There’s a moment, in every platform migration I’ve worked on, where the search traffic dips. It’s predictable. Google has to recrawl the new site, follow the redirects, decide the new pages mean what the old pages meant, and rebuild its picture of who you are. That takes weeks. Sometimes longer. During that period, organic visitors fall, and the only question is how far and for how long.
The answer to that question isn’t “wait it out.” It’s “keep paid traffic running so customers can still find you.”
This is the part nobody puts in the migration brochure.
The Safety Net
Paid traffic — Google Shopping, search ads, Performance Max — is the bridge between two stable states. Before the migration, organic search is doing the heavy lifting. After the migration, organic search recovers. In between, paid traffic carries the weight. If it’s running, the customer journey doesn’t break. The phone still rings, the orders still come in, and the dip in organic traffic looks like a small dent rather than a gap in the road.
If paid traffic isn’t running — or if it stops running at the wrong moment — there’s nothing else holding the line.
I learned this the hard way last month. I helped an ecommerce client move their store from one platform to another. They were a manufacturer with steady search traffic and a long-running paid campaign. The migration itself was clean. The site went live, the redirects worked, the new platform performed faster than the old one. By every visible measure, it was a successful cutover.
What we didn’t notice — for eight quiet days — was that the paid side had stopped working.
Twenty-Eight Versions in Seventeen Days
The reason it had stopped was the merchant feed. Anyone selling products through Google Shopping has one. It’s a structured file — a spreadsheet, essentially — that tells Google what you sell, how much it costs, how it ships, and where to find each product page. Google reads it on a schedule, builds your listings from it, and matches them against the searches people run.
The feed sounds like a simple thing. It isn’t.
When you migrate platforms, every line in the feed has to be rebuilt. The product URLs change. The product identifiers change. The shipping rules change. Google’s category tree may have moved on since the last time anyone looked. The image cache has to be busted. The feed labels have to match the campaigns, which have to match the account-level shipping settings, which have to match the country settings on the products themselves. And buried in dashboards nobody documents are manual edits — stored overrides set up years ago by people who’ve since moved on — that silently rewrite what you upload.
I rebuilt the feed twenty-eight times in seventeen days. Each version fixed something the last one exposed. I have the version history. I am not exaggerating.
That’s the gap between “just install the app and it’ll work” and what actually has to happen for the safety net to hold.
When the agency or the platform tells you the feed is automatic, what they mean is it’s automatic until it isn’t. The day it isn’t, you find out by losing your paid traffic, which means you find out by losing your revenue.
What This Means If You’re About to Migrate
If you’re thinking about moving your website to a new platform — Wix to Shopify, EKM to Shopify, WooCommerce to anything — there are four disciplines worth applying before the cutover.
Plan the paid side first. Before you book the migration date, work out who is going to keep your Google Shopping and Google Ads campaigns alive across the cutover, what the feed migration actually involves, and who is responsible for verifying it end-to-end on the new platform. If the answer is “the platform handles that automatically,” ask who is going to verify the automation actually worked. Twice.
Run paid traffic at full strength through the migration window. This isn’t the moment to economise on ad spend. Organic traffic is going to dip. Paid traffic is what keeps the customer journey intact while Google catches up on the new site. The pound you save by pausing campaigns during a migration is a pound you lose three times over in missed sales.
Insist on a parallel-running window if you’re switching agencies. Don’t fire the old supplier on the day you sign with the new one. Run them in parallel for thirty days. Pay for the overlap. The cost of the bridge is much smaller than the cost of an eight-day silent gap where neither thing is fully working and nobody knows yet.
Refuse the recipe. The most dangerous handover anyone gives you is a single sentence: “just install the app and it’ll handle it.” That isn’t a handover, it’s a recipe. A real handover is a written document — account access, campaign architecture, manual overrides, conversion tracking, the lot. If your outgoing supplier won’t produce one, that’s data. It tells you the institutional knowledge is in their head and it’s about to walk out the door.
The principle in one line
A platform migration isn’t a website project with a paid-traffic afterthought. It’s a paid-traffic project with a website cutover inside it. Plan it that way and the migration is uneventful. Plan it the other way and you find out what’s broken when the revenue stops.
I came out of the last month believing two things I didn’t believe at the start of it.
First, that organic traffic and paid traffic are not separate concerns during a migration. They’re the same concern, viewed from two sides. The migration succeeds when both survive the move. It fails when either one breaks while the other isn’t watching.
Second, that the work to keep paid traffic alive during a migration is substantially harder than anyone names it. Twenty-eight feed versions is what it actually takes. If somebody tells you it’s a five-minute job, they haven’t done it.
If you’re about to migrate anything load-bearing, that’s the conversation worth having before the cutover, not after.
Tony Cooper
We Build Stores
tony.cooper@webuildstores.co.uk
07963 242210
P.S. If you’re considering a platform migration this year — Shopify, WooCommerce, BigCommerce, anywhere — reply to this email and tell me what you’re moving from and to. I’ll send you the migration checklist I now use, including the merchant-feed and paid-traffic sections I learned the hard way to put first. No charge.
