Insights · 9 min read
A realistic breakdown of e-commerce platform costs: build vs buy, the cost drivers that move the number, typical budget ranges, and where projects overspend.
By GGP Editorial
If you're planning an online store, one of your first questions is what it costs to build. The short answer is that a custom e-commerce platform typically lands between $30,000 and $250,000, and enterprise builds go well past that. The longer answer, which matters more, is about where the money goes and which of your decisions move the number.
This guide breaks down the cost drivers, the three main ways to build, realistic budget ranges, and the costs people forget until after launch. By the end you should be able to produce a rough estimate of your own and have a sharper conversation with any development company you talk to.
Retailers often assume the gap between two quotes is markup. Sometimes it is. More often, the two companies are quoting two different projects. One quotes a storefront with a standard catalog, a checkout, and a few integrations. The other quotes a platform: multi-market pricing, a custom admin, ERP and warehouse sync, subscription billing, and a loyalty layer.
When you compare quotes, check scope before you check price. I wrote about how to read these quotes in more depth in Why custom software quotes vary by 3x.
A store selling 50 physical SKUs is a different build from one selling configurable products with variants, bundles, digital downloads, or subscriptions. Every product type adds data modeling, pricing logic, and checkout rules.
A template gets you to launch fast and cheap. A custom storefront that matches a specific brand, with its own interaction patterns, takes real design and front-end hours. This is one of the biggest single cost levers.
A basic card checkout through Stripe or PayPal is cheap to integrate. What adds cost is everything around it: multiple payment methods, saved cards, split payments, buy-now-pay-later, multi-currency settlement, and regional methods like Pix in Brazil. Each gateway has its own fees and its own integration effort. See Online payment gateway integration for what this involves.
In a serious build, most of the budget goes to connecting systems rather than building pages. Common ones are ERP, CRM, warehouse management, shipping carriers such as UPS and DHL, accounting tools like QuickBooks and Xero, and tax engines. The fewer clean APIs these systems have, the more custom glue you write.
The storefront is what customers see. The admin is what your team lives in: order management, refunds, inventory, promotions, customer service tools, and reporting. An underbuilt admin means your team ends up running the business on spreadsheets.
Selling in one country with one currency and one language is simple. Adding markets means currencies, exchange rates, translations, local tax rules, and often different payment methods per region. This is a step change in complexity, not a small add-on. If cross-border is part of the plan from day one, read this before scoping.
If you store card data yourself, you take on PCI DSS obligations. Most small and mid-market stores avoid that by using a tokenized gateway like Stripe or Adyen, which keeps card data off your servers and shrinks your compliance scope. Selling into the EU brings GDPR data handling requirements. These don't always change the build cost much, but they change how you build and add ongoing obligations.
These are three realistic routes. They are not all "custom development" in the same sense.
You pay a monthly subscription plus transaction fees and build on top with themes and apps. Launch is fast and the upfront cost is low, often a few thousand dollars for setup. The tradeoff is that you rent the platform: you pay more as you grow, you're limited by what the platform and its app ecosystem allow, and you don't own the code. For a store with standard needs this is often the right call. Custom e-commerce website vs Shopify covers when building your own makes more sense.
You keep a commerce engine for the hard parts (cart, checkout, tax, inventory) and build your own front-end on top. You get a custom storefront and customer experience without rebuilding payment and tax logic from scratch. The cost sits between SaaS and fully custom. This is a common choice for brands that have outgrown a theme but don't want to own payment infrastructure.
You own everything: catalog, cart, checkout, payment integration, admin, and reporting. The build cost is the highest of the three, and you own the ongoing maintenance too. This makes sense when your business model doesn't fit a standard engine, such as a B2B portal with negotiated pricing, a marketplace with commission logic, or a multi-market operation with unusual tax and settlement rules.
| Route | Upfront build cost | Time to launch | You own the code | Best fit |
|---|---|---|---|---|
| SaaS (Shopify) | Low | Days to weeks | No | Standard stores, quick start |
| Headless | Medium | 2-4 months | Partly | Brands outgrowing themes |
| Fully custom | High | 3-9+ months | Yes | Unique business models |
These are ranges we see in the market, not quotes. Your number depends on scope, region, and team. Treat them as a starting point.
| Project size | What's included | Typical range |
|---|---|---|
| MVP or simple store | Catalog, cart, checkout, one payment method, basic admin | $30,000-$60,000 |
| Mid-market platform | Custom design, several integrations, multi-currency or multi-language, reporting | $60,000-$150,000 |
| Enterprise or marketplace | ERP and WMS sync, multi-region tax and payments, custom admin, scale requirements | $150,000-$500,000+ |
An MVP is the cheapest way to test a store concept before committing to the full platform. See MVP development: what to build first and what it costs for that approach.
For a mid-market build, a typical team looks like this: a project manager, one or two back-end developers, one or two front-end developers, a designer, and a QA engineer. The roles overlap, and not everyone works full-time for the whole project.
The back end, which covers catalog, orders, payments, and integrations, takes the largest share of hours. Front-end and design come next, then QA and project management. If your project is heavy on integrations, the back end dominates even more.
Rates are the other half of the equation. An onshore US or UK team costs more per hour than an offshore one, which is why many companies now build with teams in China, Eastern Europe, or Latin America. GlobeSoft's team is based in China, which keeps rates competitive without dropping seniority. Software development cost in China explains what you actually pay there.
A simple store can launch in two to three months. A mid-market platform usually takes four to eight months. Enterprise and multi-market builds can run nine months to a year or more.
Time and cost move together for one reason: you're paying for people's time. A compressed timeline usually means a bigger team working in parallel, which costs more and adds coordination overhead. A relaxed timeline lets a smaller team build sequentially, which is cheaper but slower. How long custom software development takes covers the tradeoffs in detail.
The build price is not the total cost of ownership. These are the items that surprise people after launch.
The single most effective cost control is scope discipline. Most budget overruns come from adding features mid-build, not from initial misquoting. A few rules that help:
A simple store built from scratch usually runs $30,000 to $60,000. Built on Shopify with a theme, setup can be a few thousand dollars or less, plus the monthly subscription.
Shopify is cheaper to launch. Custom is more expensive upfront but removes platform fees and limits as you scale. The better question is which one fits your business model over the next three years.
A simple store takes two to three months. A mid-market platform with integrations usually takes four to eight months. Enterprise builds run longer.
A tokenized gateway keeps card data off your servers, which reduces your PCI scope to a self-assessment. You still have obligations, but they're far smaller than if you stored card data yourself. Confirm the details with your gateway and a qualified assessor, because this depends on how you integrate.
The big drivers are custom design, integrations with ERP, CRM, and warehouse systems, multi-market features, and unusual product or pricing models. These add far more than the storefront itself.
Yes. GlobeSoft builds custom e-commerce platforms, including multi-market and cross-border stores, with integrations to payments, ERP, and logistics. We work with clients across Brazil, South Africa, Singapore, and the US.
Tell us what you are building and where you are today. We typically reply within 24 hours.
