Shopify Migration Agency

WBD is a small e-commerce agency in Oxford, and a good share of our work is migration: onto Shopify from Magento, WooCommerce or BigCommerce, up from a standard plan to Plus, and occasionally — since we sell what suits the brand rather than what suits us — off Shopify altogether. The first useful thing we can tell you is where the risk isn't. It isn't the import. Data moves; the formats are known and the tools are mature. Most of what a migration agency sells is the import, and the import is the easy part.

The hard part is what a mature store has grown that never appears in an export file. The search equity. The internal links. The habits of the people who run it every day, and the reasons behind decisions nobody wrote down. This chapter is what we've learned carrying that across; the platform-by-platform detail hangs off it.

Migrations arrive from a handful of directions, and the direction changes the job more than the destination does.

Always Riding — the D2C cycling boutique we founded in 2008 and sold in 2018, a big Magento Community store running five languages and several currencies — ranked first on Google for "cycling jerseys" for a long stretch, which is to say a meaningful share of new customers each month arrived free. We moved platforms carefully, with redirects in place, and lost most of that position anyway. It came back eventually. The interim was educational in the way that only expensive things are, and the full account is in When (Not) to Leave Shopify. Then three years as consultant e-commerce directors for Above Category in Sausalito put us inside a working store's monthly bill. The checklist below is scar tissue, not theory.

Search rankings first, dropped through a missed redirect — the most expensive loss and the most preventable. Then the internal linking a store weaves over years: collections pointing at products, journal pages pointing at both, the whole woven-together-ness of a mature site, flattened into a menu and a grid, because an importer moves pages and not the threads between them. Customer accounts that look complete until someone actually checks them. Passwords never transfer between platforms, so returning customers meet a reset — which should be a communicated moment rather than a wall of confused support tickets. An app stack copied across one-for-one, old bill included, when half of it had earned retirement. And the details that were the brand, flattened into a stock theme. None of that shows up in an import log.

Which is why the procedure runs the other way round from how it's usually sold. Map the old store's structure before anything moves. Check migrated data against the source rather than trusting the import. Build the redirect map from every old URL to its new home and test it before launch, not after. Audit the apps with one plain question each — does this still need to exist? — because that audit alone often pays for a chunk of the project.

Here are the two things experience keeps underlining, which the brochure version of this page never mentions. First: even a migration done properly gets choppy seas for a while. Google doesn't simply follow your redirects and carry on — it re-reads everything. The code, the markup, the structure, the works. Organic traffic will dip while that happens and revenue with it, and anyone promising otherwise is selling you the brochure. Played well it comes back, and if the new store is genuinely better on what Google is actually measuring — usefulness and ease of use — it comes back ahead of where it was. Judge the move at three months, not three days.

Second: a 301 is a forwarding address, not a guarantee. If the page it lands on is built worse than the one it left — thinner markup, slower, less to say — you've moved the sign and lost the shop. There is migration, and there is real migration. Everything can appear to work; working is merely the first thing to check. The job isn't done when nothing's broken. It's done when the new store is better than the one it replaced.

The shape is the same whichever platform you're leaving. What changes is how much of it is thinking and how much is typing.

Somewhere around stage 04 sits the part most quotes leave out: the password reset. Returning customers cannot bring their passwords, so the first thing your most loyal buyers meet on the new store is a hurdle. Design that moment — the email, the wording, the timing — and it reads as care. Ignore it and it reads as a broken site. We've built the generous version of it before: hundreds of legacy Pegoretti owner records held in abeyance until each owner signed in and found their bicycle already there.

Ours included. Where's the redirect map, in writing, before launch? How is migrated data verified against the source? How will the password reset be communicated to customers? What happens to the internal linking? And will you tell me not to migrate if the numbers say stay? A partner who talks only about the import is telling you where your traffic is about to go.

A fair share of migration enquiries turn out to be about something else. The site looks templated; the conversion rate slipped; a competitor launched something handsome. None of those is a platform fault, and a migration is the most expensive available way to find that out — the longer argument is When (Not) to Leave Shopify, and it applies equally to WooCommerce and Magento. If a rebuild on the platform you already have gets you what you want for a third of the money, that's the advice you'll get from us, and you can have it on a call.