The situation

A UK fashion brand doing £2m annually on WooCommerce had been accumulating technical debt for eight years. Their site loaded in 7+ seconds on mobile. Their checkout was a plugin stack that broke every other WooCommerce update. They had 40,000 products, 8 years of order history, and a peak season twelve weeks away.

Their developers told them it would take six months. They came to us.

The problem, properly diagnosed

The conversation that matters at the start of a migration is not "what does the new site need to do?" It's "what is the thing most likely to go wrong that will end this project?"

For this brand, the answer was the product data. Their WooCommerce database had been modified by a succession of plugins, and the variant data — the size/colour combinations across 40,000 SKUs — was in a non-standard schema. Getting that data out and into a shape Shopify would accept cleanly was the real engineering problem. Everything else was execution.

We spent the first week on data archaeology: exporting, mapping, identifying the edge cases (products with more than three options, products with shared variant images, products with custom fields that had no native Shopify equivalent). We built a migration script against a staging environment, ran it, identified the failures, iterated.

The approach

Data migration first. We don't touch the design until the data is proven. A beautiful storefront with broken product data is worse than an ugly one that works.

The migration script handled:

• 40,000+ products with full variant data
• 8 years of order history (critical for their customer accounts)
• Customer accounts and loyalty point balances
• Redirects from every significant WooCommerce URL pattern to its Shopify equivalent

Theme: built from scratch. We didn't adapt their WooCommerce theme. Their old design had accumulated too many compromises. We built a clean Shopify 2.0 theme based on their brand guidelines, performance-first, with a mobile experience that actually worked.

Target LCP: under 2 seconds on mobile. We hit 1.8s at launch.

Cutover plan. We rehearsed the cutover twice in staging before doing it live. DNS propagation, Shopify payments activation, old WooCommerce store set to maintenance mode, 301 redirects confirmed. The live cutover took 47 minutes.

The outcome

Conversion rate lifted from 1.4% to 1.9% in the first month post-launch. By month three, it was at 2.1% — a 50% relative improvement on where they started.

The 38% figure in the headline is the conversion rate improvement at the six-week post-launch mark, before seasonal effects started to complicate the comparison.

Page speed went from 7.2s LCP to 1.8s LCP on mobile. Their Google Shopping ROAS improved alongside it.

They hit their peak season on the new platform without incident.

What we'd do differently

The size guide migration was the one thing we underestimated. Their old WooCommerce size guides were stored as custom post types with non-standard relationships to products. Getting those into Shopify metaobjects took two extra days we hadn't scoped. We absorbed it, but it's now the first thing we ask about in a scoping call.

Tech stack

Shopify Plus, Shopify 2.0 theme (custom), Klaviyo (retained from WooCommerce), Recharge (subscriptions, new), custom migration scripts (Node.js).