We are not here to talk anyone off Shopify Plus. It is a genuinely good platform: fast to launch, reliable under load, enormous app ecosystem, and the operational burden is somebody else's problem. Most brands running on it should stay.
But a platform that fits at £2m does not automatically fit at £40m, and the transition is gradual enough that teams normalise the friction rather than notice it. These are the seven signals that the constraint has become structural, and — more usefully — the four that look identical but mean something else entirely.
The Seven Real Signals
1. Your Pricing Logic Lives Outside the Platform
The clearest signal there is. If contract pricing, volume tiers, customer-specific catalogues or quote-based selling are being maintained in a spreadsheet, a middleware layer or somebody's head — because the platform cannot express them — you are paying for a commerce engine and not using it as one.
Shopify Functions has widened what is possible here considerably, so check the current position before concluding. But if you have already reached its edges, that edge is real.
2. The App Bill Is Load-Bearing and Growing
Not the size of the bill — the dependency. If four apps are individually critical to how you trade, you have four vendors who can raise prices, change behaviour or shut down, and you have no leverage with any of them.
A high app count on its own is a hygiene problem rather than a platform one; we cover the cleanup in the Shopify app audit. It becomes a platform signal when the apps are load-bearing and unreplaceable.
3. You Are Building Around the Checkout
Plus gives you meaningful checkout extensibility, and that is usually enough. But if your business requires flows the checkout does not accommodate — complex fulfilment selection, quote approval before payment, split-tender arrangements, industry-specific compliance steps — and you have built a parallel process outside it, that is a structural mismatch rather than a customisation gap.
4. Multi-Entity Complexity Exceeds Multi-Market Tooling
Shopify Markets handles multi-currency, multi-language and regional pricing well. It handles multiple legal entities with different tax registrations, different fulfilment networks, different product catalogues and consolidated group reporting less well. Brands that have grown by acquisition hit this earlier than brands that have grown organically.
5. Your Catalogue Model Does Not Fit the Variant Model
Configurable products with engineering constraints, made-to-order goods with dependent options, products priced by dimension or weight, kits assembled at order time. If you have modelled your catalogue around the variant limit rather than around your products, the platform is shaping your business rather than serving it.
6. The Platform Fee Would Fund a Team
Purely arithmetic, and it turns over at a definable point. Add the Plus fee, the app subscriptions, and any payment-processing premium you pay for using a non-native gateway. When that annual total meaningfully exceeds what an engineering team plus infrastructure would cost, the economics have inverted.
Note the honest caveat: an owned platform's cost does not stop at the team, and most business cases understate ownership badly. We laid out that arithmetic in what a custom platform really costs over three years. Do it before you use this signal.
7. You Need Channels the Platform Was Not Built For
A native app sharing commerce logic with the web. In-store systems. Partner or reseller portals. Embedded commerce in someone else's product. Agent-mediated purchasing through an interface you control. Each of these is possible against Shopify's APIs; several of them together start to describe a headless architecture that you are building anyway, awkwardly.
The Four Signals That Fool Everyone
These feel identical from the inside and mean the opposite.
- The site is slow. On Shopify, this is almost always the theme and the app scripts. It is a fixable problem with a known method, and replatforming to escape it means rebuilding the same storefront somewhere with fewer guardrails.
- Development feels slow. Usually a process, capacity or access problem rather than a platform ceiling. Ask how much of the delay is waiting for decisions rather than waiting for code.
- We do not own our data. You do — it is exportable. What you do not own is the runtime. That is a legitimate strategic concern; it is not the operational emergency it is often presented as.
- A new CTO prefers something else. Sometimes right, always worth interrogating. The question is which of the seven signals above they can point at.
If your honest answer is none of the seven, but several of the four, a replatform will cost you a year and hand you the same problems with more infrastructure to run.
If Several Signals Are Real
Before committing to a full migration, work through the cheaper options in order:
- Push Shopify Functions and checkout extensibility further. The gap between what teams believe is possible and what currently is has widened significantly.
- Go headless on Shopify. Keep the managed backend, the app ecosystem and the checkout; replace the storefront. This addresses the channel and experience signals without taking on commerce-engine ownership, and it is consistently the most under-considered option.
- Move one thing. A B2B channel, a single market, or the product configurator, running alongside the existing store. You learn on something small and keep the option to stop.
- Then, if the commercial-logic and economic signals are still real, replace the engine.
That last step is where an open-source engine becomes relevant. If your team is JavaScript-native and the driving signals are commercial-logic limits and platform economics, MedusaJS is a serious candidate — Node.js and TypeScript throughout, extensible commerce modules, no per-order fee, strong B2B primitives. We set out the wider architecture decision in our headless commerce framework.
And if the signals are not real, staying is the right answer. Our Shopify team spends most of its time making Plus stores work harder, and for most retail brands that is where the return is. Our MedusaJS practice exists for the ones who have genuinely outgrown it.
Frequently asked questions
What is the revenue point where brands outgrow Shopify Plus?
There is not one. We have seen £50m brands perfectly served by Plus and £8m brands genuinely constrained by it. Complexity, not turnover, is what determines fit — specifically the complexity of your pricing logic, catalogue model and channel mix.
Is Shopify Plus expensive at scale?
The platform fee plus app subscriptions plus any payment-processing premium becomes a meaningful annual figure at high volume. Whether that is expensive depends on the alternative, and most comparisons understate what owning a platform costs to run. Model both totals over three years before concluding.
Can we do B2B properly on Shopify Plus?
For conventional B2B — company accounts, catalogue-specific pricing, payment terms — yes, and it has improved substantially. Quote-to-order workflows, contract pricing with approval chains and complex entitlement rules are where teams still hit edges and end up maintaining logic outside the platform.
Should we go headless on Shopify or replatform entirely?
Headless on Shopify first, in almost every case. It addresses the front-end and channel constraints while keeping the managed backend, the checkout and the app ecosystem. Replacing the commerce engine is only justified when the constraints are in the commercial logic or the economics, not the storefront.
How long does a migration off Shopify Plus take?
Six to twelve months for a mid-market brand doing it properly, and the commerce engine is rarely the long pole — integrations, data migration, subscription contracts and retraining the operational teams are. Doing it in stages, one market or channel at a time, materially reduces the risk.
Get an Honest Read
Tell us which of the seven you recognise and we will tell you whether they are structural or fixable. Most are fixable. Talk to our team; we reply within a business day.