How to move a live store off a platform you don't own
Replatforming an online store is mostly a nerve problem, not a technical one. The trick is that the switch itself should be the least eventful part.
Businesses stay on platforms they've outgrown for one reason: the migration looks terrifying. The catalogue is live, customers are ordering today, and the platform holds the storefront, the checkout and the admin all at once. Moving feels like changing the engine while driving.
It isn't, if you sequence it properly. The cut-over should be the smallest step in the project, not the biggest.
What we did for Forget Me Not
Forget Me Not is a New Zealand souvenir and gift store at Cathedral Junction in Christchurch, with a catalogue running into the hundreds of products. They were on a legacy platform. We rebuilt the catalogue, the checkout and the store admin, and cut it over live.
The work that made the switch uneventful all happened before it. The new store was built and populated while the old one kept trading. Product data was migrated and checked against the live catalogue rather than eyeballed. The admin — the part staff actually touch every day — was working before anything pointed at it.
The order that keeps you trading
- Get the product data out first, and treat it as the real asset. It's the thing that's genuinely expensive to recreate.
- Build the new store to the point where you could take an order on it, while the old one is still taking them.
- Have the people who run the shop use the new admin before launch, not after. If they can't add a product without help, you aren't ready.
- Keep the URLs. Search engines and returning customers both have opinions about this.
- Switch when it's quiet, and have the old platform still standing behind you for a while.
What you're really buying
The reason to do any of this is ownership. On a rented platform, the storefront, the customer data and the checkout are all held by someone whose pricing and roadmap you don't control. Every year you stay, the migration gets a little bigger.
Doing it once, properly, and ending up with something in your own name is the cheapest version of this decision you will ever get. It does not get easier by waiting.
The migration you're avoiding is smaller today than it will ever be again.