A Mailchimp to Klaviyo migration: what breaks, and what to check first

What comes across, and what quietly doesn't

The export itself takes an afternoon. Contacts, tags and unsubscribes come out of Mailchimp as a file and go into Klaviyo without much drama. Everything that isn't in that file is the actual project.

So here's the honest framing. A migration isn't a data job. It's a permissions and reputation job, and the file is the easy half of it.

The export nobody checks is the one that matters

Take the suppression list before you take the subscribers. Unsubscribes, complaints and hard bounces have to be sitting in Klaviyo before the first send goes out, and it's the item most likely to get skipped on a Friday afternoon.

• Everyone who unsubscribed in Mailchimp, imported as suppressed, instead of simply left behind
• Complaints and hard bounces, for exactly the same reason
• Where each address came from, so a consent question has an answer in a year
• Anything Mailchimp itself imported from an older platform that nobody can now account for
• The unengaged, who are a decision to make, not a box to tick

That last line is the opportunity. A migration is the cheapest chance you'll ever get to leave dead weight behind, because nobody has to be talked into deleting anything. They just don't get invited across.

Sending reputation doesn't travel with the list

Reputation belongs to a domain and to the infrastructure sending for it. Never to the addresses. Change platform and, as far as Gmail's concerned, you're a new sender who's turned up with a suspiciously large audience.

So publish SPF, DKIM and DMARC on the new sending domain, and confirm they pass on a real send, not on a preview. Then start with the people most likely to open, and widen from there over days.

And if deliverability is already poor in Mailchimp, moving won't fix it. It'll hide it for a fortnight. Clean the list first, move second. What we do inside Klaviyo starts in the same place, for the same reason.

The pop-up that stops working the day you switch

Forms are the part that breaks silently. The Mailchimp pop-up dies the moment its script comes off the theme, and if the Klaviyo form isn't live and tracking the same day, you spend a week acquiring nobody.

Rebuild the form, point it at the list the welcome series reads, then watch the submit rate for a week instead of assuming parity. And check the two quiet sources while you're in there: the marketing box at checkout, and account creation.

Rebuilding the flows is the good news here

Nothing automated ports. A Mailchimp customer journey and a Klaviyo flow don't mean the same thing, so a copy would only be a translation of somebody's old compromises.

Rebuild in the order the money sits: welcome series, abandoned checkout, then browse and post purchase. The flow library is what that stack looks like finished, and the first two usually cover the cost of the move on their own.

The gain is real, and it's why most brands move in the first place. Trigger filters, flow filters and conditional splits let you describe a customer situation Mailchimp could only approximate with tags.

The order to work in, and the week it goes wrong

1. Export contacts, tags and the suppression list, and agree out loud who isn't coming with you
2. Build the Klaviyo account and connect Shopify first, so orders and profiles are there before the people are
3. Publish authentication on the sending domain and check it passes on a real send
4. Import suppressions, then subscribers, then switch the sign up forms over on a single day
5. Rebuild welcome and checkout, turn them on, and leave the campaign calendar alone for a fortnight
6. Warm up on engaged segments, widen slowly, then close the Mailchimp account instead of leaving it quietly running

The week it goes wrong is always the one where both platforms are live and nobody owns the calendar. Pick a date, tell everyone, and send from one place from that morning.