Platform selection is one of the highest-stakes technology decisions a retail or e-commerce business makes. Get it right and you have a foundation that supports growth for years. Get it wrong and you’re looking at a painful, expensive re-platform within 24 to 36 months — or worse, years of working around a platform that doesn’t fit your business.
Having led or advised on platform selections for businesses ranging from $10M to $2B+ in revenue, I’ve seen the decision made well and made badly. The patterns are surprisingly consistent.
Why platform selection goes wrong
The most common failure mode isn’t choosing the wrong platform — it’s choosing a platform for the wrong reasons. This usually happens in a few predictable ways.
Vendor-led evaluation. When the platform vendor controls the evaluation process — running the demos, defining the success criteria, selecting the reference customers — the outcome is predetermined. Every platform looks good in its own demo environment. The question is whether it works for your specific business model, data structures, and operational constraints.
Feature-list comparison. Evaluating platforms by ticking boxes on a feature matrix is superficial. Two platforms might both claim to support multi-brand, multi-currency, or marketplace capabilities, but the depth and maturity of those capabilities can vary enormously. Features are a starting point, not a decision framework.
Ignoring total cost of ownership. Licence fees are typically 20–30% of the total cost of a platform over five years. The rest is implementation, integration, customisation, ongoing maintenance, and the operational cost of running the platform. Businesses that select a platform based primarily on licence cost often end up spending more in the long run.
Underweighting organisational readiness. A platform is only as good as the team operating it. Selecting a highly flexible, composable architecture when your team has only worked on monolithic platforms — without a plan for building that capability — is a recipe for a difficult implementation.
A better framework
Over the years, I’ve developed a platform selection framework that consistently leads to better decisions. It’s built around five dimensions.
Business model fit is the non-negotiable starting point. The platform must natively support your business model — not through workarounds or heavy customisation. If you’re a multi-brand retailer, the platform needs genuine multi-brand architecture, not a multi-site bolt-on. If you’re a marketplace, the platform needs native marketplace capabilities, not a third-party integration.
Integration architecture matters more than features. Modern e-commerce is rarely a single platform — it’s a composition of services. The platform’s ability to integrate cleanly with your OMS, ERP, PIM, CDP, and marketing tools is often more important than its native feature set. Look for well-documented APIs, event-driven integration patterns, and a genuine ecosystem of integration partners.
Operational complexity is the dimension most businesses underweight. How much ongoing effort does the platform require? How easy is it to make content changes, update product data, configure promotions, and manage the day-to-day operation? The platform your marketing and merchandising teams interact with every day needs to be one they can use effectively.
Scalability and performance should be evaluated against your actual growth trajectory, not hypothetical scenarios. If you’re processing 10,000 orders per day and growing at 30% annually, you need to know the platform handles 30,000 orders per day comfortably. Peak trading performance is particularly critical — the platform needs to handle 5–10x normal volume without degradation.
Total cost of ownership over five years, including implementation, integration, customisation, licensing, hosting, maintenance, and the team required to operate it. This analysis often changes the ranking significantly compared to a licence-cost-only comparison.
The evaluation process
Run the evaluation yourself or with independent advisory support — not through the vendor’s sales process. This means writing your own requirements, defining your own evaluation criteria, and conducting your own reference checks with businesses similar to yours.
Include a proof of concept phase for your top two candidates. Not a vendor-run demo — a real POC that tests the platform against your specific business scenarios. This investment in evaluation quality pays for itself many times over by reducing implementation risk.
And involve the people who will actually operate the platform in the evaluation. Merchandisers, marketers, and customer service teams all interact with the platform daily. Their perspective on usability and operational fit is as important as engineering’s perspective on technical architecture.
The platforms I see working well
I’m deliberately not recommending specific platforms here because the right answer depends entirely on your business. But I will say that the platforms I see consistently working well in the Australian market share some common characteristics: genuine API-first architecture, mature multi-brand or multi-market capabilities, an active ecosystem of integration partners, and a realistic understanding of what they’re good at and what they’re not.
The most important thing is to make the decision based on your specific business reality — not on market positioning, analyst reports, or the platform your competitor chose.