Every growing retailer eventually hits the same wall: the platform that got you to your first million in GMV starts fighting you past your fifth. The question isn't "which platform is best" — it's which architecture matches your GMV, your roadmap, and how much control you actually need over checkout, content, and integrations. This guide gives you a structured way to answer that, plus the specific questions to ask before you commit budget.
Who this is for
Ecommerce leaders and IT decision-makers evaluating a replatform or rebuild — typically triggered by a growth plateau, a checkout limitation, slow page speed, or an ERP/CRM integration that's outgrown the platform's native capabilities.
Step 1: Know where you actually stand
- [ ] Annual GMV — under $1M, $1M–$5M, $5M–$10M, or $10M+? This single number eliminates more options than any feature comparison.
- [ ] Checkout customization needs — do you need subscription logic, complex tax rules, B2B pricing tiers, or multi-entity checkout that your current platform's checkout API can't handle?
- [ ] Content and merchandising velocity — how often does marketing need to ship new landing pages, campaigns, or storefront experiments without an engineering ticket?
- [ ] Existing integration debt — how many systems (ERP, CRM, PIM, loyalty, tax engine) currently talk to your storefront, and how brittle are those connections?
- [ ] Internal engineering capacity — do you have (or are you willing to hire) a team that can own a custom frontend and its uptime?
- [ ] International/multi-brand complexity — multiple storefronts, currencies, or regional catalogs from one backend?
Step 2: Match your profile to an architecture
Choose Shopify Plus (or an equivalent managed platform) if:
- GMV is roughly $1M–$5M and growing but not yet at a scale where platform fees outweigh a custom build's cost.
- You need moderate checkout customization (Shopify's Checkout Extensibility API covers most of it) without owning the underlying commerce engine.
- You want dedicated merchant support and a faster time-to-market than a custom build allows.
- Your content/merchandising needs are served fine by themes and apps.
Choose headless commerce if:
- A themed storefront genuinely can't do the job — not "would be nicer," but a real blocker: independent channel experiences (app, kiosk, marketplace), frequent frontend experimentation, complex subscription logic, or multiple payment paths across regions.
- Page speed and Core Web Vitals are a competitive factor in your category (cached headless pages can serve in well under 100ms vs. 300–800ms for typical themed pages).
- You're expanding into multiple markets or brands from a shared commerce backend and need independent frontends per market.
- You have the engineering capacity to own a decoupled frontend (Next.js/Nuxt or similar) plus its hosting, monitoring, and release process.
Choose a fully custom build if:
- GMV exceeds roughly $10M and platform + transaction fees on a managed plan now exceed the annualized cost of ownership.
- Your business logic (pricing, inventory allocation, fulfillment routing, loyalty) is a genuine differentiator that off-the-shelf platforms force you to work around.
- You need 100% IP ownership and zero vendor lock-in as a strategic requirement, not a preference.
- You're prepared for a longer build timeline in exchange for exact workflow fit and no per-seat or GMV-based licensing ceiling.
Step 3: Questions to ask any vendor or internal team before you commit
- What is the fully loaded first-year cost — platform/license fees, build cost, hosting, and ongoing maintenance — not just the sticker price?
- What happens to page performance and SEO rankings during and immediately after migration?
- Who owns uptime and incident response after launch — the vendor, an agency, or your internal team?
- How does the proposed architecture handle your specific ERP, CRM, and tax-engine integrations today, not hypothetically?
- What's the realistic timeline to first production release, and what's the cost of slippage?
- If you choose headless, which parts of the "backend" (inventory, promotions, subscriptions) still live on the commerce platform vs. custom services?
- What's the exit path if this platform stops fitting in three years — how portable is your data and business logic?
Red flags to watch for
- [ ] A vendor who can't give you a real number for ongoing maintenance cost, only the build estimate.
- [ ] A headless proposal with no plan for who maintains the frontend after launch.
- [ ] A custom-build pitch that ignores your existing ERP/CRM integration complexity.
- [ ] Any estimate that doesn't account for SEO migration risk if you're replatforming an existing storefront.
How to use this
Score your business against Step 1 first — most of the decision falls out of GMV and checkout complexity alone. Use Step 3's questions verbatim in vendor conversations; a vendor who can't answer them clearly is a signal, not just a data point. If you're already leaning toward headless, pair this with our headless commerce vendor evaluation scorecard before you shortlist partners.