Shopify alternatives for growing brands: seven routes and one test

If the pain fits on an invoice as a number, the best Shopify alternative is usually a higher Shopify plan: the fee for settling payments outside Shopify Payments drops from 2% on Basic to 0.2% on Plus. Leave when the block is the checkout, the product model or your B2B rules, because no plan removes those. From there the realistic routes are BigCommerce if you want to keep renting, and WooCommerce, Saleor, Vendure, commercetools or Medusa if you are taking control. Price two lines on each of them, not one: the platform and the build.

Every price, quota and deadline below was checked at source on 11 August 2026. The implementation cost figures were collected on 12 August 2026, and each one says who publishes it, because a platform vendor has an interest in understating that number and an agency in inflating it. Shopify displays plan prices in the currency of your market, so the only plan-level figures quoted here are the percentages that apply everywhere plus the published Plus starting price in euro.

What actually pushes brands off Shopify

It is almost never that Shopify stopped working. It is that revenue grew, and with it grew line items that were invisible at a smaller size. Six reasons that keep coming up, each with a figure attached.

• The third-party payment provider fee: settle outside Shopify Payments and an extra charge lands on top of your processor rate: 2% on Basic, 1% on Grow, 0.6% on Advanced and 0.2% on Plus. It scales linearly with revenue.
• Checkout is closed below Plus: Shopify documentation restricts checkout styling customisations to Plus merchants. Everything else moves through defined extension points rather than page code.
• The deadlines are set for you: scripts and the checkout.liquid template on the Thank you and Order status pages were sunset on 28 August 2025 for Plus stores, and script tags are sunset for everyone else on 26 August 2026.
• API quotas follow the plan: the GraphQL Admin API runs on a points budget of 100 per second on standard plans, 200 on Advanced, 1,000 on Plus and 2,000 on the enterprise offering. With several integrations running, that is a real speed limit.
• Apps stop being loose change: every capability outside the standard is a separate subscription from a separate vendor, with its own price and its own lifecycle.
• The roadmap belongs to somebody else: your change list is a subset of the vendor list, and the vendor sets the order.

Those six split into two groups, and that split is the whole mechanic of this decision. Some of them are the bill: processing, fees, subscriptions. The rest are the ceiling: checkout, data model, quotas, calendar. A bill can be lowered without migrating. A ceiling cannot.

The upgrade test: what a higher plan fixes and what it never will

Price the plan above you before you price a migration. It settles the question in a large share of cases, and it takes an afternoon rather than a quarter. Three things move, three things do not.

• It cuts the percentage: from 2% on Basic to 0.2% on Plus. On growing revenue that is usually the largest single line on the invoice.
• It raises the integration ceiling: 100 points per second on standard plans against 1,000 on Plus.
• It opens the checkout: styling customisations become available on Plus.
• It does not change the product model: made-to-measure goods, bundles, parameter-driven prices and technical files attached to products remain the same problem on every plan.
• It does not change your B2B rules: per-customer price lists, credit limits and order approvals stay outside the standard on any tier.
• It does not hand back the calendar: sunset dates and product direction remain with the vendor.

The app bill nobody will calculate for you

Shopify publishes no average app spend per store, and we will not invent one. The only credible data sits on your own invoices, and it is worth collecting before anyone starts talking about migration.

Add up the monthly subscriptions and divide by the orders from the same month. That single number tells you more than any pricing page, because it shows whether cost per order rises or falls as the store grows.

Before you pick a direction, check whether leaving pays at all

A migration is expensive whichever destination you choose, so the first question is not “where to” but “whether”. Five markers follow. Recognise two or more and the rest of this page is written for you. Recognise none while the bill keeps climbing, and skip straight to the section on when not to leave.

• The data model does not fit the range: made-to-measure goods, bundles, parameter-driven prices, technical files on the product card. You work around it today with text fields and spreadsheet exports.
• Selling logic is your advantage: per-customer contract prices, credit limits, order approvals, configurators. The thing you win on lives beside the platform instead of inside it.
• ERP integrations set the pace: how fast a change ships is decided by the queue on the ERP, warehouse and API-quota side, not by the storefront.
• You sell through several channels: your own store, marketplaces, wholesale, sometimes physical retail. Each channel wants a different set of data about the same product.
• Revenue is large enough for a percentage to matter: a fee charged on sales stops being a rounding item and becomes a line worth calculating.

What these markers share is that none of them can be written down as an amount to negotiate. They are limits of the model rather than lines on an invoice, which is exactly why a plan upgrade leaves them where they are.

Seven routes, including the six we do not sell

The order runs from the smallest change to the largest, not from worst to best. Each entry carries two numbers: the platform cost from the vendor pricing page, and the build cost, meaning what implementation itself goes for.

1. Stay on Shopify and move up, as far as Plus

The cheapest change and the most often skipped, because it does not look like a decision. A higher plan cuts the third-party payment fee and raises the API quota, and Plus opens checkout styling. The published Plus price starts at 2,250 euro a month on a one-year term and 2,100 euro on a three-year term, with a variable fee for more complex businesses. Build cost: Shopify prices its own custom development at “$2,000 to $20,000 or more”, and agencies quote $30,000 to $50,000 for the Advanced to Plus move. Pick this if your problem can be written as an amount rather than as “we cannot”.

2. BigCommerce

The closest hosted equivalent, with one real difference: the documentation presents Open Checkout and the Checkout SDK as supported routes to your own checkout and ties neither to a plan. Pricing runs across four tiers, from $29 a month billed annually to Performance from $1,499, with an open payment provider fee of 2%, 1% or 0.6%. Build cost: the only vendor on this list that publishes fixed implementation prices, in packages at $15,000, $30,000 and $50,000, with ERP integrations and a stated timeline of 90 days or less. Pick this if checkout is your ceiling but hosting and maintenance are not what you want to take on.

3. WooCommerce

A free open-source plugin for WordPress with no revenue share, a very large extension market and an equally large pool of people who can work on it. Platform cost moves entirely into hosting and extensions: WooCommerce publishes $25 to $350 a month for hosting and $29 to $299 a year per extension. Build cost: the vendor publishes nothing, so the only ranges come from agencies, and they are local. Polish agencies quote roughly 2,800 to 5,800 EUR for a simple build and 5,800 to 14,000 EUR with integrations. Pick this if the catalogue fits the standard and maintenance has to stay cheap and hireable almost anywhere.

4. Saleor

Headless with a BSD 3-Clause core and GraphQL as the primary interface, plus a vendor cloud with published pricing: Select at $1,599 a month up to $200,000 monthly GMV and Volume at $3,999 up to a million, with a percentage on anything above the cap. Build cost: nobody publishes one. Saleor sells its own services (Accelerator at $6,000, Forward Deployed Engineering at $12,000), but those are support packages rather than the price of a finished store. Pick this if you are building a custom storefront and prefer an open core with the option of vendor hosting.

5. Vendure

A TypeScript commerce backend closest in architecture to Medusa, and the licence is the thing to read first: the core is dual licensed, GPLv3 by default with a separate Vendure Commercial Licence available by agreement. The pricing page quotes each project individually rather than publishing tiers. Build cost: not published by the vendor and not published by anyone else either, which puts it in the same position as Saleor. Pick this if you want a Node.js core with a commercial licence path and your legal team is comfortable with GPLv3 in the default case.

6. commercetools

A commercial composable platform aimed at large organisations: API-first working, order-based billing and implementation normally through a partner. No pricing is published; the page promises “transparent, predictable pricing” and routes you to sales without quoting a rate. Build cost: equally unavailable. This is the one entry on the list where neither the licence nor the implementation can be priced from public sources, and the numbers circulating online come from competitors rather than the vendor. Pick this if you run several brands or markets, have an eCommerce team and an enterprise budget to match.

7. Medusa

A modular commerce backend in TypeScript with an MIT-licensed core. Platform cost: zero. No licence fee, no revenue share, hosting a separate line from $29 a month on Medusa Cloud. Build cost: nobody publishes one, so here is ours. A build from about 14,000 EUR, typically 18,500 to 46,500 EUR, in 8 to 12 weeks. Separately, AI automation from about 1,400 EUR per workflow: product descriptions, catalogue translation, marketplace feeds, order status, attribute mapping. Plus AI agents running the store on your own backend and data. We are not the only team that builds on Medusa. Pick this if your selling rules belong in the core.

Platform cost, build cost and the ceiling that stays

The table below runs on three axes rather than one. What the platform charges, what it costs to stand it up, and which ceiling moves with you. Only together do they describe the decision, because every option here has to be built by someone, including the one with no subscription.

Platform prices checked on vendor pages on 11 August 2026; implementation cost data collected on 12 August 2026. The build cost column mixes figures of different origin from different markets, so read it as an order of magnitude rather than a quote. Our own ranges are set in Polish złoty (from 60,000 PLN, typically 80,000 to 200,000 PLN, from 6,000 PLN per workflow) and converted at a rounded 4.30 PLN to 1 EUR, the NBP mid rate of 11 August 2026. No performance comparisons, because we have no measurements of our own, and no negotiated terms, because they are not public.

What these numbers do not prove

We write this as a company that builds on Medusa, which is precisely why it needs stating. The public data does not support a claim that building on Medusa is cheaper than building on Shopify Plus at comparable scope. Four caveats belong next to that table.

• The ranges overlap: the lower end of the published Plus replatforming bands sits in the same territory as quotes for open-core builds. Medusa publishes no implementation costs, and a single data point from an interested agency is not a range.
• Part of the gap is geography, not technology: agencies in Poland quote $50 to $99 an hour on Clutch, certified Shopify Plus partners $175 to $300. The same work costs less in Poland whatever the platform is.
• We are not faster: a store on a ready-made theme goes live in 2 to 4 weeks and we will not match that. On complex scope we are comparable: 8 to 12 weeks against 12 to 16 for a move from WooCommerce or BigCommerce to Plus.
• Migrations measured in months are a different project: the 20 to 28 week figure applies to leaving Salesforce Commerce Cloud or SAP, quoted separately at $300,000 and up. Setting it beside a mid-market number compares two unrelated projects.

The argument that holds is elsewhere: not in the cost of going live once, but in the three-year bill and the cost of changing things afterwards. That one is laid out line by line in BEAM and Medusa against Shopify Plus and an agency.

When not to leave

This section is here in earnest rather than for balance. We build owned platforms, and in four situations we still advise against leaving Shopify, even when the list at the top of this page sounds familiar.

• The problem is the app bill, not the platform ceiling: reviewing subscriptions and cancelling three you no longer use costs an afternoon. A migration costs a quarter.
• Nobody is available to run an owned platform: on Shopify, hosting, updates and payment compliance sit with the vendor. An owned platform without an owner degrades more quietly and more expensively.
• A migration would eat the budget that sells: if the same money could go into range, a new market or people, and the platform is merely irritating, the order of priorities is obvious.
• The plan above costs less than the move: set the annual difference in plan fee and payment fee against the cost of building and running something new. The table above gives you the order of magnitude for that second figure. Very often that comparison is the entire conversation.

The cheapest good decision is sometimes to change nothing and repeat the arithmetic in a year. If a genuine block appears in the meantime, it will show up in the numbers rather than in the mood. The symptoms are catalogued in the signs you have outgrown a SaaS platform.

What this list leaves out

The boundaries, stated plainly, so the page can be used as a source:

• Performance under load: we have measured none of these platforms and we do not repeat other people’s benchmarks.
• Negotiated terms: larger Shopify Plus accounts move to a variable platform fee based on revenue, BigCommerce Performance runs on contracted terms, and commercetools publishes no rates at all.
• Local pricing: Shopify shows plan prices in the currency of the visitor market, so only the percentages and the published Plus starting price are quoted here.
• Your own quote: the figures in the build cost column describe other people’s projects in other people’s markets. Yours follows the catalogue, the number of integrations and how many processes move with the store.
• Other candidates: PrestaShop, Sylius, Spryker and Adobe Commerce follow a different cost logic and deserve a comparison of their own.

Settling it from your own numbers

Four steps, each of which needs only data you already have. No industry averages, because in this decision they are worth nothing.

• Work out platform cost per order: plan fee, processing, third-party provider fee and app subscriptions across twelve months, divided by orders.
• Keep a “we cannot” list: for one month, write down everything the platform or your agency ruled out. That list is your real specification.
• Check how much of it a higher plan removes: if it clears more than half, the rest of the conversation is about pricing tiers rather than migration.
• Price both ends, the build and the running: an owned platform has an entry cost and an annual cost, and a comparison missing either one is unfair to somebody.

The three-year arithmetic is laid out in the true cost of owning your platform, the way to pick a destination in criteria instead of a ranking, and the head-to-head between an open core and Shopify in Medusa vs Shopify.

If the numbers point to leaving, the mechanics of the move are on our Shopify migration page and the sequencing that keeps the store trading is in migration without stopping sales. Our own commercial model is a fixed scope at a fixed price, quoted after a short conversation: the platform families are compared on eCommerce platforms and the way we price is on the pricing page.

Checked in August 2026

Platform prices and quotas were checked at source on 11 August 2026, and the implementation cost data was collected on 12 August 2026. We revisit this page quarterly, because an out-of-date comparison does more harm than good. If something has moved since, tell us and we will correct it together with the check date.

• Shopify: pricing, Plus pricing, rate limits and checkout styling.
• BigCommerce: pricing, checkout customisability and API rate limits. WooCommerce: hosting and extension pricing.
• Saleor: pricing (vendor service packages are listed there too) and core licence. Vendure: pricing and core licence. commercetools: pricing page. Medusa: Medusa Cloud pricing and repository licence.
• Implementation costs: Shopify on custom development (“$2,000 to $20,000 or more”) in its guide to ecommerce website costs. The platform licence as 20 to 40% of total spend: Elogic, Ecommerce Replatforming Cost Index 2026 (updated 3 August 2026).
• Attributed to the publisher: the B2B Quick Start Accelerator packages ($15,000 / $30,000 / $50,000, ERP integrations, “90 days or less”) come from a BigCommerce announcement dated 24 July 2025 and describe a partner programme for B2B sellers, not general implementation pricing. Migration bands to Plus come from the Elogic Replatforming Cost Index. Polish WooCommerce ranges and hourly rates ($50-99 in Poland, $175-300 for certified Plus partners) come from published agency guides and Clutch profiles, collected 12 August 2026.
• Our ranges and pricing model: the pricing page. Set in Polish złoty and converted at a rounded 4.30 PLN to 1 EUR, the NBP mid rate of 11 August 2026 (4.2996), table 154/A/NBP/2026.