Stay by default, move deliberately
Replatforming an established store is expensive, risky, and often driven by the wrong signal. A store underperforming because of a bad checkout flow, slow product images or weak merchandising will underperform equally on whatever you migrate it to. The platform is rarely the thing holding the business back, until quite suddenly it is.
This is our working framework for knowing the difference. We use it with clients considering a move from Shopify or WooCommerce to something more custom, and with clients who are already mid-flight and having second thoughts.
When Shopify is still the right answer
Shopify is the right answer for a remarkable range of businesses. A small or mid-sized DTC brand with 50-10,000 SKUs, a straightforward B2C flow, one or two currencies, and a willingness to work within the Shopify ecosystem gets serious leverage. The platform handles hosting, payments, PCI scope, fraud screening, tax, and a credible set of integrations out of the box. The real TCO of getting equivalent capability on a custom stack is three to five times higher in the first two years.
Do not leave Shopify because you dislike the theme editor, cannot find a developer for your obscure app, or want a bespoke product page. Those are content and design problems, not platform problems.
When Shopify starts to crack
Checkout customisation
The big one. Shopify has steadily opened up checkout via Checkout Extensibility and Shopify Functions, but the ceiling is real. Complex conditional logic across shipping, tax, discounts and split-tender payments still hits limits that Plus merchants eventually run into. If your business model requires the checkout to behave in ways the platform will not permit, you are on borrowed time.
B2B complexity
Shopify has a credible B2B offering via Plus, but true B2B estates (negotiated pricing per customer, multi-tier approvals, quote-to-order flows, net terms, EDI integration, customer-specific catalogues) push the platform hard. You can force it. It will not be comfortable.
Multi-currency and multi-region nuance
Markets has improved dramatically. For a brand selling in four or five currencies with locally-hosted payment methods, Shopify is fine. For a brand running genuinely distinct regional businesses with different catalogues, tax regimes (think AU GST versus EU VAT versus US sales tax), fulfilment networks and promotional calendars, the seams show.
Shopify Plus pricing
The move from standard Shopify to Plus is not linear. Plus starts around USD 2,500 per month and scales with GMV. If you are being pushed to Plus purely to unlock one feature (custom checkout, wholesale, headless) do the maths on whether a custom build amortises better over three years. Sometimes it does. Sometimes it does not. Do the actual spreadsheet.
When WooCommerce starts to crack
WooCommerce is the other side of the same coin. Maximum flexibility, minimum operational support, and a cost profile dominated by the hosting and dev time you do not see until you hit scale.
Performance at 50k+ SKUs
A clean Woo install with a small catalogue performs well. Around 50,000 SKUs with variants, attributes and meta, the wp_posts table becomes a liability. Queries that were fine at launch grind. CDN and object caching help. They do not make the underlying architecture horizontal.
Plugin fragility
Woo ecosystems accumulate plugins. Each update cycle, something breaks something else. For a small store this is annoying. For a store doing seven figures, a single plugin-induced checkout outage during a Black Friday or Click Frenzy window can cost more than a replatform.
Hosting and operational burden
You own the stack. Uptime, backups, security patches, PHP version upgrades, WAF rules, DDoS mitigation. For a business without in-house engineering, that burden falls on whichever agency is cheapest, which is rarely a good outcome at scale.
The alternative stacks worth knowing
Replatforming is never "anywhere but here". It is always a specific next platform with a specific fit. A short, opinionated tour.
Shopify Hydrogen
Headless Shopify with a React frontend on Oxygen (or your own hosting). Keeps Shopify checkout and commerce engine, replaces the storefront. Right when you want full design control and better performance without leaving the Shopify data model.
Medusa
Open-source, Node-based, modular commerce. Strong fit for custom product experiences where the team has Node capability in-house or via an agency. Self-hosted or via Medusa Cloud.
Commerce Layer
Headless commerce-as-API. Best suited to enterprise-grade multi-market, multi-currency estates where commerce sits inside a broader composable stack. Not the right shape for a typical Shopify refugee.
Saleor
Open-source, Python-based, GraphQL-native. Strong for teams already comfortable in Python and Django, or for complex catalogue and pricing models.
BigCommerce
The nearest head-to-head with Shopify Plus. Stronger at B2B out of the box, more generous with API limits, less rich app ecosystem. Worth quoting against Plus on any serious enterprise RFP.
Custom with Stripe
The nuclear option. A bespoke commerce application built on a modern framework with Stripe for payments, tax and subscriptions. Only the right answer for businesses whose commerce flow genuinely does not fit an off-the-shelf platform: subscription boxes with unusual mechanics, marketplace models, highly configurable products, B2B estates with bespoke workflows.
The decision framework
When a client asks whether to replatform, we run four numbers.
- Lost revenue. What is the current platform costing you in blocked features, conversion friction or outages? Be specific and quantified. "We think it is hurting us" is not a number.
- Migration cost. A realistic three-year TCO of the new platform in AUD, including build, data migration, integrations, retraining and an honest contingency.
- Time to parity. How long until the new stack does everything the old one does? Nine months is normal. Twelve is common. Anything quoted under six for a serious migration is optimistic.
- Risk of staying. What happens over the next 24 months if you do not move? Slow compounding pain is the real cost.
If lost revenue plus risk exceeds migration cost within 24-30 months, you have a case. If not, invest in the current platform instead.
The honest conclusion
Most businesses that think they need to replatform would be better served by a 90-day investment in their current stack: checkout optimisation, performance, merchandising, analytics. A minority genuinely have outgrown it. Telling those apart is most of the value of an honest conversation with a partner. If you want that conversation, we are up for it.
