Shopware 6 vs Endora Commerce — which one to pick for B2B

Shopware 6 is the platform that makes a conversation about B2B genuinely interesting. A German vendor, PHP and Symfony underneath, an API-first architecture with a separate Store API for the storefront and an Admin API for integrations. On top of that, two tools that make a real difference to a commercial team’s day-to-day work: the Rule Builder and the Flow Builder. Commercial rules and automations are assembled in the admin panel with no developer involved, and that is a capability most of the competition does not have at this level. On the Evolve and Beyond plans you also get B2B Components: six modules you can switch on independently (employee management, organization units, quote management, order approval, quick order and shopping lists). Add a free self-hosted Community Edition and a very deep partner ecosystem across the German-speaking countries.

That matters, because it changes the character of this comparison. With Magento and Shopify the axis ran through what B2B is simply missing: price negotiation, order approval on the buyer’s side, a hierarchy of companies. Shopware has all of that. Quote Management is a fully fledged module, and so is Order Approval. There is no point pretending Endora wins here on the length of the feature list, because it does not.

The point of origin, though, is different, and it comes back in every criterion below. Shopware 6 was built as a shop engine, with B2B bolted on as a set of modules unlocked by a pricing tier: in the Community Edition and on the Rise plan, B2B Components are simply absent. Endora went the other way: the buyer is an organization, the price comes from a contract, an order can be the outcome of a negotiation, and retail selling runs as an extra channel on the same model. On top of that comes a second axis, which any Polish company notices faster than architecture: how much of your stack (KSeF, Comarch, Subiekt, PayU, tpay) you get off the shelf.

The short version

If you only read three sentences: pick Shopware when you want a mature product with an extension marketplace, automations your team clicks together without a developer, and a deep market of implementers, especially if you operate in Germany, Austria or Switzerland. Pick Endora when the full B2B set should be the starting point rather than a EUR 2,400-a-month plan, and when the Polish stack (KSeF, Comarch or Subiekt, PayU and tpay) has to work natively rather than as an integration written from scratch. If you already run Shopware and have a team that knows it, you most likely should not migrate anywhere. There is one exception, which I describe below.

What each platform actually is

Shopware 6

Shopware 6 is a PHP application on Symfony, with MySQL or MariaDB underneath, Redis, a message queue served by worker processes, and an HTTP cache in front of the application. The stock search runs on the database. A large catalog gets Elasticsearch or OpenSearch wired in, and the Evolve plan offers Advanced Search with autocomplete and relevance tuning.

The tier distinction is the crucial part, though, because it decides what you get in B2B (as of the second quarter of 2026):

• Community Edition: no licence fee, self-hosted. It does not include B2B Components. Since March 2025 a Fair Usage Policy applies: free use ends at EUR 1m of annual GMV. Above that you have to take a paid plan to keep access to your Shopware account and the extension store. Every CE user is required to report GMV.
• Rise: about EUR 600 a month. Flow Builder Professional, the Rule Builder, Shopware AI (formerly AI Copilot), support with an 8-hour response on business days. Still no B2B Components.
• Evolve: about EUR 2,400 a month. This is where B2B Components arrive, along with Advanced Search, webhooks in the Flow Builder and a 4-hour support response.
• Beyond: from about EUR 6,500 a month. Digital Sales Rooms, multi-inventory, subscriptions, 24/7 support.

An important caveat on those figures: these are floors, not fixed prices. Commercial plan pricing scales with revenue, so a shop turning over EUR 5m will pay more on the same tier than a shop turning over EUR 500k.

There is a separate thing worth knowing if you are looking at older builds or at material written two years ago: the old B2B Suite, the monolithic B2B plugin, is in maintenance mode. It gets security patches but no new features, and support ends with Shopware 6.8. New projects are meant to start on B2B Components straight away. For companies sitting on B2B Suite the data migration is not trivial, and the longer you wait the closer you get to a forced cutover.

Endora Commerce

Endora Commerce is a B2B platform written in TypeScript: a Fastify backend with MikroORM and Zod contracts, a server-rendered Next.js storefront, and a React admin panel. Data in PostgreSQL, cache and queues in Redis, search in Meilisearch. Over 60 modules ship as standard: from catalog and price lists to KSeF e-invoicing, product feeds and an AI assistant in the admin panel.

There are no pricing tiers here. Quote requests, the organization hierarchy, credit limits, cart approval, packaging units, shopping lists and quick ordering are not unlocked by a higher plan. They are part of the platform you get.

B2C selling runs on the same installation and the same data model: registering an individual customer quietly creates a one-person organization, so orders, quote requests, invoices and addresses follow the same path as in B2B. B2C is switched on per sales channel.

The delivery model is not a subscription: you do not pay a licence, you get a platform rolled out around your processes, with the code in your own Git repository and the data in your own database. The first working prototype takes four weeks, built with AI-assisted Spec-Driven Development, and an average rollout lands in the region of PLN 60–80k, excluding integrations with external systems, which are quoted separately.

Performance

Split this into two questions, because the answers differ.

An anonymous catalog

Here Shopware with a warm HTTP cache is very strong. A category page served from the full-page cache is a response out of memory, with no PHP and no database involved. It is exactly the mechanism that has powered large German builds for years. Plenty of optimisation playbooks exist for Shopware, and Advanced Search on the Evolve plan copes with catalogs that choke the built-in database search.

Endora renders the storefront server-side in Next.js, with Core Web Vitals targets written directly into the platform requirements (LCP under 2.5 s, INP under 200 ms, CLS under 0.1). Search stays under 200 ms at the 95th percentile, on Meilisearch, which is part of the standard stack rather than an add-on from a higher tier. The demo runs on a catalog of over 120k SKUs, and generating a feed for 100,000 products produces a 38 MB file with peak memory use under 20 MB, because serialization streams.

A catalog with custom prices

Here the picture changes, and this is everyday B2B.

A full-page cache works well as long as everyone sees the same thing. The moment every company has its own price list, contract and volume break, you end up with as many variants of the same page as you have sets of commercial terms, and you hit a cold variant that much more often. On top of that come the pages you cannot cache at all: the cart, the checkout, quotes, order history. That is not a Shopware flaw, it is a consequence of a model where performance rests on caching.

Endora does not have that problem structurally, because it does not build its performance on a full-page cache. The pricing engine settles the price on every request: from the group price list, the individual contract, the volume break and the time window, with deterministic resolution of overlapping rules. A quote pulled through the API goes through the same engine as the cart, so the two match to the last grosz. Personalisation is the base case here, not an exception to the optimisation.

The hardware needed to run it

A caveat: Shopware has years of production builds behind it and well-mapped bottlenecks. Endora’s numbers are architectural targets confirmed by tests and by the scale of the demo, not a decade of production peak-season traffic.

Cost of rollout and running

This is the criterion where the gap is widest and (unusually) the easiest to compute, because Shopware publishes its pricing.

Let us work it through on a concrete example

Take a distributor that needs the full B2B set: company accounts with employees and roles, quote negotiation, order approval and quick ordering. On Shopware that means the Evolve plan, because below it B2B Components simply do not exist.

Exchange rate assumption: PLN 4.30 to the euro. That is the level of the Polish central bank’s mid rate in early August 2026 (PLN 4.3013 in the table of 3 August and PLN 4.3118 in the table of 4 August). Every figure below is the licence alone; on the Shopware side an agency build comes on top.

For completeness: the Beyond plan runs to about PLN 335k a year and about PLN 1.01m over three years, again licence only.

Three conclusions follow, including one that is inconvenient.

First: over three years the gap on the Evolve licence alone is roughly PLN 260–300k in Endora’s favour, and that is before the Shopware-side build is added. The licence recurs and scales with revenue, so as you grow you pay more for the same software. Endora’s cost is one-off plus infrastructure in the region of a few thousand PLN a year.

Second: the Community Edition is not the same platform. Free self-hosted Shopware is a real option, but B2B Components drop out of it entirely: quote management, order approval, employee accounts, organization units, shopping lists and quick ordering. What is left is a shop with the Rule Builder and the option of buying a third-party B2B plugin or building it yourself. Writing just the quote negotiation loop and order approval is not a one-quarter job, and a third-party plugin is another dependency with a lifecycle of its own. On top of that comes the EUR 1m annual GMV threshold (about PLN 4.3m), above which the Fair Usage Policy requires a paid plan.

Third, on the other side: no licence fee does not make Endora free. PLN 60–80k is a real expense and a starting point rather than a ceiling. Integrations with external systems come on top, development after launch costs too, and somebody has to look after the server. The difference is that the budget goes into developing your platform rather than renewing somebody else’s licence, and that it does not grow automatically with your revenue.

Ease of extending the platform

Shopware has an advantage here that belongs at the top, because it is real: the Rule Builder and the Flow Builder. Commercial rules are assembled from conditions in the admin panel, and automations (“when an order exceeds amount X, send a notification and apply a tag”) are built visually, with no developer. On Evolve and Beyond the Flow Builder can also fire a webhook at an external system. For a commercial team that wants to steer promotions and processes itself, that is value no amount of API documentation replaces.

Endora has no Flow Builder equivalent and there is no point hiding it. What it does have is two other things. The first is a pricing rule builder: an AST condition tree over channel, customer group, organization, category and currency, with AND and OR operators down to five levels and deterministic resolution of overlapping rules. The second is an AI assistant in the admin panel. You open the command palette, describe in plain language what should happen, the assistant shows the plan and executes it after your confirmation, within your permissions and with an entry in the audit log. Shopware AI, formerly AI Copilot, is today primarily a content generator: product descriptions, translations, property suggestions, review summaries. Agentic action on store data is on the 2026 roadmap and in beta.

Shopware’s second advantage is vendor replaceability. There are over 1,200 certified agencies, the documentation is extensive and public, and changing partners is a tender rather than a rewrite. Endora hands you the code, but you will not find a team that knows this platform inside out from the first job ad. It is softened by the code being ordinary TypeScript with Zod contracts and documentation rather than a bespoke framework, though not removed entirely.

Shopware does carry a cost that recurs, though: major version jumps. Raising the version means testing every plugin and extension in the new environment, and moving to 6.8 adds the migration from B2B Suite to B2B Components if you are on the old module.

There is a number behind that, because “fast” on its own means nothing: a less complex module takes 1-3 days in Endora, a larger one up to a week. That excludes integrations with an ERP, WMS, CRM, PIM or OMS: those are always heavily bespoke, and writing one takes up to two weeks. Those are always heavily bespoke and can take considerably longer. This is not an advantage over open code as such, because there you can write whatever you want too. It is an advantage of the starting point: you add a module to a working B2B platform instead of first building the thing that is meant to be extended.

Integrations with external systems

In B2B this is usually the deciding criterion, because the platform is almost never the source of truth. The ERP is, and next to it the CRM, the WMS and increasingly an OMS.

Where Shopware wins

An API-first architecture and an app system. Shopware separates the Store API for the customer-facing layer from the Admin API for integrations, with a Sync API for bulk operations. Apps install without forking the core, communicate through a defined contract, and survive platform updates. That is a mature, well-considered extensibility model.

An extension marketplace and a deep DACH market. Plugins for the ERPs used in Germany, for German carriers, for local payment gateways and marketing tools are ready and maintained. If your stack is German or international, Shopware genuinely lets you buy ready-made blocks. Add webhooks in the Flow Builder on Evolve and Beyond, which wire up external systems without writing code.

Where Endora wins

Start with what a few specific searches in the official Shopware store turn up. Searching for “KSeF” returns no extension. The queries “Comarch”, “Subiekt” and “tpay” end the same way. For “PayU” there is a single entry, but it is PayUnity from Wallee, an Austrian product rather than Polish PayU. There is a Przelewy24 gateway. In other words: the Polish settlement and accounting stack largely has to be written yourself on Shopware, or commissioned, and with KSeF mandatory in Poland from February 2026, that is not a detail.

On Endora these things sit in the core, and the ready-made integrations cover:

• KSeF in the platform core: invoices and corrections in the FA(3) format sent through a durable queue, the KSeF number on the document, the official receipt downloadable and a verification QR code on the PDF.
• Comarch ERP Optima and Subiekt GT: adapters as standard, covering item records, stock, prices and sales documents.
• PayU, tpay and Stripe: payment adapters as standard, with no intermediary surcharge.
• A partner API with a live OpenAPI document: generated from contracts in the code, so an integrator never works from an out-of-date PDF.
• API keys pinned to an organization and a channel: the integration sees exactly that company’s prices and limits, and can do no more than its own buyer. There is no parameter that bypasses the rules.
• Idempotent orders: a repeated request returns the same order, never a second one. That holds for concurrent repeats too.
• Webhooks signed with HMAC-SHA-256, up to eight attempts with growing back-off, and a separate view of failed deliveries with manual retry. All of it as standard, not from the Evolve plan up.
• Bulk pricing: one request prices up to 200 lines for a specific organization, guaranteed to match the cart to the last grosz.
• Credentials encrypted with AES-256-GCM, tested on save, masked on read and redacted in the audit log.

The most important part, though, is that the list is not closed. Endora’s integration layer is generic: there is no roster of “supported ERPs”, there is a documented contract. Any ERP, CRM, WMS or OMS is wired up as an adapter module, whether it is Comarch, SAP, IFS, a bespoke system written in-house twenty years ago, or a platform nobody has heard of yet. The adapter registers with the platform, gets its own settings and permissions, and the core stays untouched.

To sum this section up: if your stack is German or international, the advantage sits with Shopware. If it is Polish (KSeF, Optima or Subiekt, PayU, tpay), the advantage moves clearly to Endora, because on Shopware you commission the code anyway, just inside somebody else’s data model.

B2B: where Endora Commerce genuinely stands out

Let me start with what there is no point hiding, because this time the list is long. Shopware’s B2B Components are a serious, native set: employee accounts with roles and permissions, organization units for companies with complex structures, quote management with prices exchanged between buyer and merchant, an order approval workflow, quick ordering by typing part numbers or uploading a file, shared shopping lists, volume pricing, payment on invoice with terms. On top of that, sales channels have been a first-class concept in Shopware from the start, and custom fields are added in the admin panel with no migration. Endora does not win here on the length of the feature list.

It wins on something else, and these differences are worth knowing precisely.

1. B2B is not a pricing tier. On Shopware the full B2B set starts at the Evolve plan, which is about PLN 124k a year in licence alone, at a price that scales with revenue. On Endora, quote requests, the organization hierarchy, credit limits and cart approval are in the platform you get. For a company with online revenue in the single-digit millions of PLN, that is the difference between “we can afford it” and “not this year”.

2. An organization hierarchy that inherits commercial terms. Organization units on Shopware primarily serve to differentiate permissions within a company account. On Endora, headquarters and branches form a tree that commercial terms flow down: order visibility goes downwards, while the price list and credit limit are inherited from the nearest ancestor. A group with several levels does not need its configuration duplicated at every branch.

3. A credit limit as risk control, not an employee budget. Endora gives you a credit limit granted to an organization, reserved atomically at order placement and released once the invoice is paid or the order is cancelled, with payment methods filtered out automatically when the limit runs out. That is a different layer from payment terms, and different from a spending budget assigned to an employee.

4. The none price mode and a negotiation loop closed inside the core. Both systems have quote requests, so the difference is in the detail. On Endora the sales rep quotes with a lead time and a validity date, both sides can revise the terms, and acceptance turns the quote into an order without re-keying a line. Add the none price-display mode, which hides every price element (prices, cart, buy buttons) and routes purchase intent into a quote request. For the range you never sell from a price list on principle, that is a setting rather than a project.

5. The realities of wholesale orders handled directly. Packaging units (pallets, cartons, shrink packs) with unit conversion carried through to the order and the quote. Cart import from CSV or Excel with variant matching. A product comparison tool with PDF export. Returns and complaints with RMA numbering and credit notes. Multi-warehouse with reservations and backorder, as standard, not from the Beyond plan up.

6. Your own order lifecycle, in the admin panel. Order statuses and the transitions between them are configured in the admin panel, with a business event attachable to every transition. On Shopware the order state machine is defined in code and custom transitions are added through a plugin.

There is also something a feature list does not show: every sensitive operation lands in the audit log with the state before and after the change, and a sales rep can step into the customer’s context to see their prices through their eyes, with a visible banner and an audit entry.

Where Shopware has the edge within B2B itself: order approval in B2B Components is a more mature workflow than cart approval on Endora. For corporate procurement with several value thresholds and multiple approval levels, Shopware is closer to a ready-made answer. Employee management with an elaborate permission tree on the buyer’s side is also further developed there. Advanced Search on the Evolve plan is a serious tool for large technical catalogs. And Digital Sales Rooms on the Beyond plan, sales video presentations integrated with the shop, have no Endora equivalent.

When Shopware 6 is the better choice

1. You sell in Germany, Austria or Switzerland. Shopware is the standard there: agencies know it, so do the IT departments on the customer side, and integrations with local ERPs and carriers are ready. Picking the platform your market knows is an argument in itself.

2. Your commercial team is meant to steer automations itself. The Flow Builder and the Rule Builder are Shopware’s strongest suit in day-to-day work. If marketing and sales are supposed to click rules and processes together without raising tickets with IT, Endora asks for more development work here.

3. You already run Shopware with a PHP/Symfony team. Replatforming without a costed business reason is one of the most expensive ways to achieve nothing. If the platform works, stay. And if you are on B2B Suite, plan the migration to B2B Components rather than a migration to a different platform.

4. The procurement requirement reads “a market product with a replaceable vendor”. Over 1,200 certified agencies, public documentation, an extension marketplace and a contract with the vendor is a set that passes a corporate purchasing department without debate.

5. You need a multi-level purchasing process on the buyer’s side. Elaborate employee permission trees, value thresholds, several approval levels. B2B Components cover that in more depth.

6. A large technical catalog with demanding search. Advanced Search on the Evolve plan, with autocomplete and relevance tuning, is a ready-made tool you do not have to tune from scratch.

7. The Community Edition genuinely is enough for you. A simple shop, revenue under EUR 1m a year, no need for advanced B2B features. Then the licence costs nothing, and that is hard to beat.

When Endora Commerce is the better choice

1. You work in a Polish context. Comarch ERP Optima or Subiekt GT behind the scenes, mandatory KSeF, PayU and tpay at the checkout, invoices with your own numbering. On Endora that is native; on Shopware, searching those terms in the official extension store returns nothing.

2. The full B2B set should be the starting point, not a EUR 2,400-a-month plan. If quote negotiation and order approval are things you need from day one, on Shopware you go straight to Evolve. On Endora that is simply the platform.

3. The rollout budget is a real constraint. Roughly PLN 60–80k one-off, plus separately quoted integrations, is a different shelf from a six-figure agency project with an Evolve licence on top. As you grow you pay for infrastructure and development, not for a higher licence tier.

4. Every company has its own prices with you. Group price lists, individual contracts, volume breaks, time windows, overlapping promotions. That is the scenario where a full-page cache stops helping, and Endora treats it as the base case, with API quotes guaranteed to match the cart to the last grosz.

5. Your customer structure is complex and terms are meant to be inherited. Groups of companies, branches picking up the price list and credit limit from headquarters, order visibility flowing down the tree.

6. Your processes are unusual. Selling in pallets and cartons, an order lifecycle with steps that cannot be called “paid” and “shipped”, extra fields that come out of contracts, RMAs with credit notes. On Endora most of that is admin configuration.

7. You have a system nobody supports off the shelf. A bespoke ERP, an ageing WMS, an OMS assembled from several tools. Endora’s integration layer does not assume a particular product on the other side. You write the adapter once, in your own repository.

8. You are launching a distributor portal or several channels at once. Storefront, marketplace, distributor portal, PWA and POS as separate channels of one installation, with separate products, prices, promotions, content, analytics and search index, plus API keys pinned to an organization and a channel for integration traffic.

9. You want to validate an idea quickly before signing off a large budget. A workshop and a clickable prototype on your own data after four weeks, then a joint decision on the full rollout, with no subscription started beforehand.

What Endora Commerce does not have

This section is here so the piece can be used to make a decision, not just to feel convinced.

• An extension marketplace. You will not buy a ready-made plugin for EUR 300. Every unusual feature is development work.
• A visual automation builder on the level of the Flow Builder. That is a real Shopware advantage and it cannot be talked away.
• A deep market of specialists. Certified Shopware agencies number over 1,200. Endora is developed with a partner or your own TypeScript team, who have to learn the platform.
• A decade of references at extreme scale. If your requirement is a documented case at the level of the largest European builds, Shopware has those references.
• Full API integrations with marketplaces. Allegro, Amazon and eBay are handled today with feed templates rather than two-way marketplace API integration.
• Upkeep bundled into a single price. Somebody has to maintain the server, the backups and the updates. You can run that yourself, or contract it to us or to a partner firm under an SLA.
• Multi-level purchase order approval to the degree B2B Components have it. Endora has simpler cart approval enabled per organization, with a decision carrying a reason.
• An equivalent of Digital Sales Rooms.

A short checklist

Answer these questions. The answers usually line up on one side.

1. Is your main market Poland, or the German-speaking countries?
2. Are KSeF, Comarch or Subiekt hard requirements for you, and has anyone costed out writing those integrations from scratch?
3. Do you need the full B2B set from day one, or can you start with a shop and build up?
4. What is your online revenue: do you fit under EUR 1m a year, or will the Fair Usage Policy push you onto a paid plan anyway?
5. Does a licence in the region of PLN 124k a year, growing with revenue, fit your cost model over three years?
6. Is your commercial team meant to click automations together itself, or do changes go through IT anyway?
7. Is owning the code and the data a requirement for you, or just a nice extra?
8. How unusual is your order process: does it fit the standard, or does it have steps nobody else has?

A majority of “the Polish market”, “KSeF and a Polish ERP”, “full B2B from day one”, “a limited budget” and “we want to own the code” points to Endora. A majority of “the DACH market”, “the team should click it themselves”, “we need a market product with a replaceable vendor” and “corporate procurement” points to Shopware.

In summary

Shopware 6 and Endora Commerce are closer to each other than any other pair in this comparison series, because both share the same B2B core: price negotiation and order approval. Shopware gives you a mature product with an extension marketplace, visual automations and a very deep market of implementers. You pay for that with a subscription that grows with revenue, an Evolve threshold that gates the full B2B set, and the fact that the Polish settlement and accounting stack still has to be written. Endora gives you the complete B2B feature set with no pricing tiers, KSeF and Polish ERPs in the core, the code in your repository and a rollout in the tens of thousands of PLN. You pay for that with the platform’s youth, a narrower market of specialists, no marketplace, and a server somebody has to look after.

The simplest test: if your market and your stack are German or international, Shopware will be the more natural choice. If you sell in Poland, to Polish companies, with a Polish ERP and mandatory KSeF, cost out the second option before you sign a three-year subscription.

The Endora admin panel and storefront run online, so the fastest route is to click through the demo yourself or book a free workshop where we walk through your processes and systems and put a number on the first rollout, integrations included.

And if that conversation points to Shopware being the better fit, we will say so plainly. A failed rollout is not a good reference for anyone.