Enterprise commerce transformation

The challenge

The platform connected product data, pricing, stock, payments, fulfilment, customer service and third-party marketplaces. Replacing everything at once would have created unacceptable commercial risk, while leaving it untouched would have limited growth.

Pius modernised the platform in stages while preserving the business knowledge built into years of production behaviour.

A modernisation programme, not a rewrite

The work combined architecture, performance engineering, data migration, payment-risk controls, marketplace automation and operational tooling. We introduced changes in stages so the business could benefit early and keep trading throughout.

We prioritised revenue-critical paths, security and observability. Lower-risk structural improvements followed once the platform could be measured and trusted.

Engineering for commercial reality

The platform connected to Microsoft Dynamics NAV, multiple marketplaces, search services, payment gateways, fulfilment processes and internal tools. Pius treated those integrations as products in their own right: documented contracts, observable failures, deliberate retry behaviour and clear ownership.

That discipline turned an opaque legacy estate into a system teams could operate, improve and reason about.

The outcome

Approved results are presented anonymously because the engagement includes commercially sensitive enterprise work. They show revenue growth, lower fraud exposure and production reliability achieved without destabilising live operations.

Security, fraud and operational resilience

Security was treated as part of the platform architecture rather than as a final compliance exercise. The estate handled customer information, high-value payment activity and privileged connections to ERP, marketplace, fulfilment and payment systems.

Pius introduced and progressively refined payment-risk controls, strengthened the boundaries around sensitive workflows and third-party integrations, and improved the telemetry needed to identify suspicious behaviour and operational failure. Controls were designed around the commercial reality of the platform: reducing fraud and chargebacks without unnecessarily obstructing legitimate customers or destabilising live trading.

The programme reduced chargebacks by 65%, while the wider modernisation maintained 99.98% platform uptime and supported £12 million in additional checkout revenue.

Security work included, where applicable:

• AWS WAF rules and rate limiting;
• restricted IAM roles and environment separation;
• secrets moved out of source code;
• dependency and static-analysis checks in CI;
• privileged-action auditing;
• antifraud velocity and behavioural rules;
• payment-tokenisation and reduced card-data exposure;
• rehearsed backup restoration;
• incident alerting and documented response ownership.

Security improvements were introduced incrementally, prioritising exploitable and commercially material risks while preserving the platform's established operational behaviour.

AI-assisted analysis was used to accelerate the mapping of inherited code paths, identify areas requiring deeper manual review, generate adversarial test cases and investigate complex production behaviour. Findings remained subject to human review and reproducible technical verification.