ask before they hire help with an ecommerce platform
What drives the cost of implementing an open-source ecommerce platform?
The biggest drivers are the number of integrations (payments, tax, shipping, ERP, customer data), how much the storefront or checkout must be customized, data migration from an existing store, and the hosting footprint you choose. Open-source licenses for the core platforms covered on this page carry no license fee in most cases, so you pay for the people and infrastructure around the platform. Ask any provider for a quote that separates one-time work (planning, build, migration, hardening, testing) from recurring work (hosting, upgrades, extension maintenance, support) and states which items are priced separately.
How long does an ecommerce implementation usually take?
Duration depends on scope rather than on the platform name alone. A store that uses a standard theme, a few well-documented payment and shipping extensions and a simple catalog takes far less time than one with a custom headless storefront, multiple storefronts, a legacy catalog to migrate and several back-office systems to connect. No duration figure applies across these platforms, so ask each shortlisted provider for a timeline tied to a written scope, and ask what is excluded from it, such as load testing, security testing or a second round of data migration.
What does the work require from our own team?
Expect to supply a product owner who can make catalog, pricing and checkout decisions, someone who owns payment and tax configuration, and an engineer or administrator who will receive the handover. Platforms such as Magento Open Source and PrestaShop also need someone who understands the server stack, including PHP, the database and the search or cache services each version requires. Agree in writing who approves design, who signs off the test results and who will operate the store after go-live.
Can our in-house team implement and run the store without outside help?
Yes, if you already run PHP or Node.js applications, have a tested deployment pipeline and can dedicate people to upgrades and security patching. The main risks for in-house teams are the first major migration, performance under peak traffic, payment compliance evidence and keeping extensions compatible across upgrades. A mixed model is common: your team runs day-to-day operations while an implementation partner handles the initial build, hardening and upgrade planning under a fixed scope.
How do extension and theme costs recur after launch?
Paid extensions and themes usually carry an annual license or renewal, and each major platform upgrade can require a new version of the extension or a code change. Budget for compatibility testing every time the core platform moves to a new release, since an extension that works today may not support the next PHP or framework version. Ask each vendor which extensions are maintained for the versions you plan to run and how long their support windows last.
How do Shopware's GMV thresholds change our planning?
Shopware offers a free Community Edition, but its pricing page says shops above 1 million euros in annual gross merchandise value must move from the Community Edition to a paid plan under its Fair Usage Policy. Paid plans are priced on GMV, so a store that grows through the threshold will see a new recurring cost that should be in your forecast from the start. Model your revenue growth against that line before you choose the edition, and confirm current plan terms directly with Shopware.
What does a fixed-scope implementation from DeployOpen include?
DeployOpen plans, implements, customizes, integrates, hardens, backs up, upgrades, documents and supports self-hosted open-source platforms inside your own cloud accounts, data centers or air-gapped networks. Most engagements are fixed-scope, with scope, deliverables and timelines agreed in writing before work starts. Ongoing support, upgrades and continuous testing are available as retainers. DeployOpen does not publish a price list, so a quote follows from the written scope, and DeployOpen prepares controls and evidence for compliance assessments without issuing audit reports or certifications.
When a scoped implementation partner helps your store choice
A scoped implementation partner is worth considering when your team has shortlisted two or three ecommerce platforms but lacks the capacity to build, harden and run each one for a fair test. DeployOpen plans, implements, customizes, integrates, hardens, backs up, upgrades, documents and supports self-hosted open-source platforms inside your own infrastructure, whether that is your cloud accounts, data centers or air-gapped networks. The work happens in your environment, under your control of keys, data and access, which matters for a store that handles payment flows and customer records.
Most DeployOpen engagements are fixed-scope. Before work starts, the scope, deliverables and timelines are agreed in writing, and a quote follows from that document. DeployOpen does not publish a price list, so ask for a quote that names the platform, the version, the number of stores or sites, the payment and shipping integrations, the ERP or CRM connections, and whether a staging environment and upgrade path are included. Ask which items recur, such as hosting, patch windows and retainer-based support, and which are one-time, such as the initial build and hardening. Ongoing support, upgrades and continuous testing are available as retainers, so the choice of a retainer versus a one-off build is a decision you can make separately.
DeployOpen is not a vendor partner, reseller or certified provider of Magento Open Source, WooCommerce, PrestaShop, Shopware, Medusa or Saleor, and it does not issue audit reports or certifications for any of them. Treat any partner, including DeployOpen, as an implementer rather than a source of platform licensing or compliance evidence. Confirm license terms directly with each project and, where a shop needs an assessment, with the independent auditor who will perform it.
How DeployOpen scopes and delivers ecommerce platform work
DeployOpen plans, implements, customizes, integrates, hardens, backs up, upgrades, documents and supports self-hosted open-source platforms inside your own cloud accounts, data centers or air-gapped networks. For ecommerce work, that means agreeing the platform, hosting stack, extensions and integrations in writing before build begins, so the store you run is the one you approved.
How we scope
- Engagements are fixed-scope, with scope, deliverables and timelines agreed in writing before work starts.
- The written scope names the platform version, database, search, cache, queue and payment integrations that the store depends on.
- Extension and customization needs are listed individually so each one can be reviewed for licensing and upgrade impact.
How we build
- Installation runs in your environment, on the server stack your team has chosen and the vendor supports for that release.
- Customizations, extensions and API integrations are built against the platform's documented interfaces and tested before go-live.
- Security testing of the storefront, APIs and admin access can be scoped alongside the build, as separate written deliverables.
How we hand over
- Handover documentation covers architecture, configuration, integration points, backup and restore steps and upgrade procedures.
- Your team receives the controls and evidence that support a compliance assessment, which independent auditors then evaluate.
- Ask in the quote which artifacts, training sessions and access transfers are included in the fixed scope.
How we support
- Ongoing support, upgrades and continuous testing are available as retainers after the initial engagement closes.
- Upgrade planning covers platform release lines, extension compatibility and the PHP, database and cache versions each line requires.
- Confirm response expectations, upgrade windows and included testing in the retainer terms before you sign.

