Solidus vs Shopify vs Shopware vs Magento vs WooCommerce: Which Platform for a DACH SME

Why I am writing this one

Every few weeks I get a version of the same call. Someone runs a shop doing somewhere between half a million and five million euros a year. They are on Shopify, or on an old Shopware 5, or on a WooCommerce install that a nephew set up in 2019, and something has started to hurt. The app bill has crept past what the platform itself costs. The agency quoted four weeks for a change to the checkout that sounds like an afternoon. Or the platform vendor has announced that the version they are on is end of life and the migration path looks suspiciously like a rebuild.

And then the question: what should we be on instead?

This is my long answer. I build on Solidus, so you know where I stand. But I have migrated shops off every platform on this list, and onto most of them at some point in the last fifteen years, and I would rather you pick the right one for your business than the one I happen to like. If Shopify is right for you, I will say so. I have said so to prospects and lost the job, and I still think it was the correct call.

One more thing before we start. This is written for the person who signs the invoice. Your developer will have opinions about PHP versus Ruby and about whether Symfony is nicer than Rails, and those opinions are fine, but they are not what decides whether your shop makes money in 2029. What decides that is who owns the thing, what it costs over five years, whether it does what German and Austrian customers expect at the checkout, where the customisation ceiling is, and how hard it is to leave. So that is the order I am going to take.

The question behind the question: rent or own

Strip the marketing off all five platforms and you are left with two models.

Renting. Shopify is the pure form of this. You pay monthly, the vendor runs everything, you customise within the boundaries they give you, and the moment you stop paying the shop is gone. Shopware has been moving in this direction since it took outside investment, and its paid plans are now the centre of gravity even though the community edition still exists. Adobe Commerce, the paid Magento, is rented software you host yourself, which is the least attractive combination of the two.

Owning. Solidus, Magento Open Source and WooCommerce are code you run. Nobody can switch it off. Nobody takes a percentage. The flip side is that nobody else is responsible for it either, and if you have ever owned an old house you know exactly what that means.

Most owners I talk to have never made this decision consciously. They picked whatever their first agency knew, or whatever had the nicest onboarding video, and five years later they are discovering that the choice they did not make has a price. So make it consciously now. Here is my rough rule.

If ecommerce is a sales channel for you, one of several, and the shop itself is not where you compete, rent. Pay Shopify, keep it simple, put your energy into product and marketing.

If the shop is the business, if the way people buy from you is part of why they buy from you, own. A B2B wholesaler with customer specific price lists, a brand with a configurator, a retailer with click and collect across twelve locations, a subscription box, a marketplace: these businesses eventually hit the walls of a rented platform and then they pay twice, once for the workarounds and once for the migration.

Everything below is detail on that one distinction.

Shopify: brilliant until it is not

I want to be fair to Shopify because it is genuinely the best product in this list at what it does. The onboarding is smooth. The admin is pleasant. Shopify Payments works in Germany and Austria now, the theme store is full of decent themes, and you can have a shop live in a weekend. For a founder testing a product, there is nothing better.

The problems show up in three places, and they show up at roughly the size where a shop becomes a proper business.

The checkout. Unless you are on Shopify Plus, the checkout is not yours. You can change colours and a logo. You cannot change the flow, you cannot add a step, you cannot put a custom field where you need it, and the old way of hacking it through checkout.liquid has been switched off. Plus gives you checkout extensibility, and Plus starts at around 2,300 US dollars a month on a three year term. For a shop doing two million a year that is a meaningful line, and it is a line you pay forever.

The apps. This is the bit that nobody budgets for. A real Shopify shop in DACH runs somewhere between ten and twenty five apps. Reviews, bundles, a proper German legal texts app, a cookie banner that actually does consent mode, invoicing that produces something your Steuerberater will accept, back in stock, upsell, loyalty, a search that works, translations, a B2B add on if you sell to trade. Each one is ten to a hundred and fifty euros a month. The app bill for a mid sized shop I audited last year was eleven hundred euros a month, more than three times the plan fee, and the owner had stopped noticing because it was spread over twenty two Stripe receipts. And every app is another company with a foot in your customer data, which is a GDPR conversation on its own.

The fees. If you use Shopify Payments the transaction fee is baked into the card rate. If you want to use a different provider, say because your Kauf auf Rechnung partner needs it or because you negotiated a better rate with a German acquirer, Shopify charges an additional percentage of every order for the privilege of not using theirs. On the standard plans that is between 0.6 and 2 percent. On two million of revenue the difference between the plans is not pocket money.

Then there is the exit. Products, customers and orders come out as CSV. Your theme is Liquid and useless anywhere else. Your apps do not come with you. Your SEO is tied to a URL structure you cannot fully control. None of this is a scandal, it is just what renting means, and you should price it in.

Where Shopify is right: new brands, small catalogues, businesses where the shop is a channel and not the moat, anyone without the appetite for owning software. If that is you, stop reading, go and pick a theme.

Shopware: the DACH default, and why that is not automatically a recommendation

If you ask a German agency what to use, the answer is Shopware, and it has been for a decade. There are reasons for that. Shopware is built in Germany, it thinks in German legal requirements from the start, the Rechnungskauf integrations are mature, the plugin store has a German legal texts plugin for every eventuality, and there are hundreds of agencies who know it. For a shop that only ever sells in DACH, that ecosystem is a real advantage.

But I want you to separate two things: Shopware the software, and Shopware the market position. The market position is excellent. The software is a large Symfony application with an admin built in Vue, and it is competent. It is not, in my experience, easier to customise than the alternatives, and it is considerably heavier than it needs to be for most SMEs.

Three things worry me about it for a small or medium shop.

The direction of travel. Since the investment round in 2022 the commercial focus has shifted to the paid Rise, Evolve and Beyond plans, which start at several hundred euros a month and scale with your revenue. The community edition is still free and still open source, and I am not suggesting it will vanish, but the features that matter increasingly land in the paid tiers first. If you build on the free edition you should assume you are on your own for the interesting bits.

Migration history. Anyone who lived through Shopware 5 to Shopware 6 will tell you it was not an upgrade, it was a rebuild with a data import. Shopware 5 went end of life in 2024 and a lot of shops paid for a second build to stay on the same vendor. That is not unique to Shopware, Magento did the same, but it should temper the "we will just upgrade later" assumption.

Agency dependence. Because the ecosystem is so agency driven, most Shopware shops I see are entirely dependent on one agency, on a retainer, for anything beyond editing a product. That is fine while the relationship is good. When it sours, and I have picked up three of those in the last two years, you discover that the customisations are undocumented plugins that only that agency understands.

Where Shopware is right: a DACH only retailer with a conventional catalogue, a budget for an agency retainer, and a preference for a vendor everyone has heard of. That is a perfectly reasonable set of preferences. Just go in knowing it is closer to renting than the "open source" label suggests.

Magento: the platform that ate its own users

I ran Magento shops for years and I have a soft spot for it, so take this with that in mind.

Magento Open Source is a serious piece of software. It handles enormous catalogues, complex B2B, multiple stores and currencies, and it has been doing so for fifteen years. Adobe Commerce, the paid version, adds a licence fee that starts somewhere north of twenty thousand euros a year and scales with revenue, plus features you probably do not need.

The problem with Magento for an SME is not capability. It is that the capability is expensive to operate. A properly run Magento 2 store needs Elasticsearch or OpenSearch, Redis, a page cache like Varnish, cron jobs that behave, and a hosting environment that costs three hundred to six hundred euros a month before anyone touches the code. Upgrades are quarterly and they break things. The default frontend is slow enough that a whole company, Hyvä, exists purely to replace it, and Hyvä is itself a licence. Agencies charge Magento rates, and Magento rates are the highest in this list.

Then there is the question of where Adobe is taking it. Adobe bought Magento in 2018 for the enterprise business. The open source edition has been drifting for years, the community forked it into Mage OS out of frustration, and Adobe's own roadmap is about a cloud service that has nothing to do with the software you would be installing. I would not start a new SME shop on Magento in 2026. I would keep an existing one running if it is working, and I would plan the exit.

Where Magento is right: you already have it, it works, and you have the budget to keep it working. Or you are a large B2B operation with a catalogue in the hundreds of thousands and an in house team. Neither of those is the typical DACH SME.

WooCommerce: WordPress with a till

WooCommerce is the most installed ecommerce platform in the world, and the reason is that it is free and it bolts onto the most installed CMS in the world. If you already have a WordPress site with a good content operation and you want to sell a few dozen products, WooCommerce is the obvious answer and I would not argue with it.

For anything beyond that, I have three problems with it.

Security. WordPress plugins are the single largest attack surface on the web. Not WordPress core, which is fine, but the ecosystem of forty thousand plugins of wildly varying quality, of which a real WooCommerce shop uses twenty or thirty. Every one of them is a door. I have cleaned up more hacked WooCommerce shops than every other platform on this list combined, and a hacked shop in the EU is a data breach with a seventy two hour notification clock, not just a bad afternoon.

Performance at scale. WordPress is a blog engine with a database schema to match. Products are posts. Attributes are post meta. Past a few thousand SKUs, or a few hundred orders a day, it starts to need the kind of caching and hosting attention that erases the "free" in the price tag.

DACH compliance is a plugin, not a feature. To sell legally in Germany you need Grundpreise, the correct button text, Widerruf texts, delivery times, and a dozen other things, and in WooCommerce all of that comes from Germanized or German Market. They are good plugins. They are also a dependency on one small company for whether your shop is legal, and every WooCommerce update is a small gamble on whether they still work together.

Where WooCommerce is right: content led businesses, small catalogues, and anyone who already lives in WordPress and has someone who keeps it patched. If you are doing more than a couple of hundred thousand a year through it, start thinking about what comes next before you are forced to.

Solidus: the one you own outright

Now the one I actually build on, and I will try to be as hard on it as I was on the others.

Solidus is an open source ecommerce framework written in Ruby on Rails. It started as a fork of Spree in 2015 when a group of agencies and merchants decided they wanted a platform that prioritised stability over features, and it has been maintained that way since, with a small core team and a community of agencies who run real shops on it. It is not a product you install and click through. It is a set of well designed building blocks, orders, products, variants, stock locations, promotions, payments, shipping, tax, that you assemble into your shop, with a solid admin on top.

That description is also its biggest drawback, so let me lead with the honest bit.

You cannot launch a Solidus shop in a weekend. There is no theme store. There is no marketplace of five hundred plugins. The frontend is yours to build, which is wonderful if you want a shop that looks like your brand and not like everyone else's Shopify theme, and expensive if you just want a shop. A typical Solidus build for an SME, done properly with tests, a real design, DACH payments, invoicing and the legal texts, runs from forty to ninety thousand euros depending on how much is genuinely custom. That is more than a Shopify launch and roughly the same as a Shopware agency build.

You need someone who knows Rails. Not many agencies in DACH do. That is a real constraint. It is also, I would argue, part of why the shops that run on it tend to be well built: it selects for developers who chose the framework on merit rather than because it was what the market handed them.

The extension ecosystem is small. There are official extensions for Stripe, PayPal, Braintree, internationalisation, multi domain and a handful of others, and they are good. But if you want a specific feature, the answer is usually "build it", not "install it". For a shop with a conventional catalogue and no unusual requirements that is a cost with no offsetting benefit.

Now the reasons I chose it anyway.

Nobody owns you. No licence fee, no revenue share, no plan tier that unlocks your own checkout. The code is BSD licensed, which means you can do anything you like with it, including keeping it private, and the version you deploy today will still run in ten years whether or not anyone maintains the project. Your hosting is whatever you want it to be. A Solidus shop doing a couple of million a year runs happily on a single virtual server costing forty to a hundred euros a month, deployed with Kamal, which is the same tooling Basecamp uses to run their own products. Compare that to a Magento hosting bill.

Everything is customisable, and I mean everything. The checkout is a set of Rails controllers and views. You want a step that asks for a delivery slot? You add it. You want the order total to change based on a customer's negotiated price list? That is a promotion rule and a pricing decorator, an afternoon's work with tests. You want build your own hamper, where the customer picks eight items and the total is fixed? I built exactly that for one of my demo shops and it is a few hundred lines. On Shopify the same feature is an app with a monthly fee and a compromise. On Shopware it is a plugin and a quote.

It is a Rails app, so it behaves like one. That sounds like a developer point but it is really a business one. It means you get the same test tooling, the same deployment story, the same security scanning and the same hiring pool as every other Rails application. Your shop can share code with your internal tools, your customer portal, your subscription system, because it is all the same framework. I write about how I test these builds in my piece on Cucumber, and the short version is that every feature in the shop has an executable specification that you, the owner, can read.

The API is first class. Solidus ships with a full JSON API, and because the storefront is yours you can make the whole shop headless if and when you need a mobile app, a marketplace feed or a second storefront. That is not an upgrade tier. It is just there. I keep every demo shop I run API ready from day one for exactly this reason, and I will write a separate piece on it.

Upgrades are boring. This is the thing I appreciate most after years of Magento. Solidus follows semantic versioning, deprecations are announced a version ahead, and the core team has a stated policy of not breaking things for the sake of a new feature. A Rails upgrade on a well tested shop is a day, not a quarter. There has never been a Solidus 5 to 6 style rebuild and the project's whole identity is that there will not be one.

Where Solidus is right: shops where the buying experience is part of the product. B2B with price lists and quotes. Configurators and bundles. Multi location retail. Subscriptions. Anything that needs to integrate deeply with an ERP, a warehouse or a bespoke process. Brands who want a shop that looks like nothing else. And, frankly, any owner who has been burned by a platform vendor and has decided they would rather own the house.

The DACH checklist, because this is where platforms actually fail

Owners tend to compare platforms on features and design. Shops in Germany and Austria actually fail on the boring stuff. Before you look at a single theme, put every candidate through this list.

Kauf auf Rechnung. Roughly a quarter of German consumers want to pay by invoice after delivery, and for B2B it is not optional. Every platform can do it via Klarna, PayPal or a specialist like Unzer, but the difference is in what happens when the customer does not pay, how the dunning runs, and whether your accounting sees it correctly. I wrote about the broader point in why German buyers avoid subscriptions and expect pay on invoice. Shopify makes this harder than it should be by penalising external providers. Shopware and Solidus handle it natively with the right integration. WooCommerce does it through Germanized plus a payment plugin. Magento does it, at Magento prices.

SEPA, EPS, Twint. Direct debit for Germany, EPS for Austria, Twint for Switzerland, and Giropay is gone. If you sell in all three countries your checkout needs to show the right methods to the right customer and nothing else. On a rented platform that is a settings page if you are lucky. On Solidus it is a payment method per zone, which is how it should be.

The button rule. In Germany the final order button must say "zahlungspflichtig bestellen" or an equivalent, and the summary immediately before it must show the essential characteristics, total price and delivery costs. It sounds trivial. Shops get Abmahnungen for it every month. Any platform with a DACH ecosystem has this solved. If yours does not, it is not the platform for you.

Grundpreise and the Preisangabenverordnung. If you sell anything by weight, volume or length you must show the price per base unit next to the price. Shopware and the WooCommerce German plugins do this out of the box. Shopify needs an app or theme work. Solidus needs you to add a field and render it, which is ten minutes, but it is ten minutes someone has to remember.

OSS VAT. Since July 2021, once you sell more than ten thousand euros a year cross border within the EU, you charge the destination country's VAT and report it through the One Stop Shop. That means twenty seven VAT rates, product category exceptions, and reporting that your Steuerberater needs in a specific shape. Every platform on this list can charge the right rate. Not every platform produces reporting that does not require a spreadsheet at quarter end. Ask to see the export before you sign.

Invoices and GoBD. German customers, and German tax inspectors, expect a proper invoice with sequential numbers that cannot be changed after the fact. Shopify's built in invoicing does not meet that bar and you will need an app. WooCommerce needs a plugin. Shopware has it. Solidus needs an invoicing module, which is usually part of the build, and it is worth building right the first time because e invoicing for B2B is being phased in across Germany and your customers will start asking for XRechnung or ZUGFeRD formats.

Barrierefreiheit. The Barrierefreiheitsstärkungsgesetz has applied to B2C online shops since June 2025, with a carve out for the very smallest businesses. Your storefront needs to meet WCAG 2.1 AA in practice. Theme based platforms are hostage to their theme's quality here. A custom frontend is only as good as the people who built it, which is an argument for insisting on accessibility in the specification, not a point for or against any platform. I covered the requirements in my accessibility piece.

Data location and processors. Under GDPR every app, plugin and hosted service that touches customer data is a processor you need a contract with and a line in your records of processing. Twenty Shopify apps is twenty processors. A self hosted Solidus shop with Stripe and a transactional email provider is two. That is not the deciding factor, but if you have ever had to answer a Datenschutzbehörde enquiry, it is not nothing either.

The five year bill

Here is where I lose friends, because I am going to put numbers on it. These are ranges from real projects, rounded, for a DACH SME doing between one and three million euros a year with a catalogue of a few hundred to a few thousand SKUs, selling in two or three countries. Your shop will differ. The point is the shape, not the decimals.

Shopify, standard plans. Plan fee around four thousand euros over five years on Advanced. Apps at four hundred to a thousand a month, call it thirty to sixty thousand. Theme customisation and ongoing agency work, twenty to forty thousand. Extra transaction fees if you use an external payment provider, potentially tens of thousands depending on volume. Somewhere between sixty and a hundred and twenty thousand, with the important caveat that the checkout stays locked the whole time.

Shopify Plus. Add roughly a hundred and forty thousand in plan fees over five years to the above. Now you are at two hundred thousand plus and you still do not own anything.

Shopware. Free on the community edition, or thirty five to a hundred thousand over five years on the paid plans depending on tier and revenue. Agency build forty to eighty thousand. Plugins and retainer, another forty to seventy. Hosting ten thousand. Between a hundred and two hundred and fifty thousand, and most of the ongoing spend goes to the agency retainer.

Magento Open Source. Build sixty to a hundred and twenty thousand. Hosting twenty to forty. Upgrades and maintenance fifteen to twenty five a year, so seventy five to a hundred and twenty five. Around a hundred and sixty to two hundred and eighty thousand, and if you are on Adobe Commerce add the licence, which on its own is more than a Solidus build.

WooCommerce. Build ten to thirty thousand. Plugins and hosting fifteen to twenty over five years. Maintenance twenty five to fifty. Fifty to a hundred thousand. Plus an unquantifiable line for the year something gets hacked, which I am not being flippant about, it is the single most common reason WooCommerce shops call me.

Solidus. Build forty to ninety thousand. Hosting three to six over five years. Maintenance, meaning Rails and dependency upgrades, security scanning and small improvements, eight to fifteen a year, so forty to seventy five. No licence, no revenue share, extensions are free. Between eighty five and a hundred and seventy thousand.

Look at those honestly and Solidus is not the cheapest. WooCommerce is cheaper if nothing goes wrong. Standard Shopify is cheaper if you never need what it will not let you do. What is different about the Solidus number is what it buys. Almost all of it is spent on your shop, on features that are yours, built by someone who documents them, tested so they keep working. Almost none of it is rent.

And the number that matters most is not on the list, because it is the cost of the thing you cannot do. The B2B price list Shopify will not let you build without Plus. The checkout step that costs four weeks and a compromise. The migration you are forced into when a vendor sunsets a version. I have watched owners spend more on workarounds in year four than the entire build would have cost in year one.

The customisation ceiling

Every platform has a point where the answer to "can we do X" changes from "yes" to "sort of" to "not really". Knowing where that point is matters more than any feature list. Roughly:

Shopify. Themes: high. Product and content: high. Checkout: very low unless Plus, then medium. Business logic such as pricing, stock and order flow: low. You work around it with apps, and the apps have their own ceilings.

Shopware. Themes: high, with effort. Checkout: medium to high through plugins. Business logic: high if your agency is good, because it is a real application framework underneath. The ceiling is your agency, not the software.

Magento. Everything: high. The ceiling is your budget.

WooCommerce. Themes: high. Checkout: medium, and fragile. Business logic: medium, and every customisation makes the next update riskier.

Solidus. Everything: high, and it stays high because there is no vendor layer between you and the code. The ceiling is whether you have a Rails developer, and how much you are willing to spend building rather than installing.

How you leave

Nobody thinks about this at the start and everybody thinks about it at the end. Ask each candidate: if in four years I want to move, what do I take with me?

From Shopify you take CSVs of products, customers and orders. Not your theme, not your apps, not your reviews unless the review app lets you export, not your checkout. From Shopware you take your database and your custom plugins, which are useful if your next platform is Shopware. From Magento you take a large database and a collection of modules, and the migration project is priced accordingly. From WooCommerce you take a WordPress database in which products are posts, which is exportable but not pretty.

From Solidus you take a Rails application and a PostgreSQL database that you already own, on a server you already control. There is no leaving in the sense the others mean. You could hand it to a different Rails developer tomorrow. That is the whole point of owning.

Hosting and security, briefly

Rented platforms handle hosting and the security of the platform itself, which is a genuine benefit and the strongest argument for Shopify. What they do not handle is the security of your twenty apps, your staff accounts and your integrations, which is where the actual incidents happen.

Owned platforms are your responsibility, which means they are as secure as the people running them. For Solidus that means the standard Rails toolchain: Brakeman for static analysis, dependency auditing, a pen test per release, and a deployment that does not expose anything it should not. I bake that into every build and record it in my client portal so there is evidence when an auditor or an insurer asks. I will write a full piece on it. For Magento it means a lot of ongoing hosting attention. For WooCommerce it means keeping thirty plugins patched forever.

Who should pick what

Let me try to be as direct as the format allows.

Pick Shopify if you are launching, if your catalogue and checkout are conventional, if the shop is a channel rather than the business, or if you simply do not want to own software. Budget for the apps. Know that the checkout is not yours.

Pick Shopware if you sell only in DACH, you want a vendor every agency knows, and you have a retainer budget. Read the paid plan pricing before you fall in love with the free edition.

Stay on Magento if you are already there and it works. Do not start there.

Pick WooCommerce if you are a content business selling a small catalogue and you have someone who keeps WordPress patched. Set a revenue threshold at which you will reconsider, and write it down.

Pick Solidus if the way people buy from you is part of what you sell, if you have requirements that the others call "enterprise" and charge accordingly, if you have been burned by a vendor before, or if you have done the five year maths and decided you would rather spend the money on your own shop than on rent. And accept that it costs more in year one and you need a Rails developer you trust.

What I would do if it were my money

I did do it with my own money, which is why the demo shops on this site run on Solidus and not on a Shopify trial. When I set up the first one, a gift and hamper store with a build your own hamper feature, five currencies, two languages and regional delivery rules, I costed the same thing on Shopify for comparison. It needed Plus, four apps, and a compromise on the hamper builder that would have made it a worse product. The Solidus version is a few hundred lines of Ruby, fully tested, running on a server that costs less than one of those apps.

That is a story about one shop and you should weigh it accordingly. But the pattern behind it is the one I see every time. Rented platforms are cheaper right up until you want something of your own, and then they are the most expensive option there is.

If you want a second opinion on your own numbers, or you want to see the demo shops with a login rather than screenshots, get in touch. I will tell you which platform I would pick for you, and it will not always be mine.