The question every founder asks
Sooner or later, every growing online brand asks the same question: should we stay on Shopify, or build something custom?
The internet isn’t helpful here. Shopify’s marketing says you never need to leave. Dev agencies’ marketing says you should have left yesterday. Neither is right. The honest answer is it depends on a small number of specific factors — and once you know what they are, the decision usually makes itself.
Here’s how we walk Malaysian clients through it.
What Shopify does brilliantly
Before we argue about leaving, let’s be fair about what Shopify is actually great at:
Speed to launch. You can have a real, working store live in a weekend.
Checkout conversion. Shopify Checkout is genuinely one of the best converting checkouts in the world. That’s not marketing spin — it’s the product of a decade of A/B testing at scale.
A huge app ecosystem. Loyalty, reviews, subscriptions, shipping rules, most of it is a few clicks away.
No server ops. You don’t wake up at 2am because a server crashed during a Raya campaign.
Affordable at the small end. For businesses doing under RM 500k/year in GMV, custom is almost always the wrong answer.
If you’re reading this and those points sound like a description of your current life, the answer is probably: stay on Shopify. Ignore the rest of this post. Come back in two years.
Where Shopify hits a wall
That said, growing businesses do hit real walls. In our experience, here are the common ones:
1. Unusual products. Rental periods, made-to-order lead times, bundles with complex rules, B2B tiered pricing, ticketed services — any time the product doesn’t fit into “item + variant + price”, you start fighting Shopify’s data model.
2. Custom checkout requirements. Installments that Shopify Payments doesn’t support, B2B purchase orders, custom KYC steps, split shipments with different fulfilment logic. Shopify Plus unlocks some of this, but a lot still isn’t flexible enough.
3. Deep integration with your back-office. If your ERP, WMS, POS, accounting and CRM all need to talk to each other and to your store in real time — and if “approximately synced every 10 minutes via Zapier” isn’t good enough — you eventually run out of duct tape.
4. App overload. You know the warning signs. You’re paying for 14 apps. Two of them contradict each other. Page load is 5+ seconds. You’re scared to uninstall any of them because you can’t remember which does what.
5. Brand and UX ambition. There’s a ceiling on how distinctive a Shopify theme can look before you’re basically fighting the platform to build around it.
None of these mean Shopify is bad. They just mean you’ve outgrown it.
What “custom” actually means (and doesn’t)
“Custom e-commerce” does *not* mean “write everything from scratch in PHP in 2026.” That’s not what any competent agency builds today.
What it usually means is one of two things:
A custom storefront on top of a headless commerce engine (Shopify Hydrogen, Medusa, Saleor, commercetools, or a custom Laravel/Node backend for highly unusual businesses). You keep a mature cart/checkout engine but own the presentation, the data model extensions, and the integrations.
A fully custom stack for businesses whose catalogue, checkout, or workflow is so unusual that off-the-shelf engines cost more to bend than to rebuild.
In both cases, you gain: full control over the data model, deep integration with your operations, an interface that’s yours (not a theme), and no monthly cap on what the platform “allows.”
You also gain: more responsibility. Custom means someone has to own uptime, security updates, payment certification, and so on. That someone should be your agency, not you — but it’s still a different relationship from renting Shopify.
The real decision matrix
Here’s the short version we use with clients. If you answer yes to three or more, it’s time to seriously consider custom:
1) Your products don’t fit into a “variant + price” shape.
2) Your checkout needs steps or payment rules Shopify doesn’t support natively.
3) Your ERP or POS needs real-time (not nightly) sync.
4) You’re paying for 10+ apps and the store is slow.
5) Your GMV is >RM 2M/year and app/platform fees are becoming material.
6) Brand differentiation through the site experience is a competitive moat for you.
7) You have a technology partner who can own the platform long-term (vs. managing it yourself).
If you answered yes to one or two, you probably just need a better Shopify setup — consolidate apps, move to a faster theme, add Shopify Plus if you’re near the GMV line, and come back to this question in 12 months.
Cost reality
Ballpark figures for Malaysian businesses, as of 2026:
1) A well-built custom-themed Shopify store: RM 15k–60k build + ongoing platform fees.
2) Shopify Plus with custom checkout extensions: RM 60k–150k build + ~RM 10k/month platform.
3) Headless Shopify (Hydrogen or similar) with a custom frontend + back-office integrations: RM 80k–250k build.
4) Full custom e-commerce with ERP/POS integration: RM 150k–500k+ build.
Custom is not “always more expensive” over a 3-year horizon — once app fees, platform fees, and the cost of working around platform limits are counted, the break-even is often tighter than founders expect. But custom is always a bigger upfront commitment and needs a real build partner.
The local context
For Malaysian brands, a few specifics worth keeping in mind:
1) Payment gateway flexibility (FPX, Billplz, iPay88, DuitNow) is easier to orchestrate cleanly on a custom stack if you need non-standard flows.
2) Local courier integrations (Ninja Van, J&T, Pos Laju) are well-covered by both Shopify apps and custom builds — not a deciding factor on its own.
3) E-invoicing / LHDN requirements are handled better when your store talks directly to your accounting system — a common custom-build trigger.
The honest answer
Most businesses we talk to should stay on Shopify longer than they think. Some should have moved 12 months ago. Knowing which one you are isn’t about taste — it’s about looking at those seven matrix questions and counting.
Next step: Not sure which side you’re on? We’ll do a 45-minute review of your current stack and GMV and give you a direct answer — stay, restructure, or rebuild.
