How to migrate an online shop platform without losing customers

You've decided to move. The new thing is cheaper, or faster, or does the one thing the old thing wouldn't. What's stopping you is the quiet fear that somewhere in the switch, a chunk of the people who buy from you will simply evaporate.

They can, and here's where they go: into the gap. Not into the move itself. Almost nobody loses a customer because a product page looks different. They lose them because for eleven days there was a dead link in a bio, an old URL 404'd, a file a buyer paid for stopped downloading, and a recurring charge failed silently.

So the whole job is closing gaps, and the order matters more than the tooling. Export first, run both platforms in parallel, redirect every URL you control, and tell people twice. Budget for paying both for 60 days. At Shopify's Basic price of €19 a month on 6 August 2026 plus whatever you're moving to, that overlap costs under €60, and it's the cheapest insurance in this entire process.

Step one: export everything, before you tell anyone

Do this on day one, before you've picked the new platform, before you've announced anything, and ideally before you've cancelled anything.

Pull five things:

1. Your customer list. Names, emails, phone numbers, WhatsApp numbers, addresses. CSV.
2. Your order history. Who bought what, when, for how much. This is what lets you message the right people later.
3. Your products. Titles, descriptions, prices, variants, stock. Plus every image at full resolution, not the resized versions the platform serves.
4. Your URLs. A list of every product URL that exists today. You need this for redirects and you cannot reconstruct it after the shop is gone.
5. Your reviews. Screenshot them if you can't export them. Most of the time you can't export them.

Save all of it somewhere that isn't the platform you're leaving. A folder on your own machine, backed up. The day the account closes is the day the export button disappears, and support will not retrieve it for you.

If the export button doesn't exist, or is behind a tier you're not on, that's information about how the rest of this will go. Budget for the tier for one month rather than losing the data.

What you can take, and what you never can

Be honest with yourself about the second list before you commit.

Comes with you: customer contact details you collected, order history, product photos and copy, your own domain if you own it, your social following, your email list.

Does not come with you: marketplace reviews and ratings, seller badges, marketplace search ranking, any customer whose only contact detail is a platform-internal message thread, and usually your product page rankings in Google unless you redirect properly.

That third one is the killer. If your only way of reaching a customer is a messaging system inside the platform, that customer is not yours and never was. This is precisely why owning your customer list is worth doing before you need it rather than during a migration.

Comes with you but breaks: recurring subscriptions, digital download links already sent to buyers, coupon codes in circulation, and any QR code you have printed on packaging or business cards.

The 90-day overlap

Run both for a season. This is the single decision that separates a calm migration from a bad fortnight.

Days 1 to 30. New platform live, fully stocked, tested. Old platform still live and still taking orders. You send nothing to the new one yet except your own test orders and a handful of friendly customers. You're looking for the things that only appear under real use: a payment method that fails on one bank, a delivery option that doesn't save, an order confirmation that goes to spam.

Days 31 to 60. Switch your bio link, your posts and your signature to the new platform. Old platform still live, still reachable, no longer promoted. Anyone who arrives on an old link still gets a working shop rather than an error. Watch your order count daily, and compare it to the same days last month rather than to last week, because seasonality will otherwise lie to you.

Days 61 to 90. Old platform in read-only or products hidden, redirects in place, one final message to anyone who bought there. Then close it.

The temptation is to compress this into a weekend. Sellers who compress it are the ones who discover in week three that a payment method silently wasn't working, and by then they've lost a month of orders they'll never see and never know about.

Redirects, and what to do when you can't set them

If you own your domain, set 301 redirects from every old product URL to its new equivalent. Not to your homepage. To the specific product, because a buyer who wanted a specific thing and lands on a homepage mostly leaves.

If you don't own the domain, and the old shop was on a platform subdomain, you probably can't set redirects at all. Then do these instead:

• Leave the old shop live for as long as you can afford it, with a line at the top of every product page pointing to the new one.
• Edit the link in every social bio you have, including the ones you forgot about. Check Instagram, TikTok, X, Facebook, Pinterest, YouTube, LinkedIn, and any Telegram or WhatsApp business profile.
• Go back through your last 30 posts and update any link in a caption you can still edit.
• If you have a Google Business profile, update the link there. It's the one people find by searching your name.

Old links in old posts are the longest tail in a migration. Somebody will tap a link from a reel you made 14 months ago, and where that lands is entirely up to what you did this week.

Tell them twice, in the place they actually read

One announcement is not a migration plan. People miss things.

Message one, when you switch the bio link: short, factual, with the new link. Say what changed and what didn't. Nobody cares which platform you're on, so don't explain your reasoning; say "new shop link, same everything else" and get out.

Message two, about ten days later, on the same channel: send it to people who bought from you before, individually if your list is small enough. This is the one that actually works. A message to 140 past buyers converts better than any public post, and past buyers are exactly the people whose bookmark just broke.

If you sell on WhatsApp, put the new link in your business profile and in your away message. If you sell on Instagram, pin a post with it. If you sell on a marketplace and you're leaving, check that platform's rules before you do anything that looks like moving customers off it, because those rules are enforced and the enforcement is account-level.

Layla in Dubai: 300 products and a broken QR code

Layla sells modest-wear and skincare from Dubai. Around 300 SKUs, average order AED 240, roughly 90 orders a month. Instagram is the storefront and WhatsApp is where the conversation happens and the order gets confirmed, which is how commerce works across the Gulf.

She moved platforms in April. Most of it went well. Two things didn't.

The first: 40 of her products had descriptions written in both Arabic and English, and the export mangled the Arabic. She re-typed them over three evenings. If she'd checked one exported row before doing the full export, she'd have caught it in five minutes.

The second is the one worth remembering. She had a QR code printed on 2,000 branded pouches, pointing at the old shop. Those pouches were in stock, in boxes, in her spare room. She'd budgeted zero for it because it hadn't occurred to her that a printed link is a link.

She got lucky: the old platform stayed live long enough for her to keep the URL working for six months while she used the stock up, and she reprinted the next batch pointing at a page she controls rather than a platform. That's the fix, and it's worth doing now even if you're not migrating: any URL you print should be one you own and can repoint, not a platform address you'd have to abandon.

Her order count dipped about 12% in the switching month and was back above the previous level by the second month. That dip is normal and you should plan for it rather than panic at it.

The week-one breakage list

Check every one of these in the first seven days. They're the things that fail silently.

• Payment. Place a real order with your own card, and a second one on the payment method your buyers use most. Not a test-mode order. A real one, for real money, which you then refund.
• Digital files. If you sell files, buy one and check the download link works, the file isn't corrupted, and the download limit isn't set to something absurd.
• Order notifications. Confirm they arrive, and that they aren't landing in spam. Send one to a Gmail address and one to an Outlook address.
• Delivery options. Every zone, every price, including the one you rarely use.
• Mobile page speed. Load the shop on your phone on mobile data. If it's slower than the old one, fix it now, before the traffic arrives.
• Sold-out states. Set an item to zero stock and check the buyer sees the right thing.
• Coupon codes. Any code in circulation, especially in a printed insert or an old post.
• The bio link itself. Tap it from the actual app, not from a browser. Some in-app browsers behave differently.

When not to migrate at all

A migration costs you two to four weekends and a dip in orders. Some reasons don't justify that.

Don't move because the new thing is $8 a month cheaper. That's $96 a year against a fortnight of your life and a 12% dip. Don't move because you're bored of the design; you can change a design. Don't move in your busiest month, ever, for any reason.

Do move when the platform can't do something your business now needs, when your all-in fees have crossed the point where a different structure clearly wins, or when you can't get your customer list out and that has started to worry you. That last one is a good reason on its own.

If you're moving upward because your operations have outgrown a simple link, when Shopify is the right answer sets out the five signals. If you're not sure the destination is right, run it through what to look for in a selling link before you start exporting anything.

What migrating to Sailo actually involves, limitations included

Sailo's plans are $0 free with 10 products and 7 days of analytics, $19/mo or $180/yr for Pro with 100 products and CSV export, and $49/mo or $468/yr for Business with unlimited products and card payments.

Three things to know before you move in.

CSV export starts on Pro. On the free plan you can see your orders but you can't export them, and by this article's own standard a platform you can't export from is a platform you can't leave cleanly. If exportability matters to you, and it should, that's $19 a month you should count from day one.

The 20-product cap on free is real. If you're migrating 60 products, free is not a plan you can migrate into.

And card is Stripe only, requiring the Business plan plus a Stripe account cleared for charges, with Sailo taking 1–3% of the goods after discounts and excluding delivery and tax. There's no Paystack and no mobile money rail. If your current platform takes card in a way Stripe doesn't cover in your country, that part of your business does not migrate, and you should know that before you export anything rather than after.

On the manual rails, bank transfer, cash on delivery, and orders handed to WhatsApp, Instagram, Telegram, email or phone, Sailo takes nothing and never holds the money. It also can't tell you a transfer arrived. Only your bank can.

The customers you lose in a migration are lost in the eleven days when a link went nowhere, not on the day you switched platforms.

Start with the export, tonight

Don't pick the new platform yet. Log into the one you're on and download everything: customers, orders, products, images, URLs. Put it in a folder with today's date on it.

Then open one exported row and read it properly. Check the accented characters, the second language if you have one, the decimal separator on prices. Five minutes there saves the three evenings Layla lost.

Once you have the export, you're free. You can take your time choosing, you can compare properly using link in bio tools compared, and you're no longer negotiating from inside somebody else's account. Which is also the answer to the harder version of this question: what happens if a tool shuts down with no notice at all.