Ecommerce & Retail · Bristol

Holding organic traffic through a Shopify replatform

Most replatform traffic losses are avoidable and happen in the redirect map. This one was planned six weeks before launch rather than the week after.

Format template — not a real client

This page shows the structure a Nexclick case study takes. A homeware retailer is a description of a type of business, not an actual client, and every figure below reads ILLUSTRATIVE rather than a number. The page is noindex, emits no Article schema, and is excluded from the sitemap and the home page.

The situation

What we walked into

The retailer was moving from an ageing WooCommerce build to Shopify. The new URL structure was different, roughly 400 product URLs were changing, and several category pages that drove most of the organic revenue were being consolidated.

Client: A homeware retailer
Sector: Ecommerce & Retail
Started: 2025-04

What we did

In order, with dates

01
Six weeks pre-launch
Crawled and mapped everything before a line was changed
Full inventory of live URLs, ranked by organic sessions and revenue, so the pages that mattered were protected first.
02
Four weeks pre-launch
Built and tested the redirect map on staging
One-to-one redirects wherever a page had a successor. No blanket redirects to the homepage — that is how equity gets thrown away.
03
Two weeks pre-launch
Rebuilt category structure and internal linking
Consolidated thin categories deliberately rather than accidentally, with the merged pages absorbing the content that justified their rankings.
04
Launch to week 2
Monitored daily for the first fortnight
Crawl errors, index coverage and rankings checked every day post-launch so anything broken was caught in hours, not weeks.

The numbers

What changed

Figure pending

Organic sessions retained

Month 1 post-launch

Figure pending

Organic revenue

Months 1–3 post-launch

Figure pending

Indexed product pages

Week 4

Figures are withheld rather than estimated. A real case study states the measurement window alongside every number.

Honestly

What we would do differently

We agreed to a launch date in the week before a bank holiday. Nothing went wrong, but it meant the monitoring window overlapped with reduced cover on both sides. Worth an argument next time.