Shopify to WooCommerce Migration: $1,840/mo to $86/mo, 1,840 SKUs, Zero Lost SEO

Northdrift sold technical outdoor gear (climbing hardware, ski-touring kit, camping electronics). 1,840 SKUs, ~$3.2M/year in revenue, growing 30% YoY. They were on Shopify Plus paying $1,840/month base, plus another $400/mo on apps (subscriptions, B2B portal, advanced shipping, returns). About $27k/year in recurring platform cost for a store that needed five custom features and got them via five different apps.

Three specific pain points pushed the migration:

• Product-page rigidity. Editorial wanted long-form gear stories — a 1,200-word write-up on each climbing harness with sizing-fit videos, expert quotes, comparison tables. Shopify product templates kept fighting them. Liquid edits required either a freelancer or sacrificed page-builder flexibility.
• App lock-in. Their B2B portal app raised prices from $39 to $129/month. Migrating off it inside Shopify meant another app with a different data shape and a manual customer-data re-import.
• Subscription product checkout cost. Recharge took 1% on top of every recurring order. On $400k/year subscription revenue, that was $4k/year for functionality they’d rather own.

The team had heard “Shopify to WooCommerce kills SEO”. They asked for a hard floor: no more than 5% organic traffic drop in the first 30 days post-cutover, or roll back.

WooCommerce + Bricks Builder + a tight set of plugins we trust on production: WooCommerce Subscriptions (official), WPML for the German edition, WooCommerce B2B by WebKul, FluentSMTP for transactional mail, Wordfence + 2FA for login security.

Hosting on Cloudways with a Vultr High-Frequency 4GB instance ($55/month), Bunny CDN ($5/month base + traffic), and Cloudflare in front for DNS, edge caching, and image polish. Total recurring infrastructure: $86/month.

The migration ran in five phases over 5 weeks:

• Week 1 — Export and shape. Pulled Shopify products, customers, orders, and metafields via the Admin API. Wrote a Python ETL that mapped Shopify product variants into WooCommerce variations and preserved metafield content (gear-spec tables, fit notes) into ACF fields on each product.
• Week 2 — Schema and templates. Built the WooCommerce product template in Bricks with the long-form editorial layout the team wanted. Built 4 page templates (collection, category, blog, landing). Tied ACF gear-spec fields into Product schema with proper offers and aggregateRating.
• Week 3 — URL preservation. Every Shopify URL got a one-to-one 301 in a custom redirect plugin. /products/{handle} → /product/{slug}, /collections/{handle} → /product-category/{slug}, /blogs/news/{handle} → /blog/{slug}. 2,400 redirects, all unit-tested.
• Week 4 — Payments and subscriptions. Stripe direct (no Recharge middleman), Klarna, PayPal. Subscription customers re-tokenized via Stripe’s setup_intent flow — we ran an email campaign 7 days before cutover so existing subscribers re-authorized cards once.
• Week 5 — Staging burn-in and cutover. Full staging environment, 500-order load test, support-team training, then a 3 AM Sunday DNS flip with the old Shopify store kept warm for 72 hours in case rollback was needed.

Day 30 organic sessions: down 1.8% versus the same 30-day window pre-migration — well inside the 5% rollback threshold. Day 60: +11% versus pre-migration as new editorial content started ranking. The 301-redirect map preserved link equity; no manual reindexing dance.

Recurring platform cost: $2,240/mo to $86/mo — a 95% drop, $25,800/year saved on hosting and platform alone. Recharge fees gone. App subscriptions gone.

Editorial team published 14 long-form product stories in the first two months (versus 2 in the prior quarter on Shopify). The climbing harness category page now reads like a gear review, not a grid of identical thumbnails. Time on category page: 52 seconds to 2 minutes 14 seconds.

One real downside: WooCommerce admin is slower than Shopify admin. Order fulfillment per order is ~4 seconds slower than Shopify’s UI. For a store doing 60 orders/day that’s 4 minutes of operator time. The team accepts it.