How Do You Measure Conversion When Your Checkout Isn't Shopify's?
When your checkout is not Shopify’s, measure conversion by treating completed orders as the source of truth, tracking sessions across the storefront and checkout domains, and reconciling your analytics to the order system. Then break the funnel into steps so you can see where shoppers leave.
A checkout on another domain or platform splits the shopper’s journey across two systems. Conversion numbers can quietly break at the handoff. This guide is vendor-neutral and works whichever tools you use.
- Use completed orders from your order system as the source of truth.
- Make sure a session carries across domains. A broken handoff makes one shopper look like two.
- Measure the funnel in steps, not only a single rate.
- Reconcile analytics to orders on a schedule and investigate any gap.
Why is conversion harder to measure with an off-site checkout?
Most analytics tools follow a visitor on one domain. When checkout runs elsewhere, the visitor can appear as a new session at the handoff, the original traffic source can be lost, and the checkout domain can show itself as the referrer. The result is inflated sessions, understated conversion on the storefront and misattributed revenue.
How should you set it up?
First, decide that orders in your order system are the truth. Second, make the session identity travel across the handoff, for example through cross-domain tracking in your analytics tool, and exclude your own checkout domain from referral sources. Third, capture checkout steps on the second domain so the funnel has no blind spot.
How do you check that the numbers are right?
On a schedule, compare analytics orders and revenue with the order system for the same days. Small differences are normal. A growing gap is a signal that the handoff or a tag is failing. Why Shopify Analytics and GA4 show different numbers explains common reasons for ordinary differences, and platform ROAS versus store data shows the same reconciliation idea for ad reports.
What should you measure at each step?
Use a funnel: product view, add to cart, checkout start, payment attempt, order complete. Report each step on its own. A single overall rate hides where the loss happens. What shoppers do after an ad click shows how step-level views help, and how to prove a site change improved conversion covers confirming results.
A Noibu setup can include a checkout on its own domain alongside the storefront domain, so you can look at both parts of the journey. The CRO agent then finds where shoppers drop off, builds changes and hands them back for your approval. To see your current picture, run the free website audit.
Frequently asked questions
What is the source of truth for conversion?
Why do sessions double when checkout is on another domain?
What is cross-domain tracking?
How often should I reconcile analytics to orders?
Should I report one conversion rate or a funnel?
The short version
Trust orders, keep the session whole across domains, and measure each funnel step. Noibu is AI for ecommerce operations, the always-on optimization system that keeps online stores running at their best. A team of supervised AI agents finds what's broken, slow, or leaking revenue, does the work, and hands results back for human approval. Trusted by retailers like Free People, Guess, Liquid I.V., and Ariat.
