A new theme can fix a store that looks outdated. It cannot fix a store that is structurally incapable of doing what your business now needs. Confusing the two is one of the most expensive mistakes an online retailer can make — and it goes both ways. Some businesses spend six figures replatforming when a focused redesign would have solved the actual problem in weeks. Others spend years patching a redesign on top of a platform that was never going to support their next stage of growth.

The difference comes down to one question: is the problem cosmetic, or is it architectural? A redesign changes how your store looks and feels on the platform you already have. A replatform changes the platform itself — the underlying system that decides what your store is structurally capable of, regardless of how much design or development time you throw at it.

Below are seven signs that point specifically toward the second category. If two or three of these sound familiar, it’s worth a real conversation before you spend another budget cycle redesigning around a problem a redesign can’t fix.

Sign 1: Your checkout is structurally locked

Some platforms let you customize checkout freely. Others lock it down entirely — no added steps, no removed fields, no changes to how payment options display, no matter what a developer tries. If your business model genuinely requires checkout logic your platform doesn’t allow (a specific B2B approval flow, a regulatory disclosure, a non-standard payment sequence), that’s not a design limitation. That’s an architectural one.

The test here matters: if a developer actually can build what you need with custom code, that’s a development conversation, not a migration one. The signal to watch for is when the answer comes back as “the platform doesn’t allow this,” not “this would take more time than expected.”

Sign 2: Your site struggles or goes down under real traffic, even after hosting upgrades

Every platform has a performance ceiling. If your store consistently slows down or fails during high-traffic moments — a sale, a press mention, a seasonal spike — and you’ve already tried the obvious fixes (better hosting, a CDN, image optimization, app cleanup) without resolving it, you’re not looking at a configuration problem anymore. You’re looking at a platform that wasn’t built for the transaction volume your business now needs to handle.

Why this one is worth taking seriously

Core Web Vitals data from 2026 shows a direct, measurable line between speed and revenue: every 100 milliseconds of load-time improvement correlates with roughly 1% more conversions, and for a $10 million-a-year store, a 500ms improvement can recover an estimated $500,000 in annual revenue. If your platform has a hard ceiling on how fast it can possibly get, no amount of optimization budget closes that gap.

Sign 3: Your core business systems can’t talk to each other without manual work

If your ERP, CRM, or inventory system requires someone on your team to manually re-enter data because your platform can’t integrate with it cleanly, that’s not a productivity inconvenience — it’s a growth ceiling with a staffing cost attached. As order volume increases, manual reconciliation doesn’t scale linearly; it gets worse faster than the business grows. This is one of the clearest, least debatable replatform signals, because the workaround cost is visible on a spreadsheet, not just in customer experience.

Sign 4: Every new feature requires a workaround instead of a setting

There’s a specific moment worth paying attention to: you ask for a feature — a loyalty program, a new payment method, a subscription option — and the answer comes back as “the platform doesn’t support that natively, but we can build a workaround.” One workaround is normal. A pattern of them, stacking on top of each other release after release, is a sign that your platform’s roadmap has fundamentally diverged from your business’s roadmap.

This compounds in a way that’s easy to underestimate. Each workaround becomes its own piece of fragile, custom logic that the next developer has to understand before they can safely change anything. Eventually, simple requests take disproportionately long not because the request is complex, but because the system underneath it has become a stack of patches.

Sign 5: Maintenance and licensing costs are rising faster than revenue

Track this one in actual numbers, not impressions. If your platform fees, plugin costs, and the developer hours needed just to keep the lights on are growing faster than your revenue line, the platform has quietly become a drag on margin rather than infrastructure that supports growth. This is especially common on legacy or heavily customized setups, where the original build made sense at a smaller scale but the carrying cost compounds as the business grows around it.

Sign 6: You’re expanding into new markets, channels, or business models the platform can’t support

Adding multi-currency pricing, regional tax handling, a wholesale or B2B portal, or a true omnichannel presence across marketplaces and social commerce exposes platform limitations that a single-market, single-channel store never had to confront. If localization, B2B logic, or multi-storefront management requires bolting on a patchwork of third-party tools instead of using something the platform was actually designed to handle, that’s a sign the platform’s ceiling is lower than your roadmap.

Sign 7: A past migration or build cut corners, and you’re still paying for it

This sign is less about the platform itself and more about what’s underneath it. A meaningful share of replatforming projects fail or run over budget — industry research puts the figure at over 80% — and the most common cause is what’s known as the lift-and-shift trap: old workarounds, broken logic, and platform-specific patches get carried over unchanged into the new system instead of being fixed. If your business is still running on operational duct tape from a rushed build or a previous migration that didn’t address root causes, the problem isn’t going to disappear with another redesign. It needs to be diagnosed and rebuilt properly, not relocated.

This is also the sign most likely to involve genuine data and SEO risk, since the original migration may have already introduced broken redirects, lost product data, or fragile integrations. A proper platform migration addresses those root causes directly, rather than carrying them forward into yet another rebuild.

What does NOT require a replatform

It’s worth being equally direct about the other side, since over-correcting is its own expensive mistake. A redesign — not a migration — is almost always the right call when the issue is:

• A dated visual design or an inconsistent brand feel
• A high cart abandonment rate around the 70% industry average, which is almost always a checkout UX, trust signal, or shipping cost presentation issue, not a platform issue
• Confusing navigation or a cluttered layout
• A mobile experience that feels clunky but is technically functional
• A single slow integration or an isolated UX problem that a focused fix can resolve

Replatforming to fix a conversion or UX problem usually doesn’t fix it — the same weak checkout flow or confusing navigation tends to get rebuilt, just on new infrastructure, while the migration itself introduces fresh risk: broken redirects, SEO traffic loss, and months of disruption for a problem a redesign would have solved in weeks at a fraction of the cost.

How to actually decide

If you recognized two or three of the seven signs above, the right next step isn’t a redesign proposal and it isn’t a migration proposal either — it’s a diagnostic. Before any code gets written, the questions worth answering are: is this problem cosmetic or architectural? Can a developer actually build the fix on the current platform, or is the platform itself refusing to allow it? And if it is a genuine replatform, what from the current setup is a workaround that needs to be redesigned, not lifted over as-is?

That diagnostic is exactly what we run with clients before recommending anything — a clear read on whether your current platform is a cosmetic problem, an architectural one, or some of both, and a concrete plan for whichever one it turns out to be.

Get in touch for a free consultation, or explore our platform migration services to see how we approach platform decisions like this one.