How to Track Email Marketing Attribution on Shopify (Klaviyo Guide)

How to Track Email Marketing Attribution on Shopify (Klaviyo Guide)

Quick Answer: By default, Klaviyo credits a Shopify order to the last email someone opened or clicked in the five days before it (Klaviyo Help Center). That makes its attributed revenue generous: an order an ad or a search brought in still counts as email revenue if an email was opened that week. Track email a second way in GA4, with UTM parameters whose medium is "email", and measure what email adds by holding out a slice of your list. That gives you marketing attribution from two sources and one test of cause.

Klaviyo's number and GA4's number will disagree, and both are credit, not cause. The setup below makes each as clean as it can be; the last sections cover what neither can tell you.

Shopify, your ecommerce platform, holds the orders. Klaviyo sends the emails and records opens and clicks against each customer profile. Attribution lives in the join between the two, and the default settings decide most of what you'll see.

Setting Up Klaviyo for Shopify Integration

Klaviyo's native Shopify integration pulls in orders, products, customers and fulfilment events. It's the foundation for both personalization and attribution: without it, Klaviyo can't link email engagement to purchases.

Klaviyo then syncs historical data and keeps new events coming, including Placed Order, Fulfilled Order and Refunded Order, which attribution depends on. Match your Shopify and Klaviyo time zones: a mismatch can put orders on the wrong day, or against the wrong campaign.

Then check that Klaviyo's onsite tracking script (Klaviyo.js) runs on your store. It records Viewed Product, Added to Cart and Started Checkout: not attribution itself, but the customer journey behind your segments and flows. Without it, Klaviyo sees little beyond email clicks.

Understanding UTM Parameters for Email Attribution

UTM parameters tell Google Analytics where a visit came from: source, medium, campaign, content and term. Without them, email clicks can land in Direct or Referral, and GA4 can't credit email at all.

Klaviyo can add them to every link for you: in Settings, under Other > UTM tracking, turn on "Automatically add UTM parameters to links" (Klaviyo Help Center). utm_source and utm_medium are always on. A setup GA4 reads correctly:

utm_source: "klaviyo", so you can tell Klaviyo's email apart from any other email you send.

utm_medium: "email". GA4 files a visit under its Email channel only when the source or the medium is "email" (or a spelling like "e-mail"), so a medium like "newsletter" or "abandoned_cart" lands in Unassigned. Klaviyo's default for new settings, Message type, already sends "email"; accounts that saved their UTM settings before 2 June 2025 may still send "campaign" or "flow", so check yours.

utm_campaign: The campaign or flow name, such as "summer_sale" or "welcome_flow_1". This is where the type of email belongs.

utm_content: Used to differentiate between different links within the same email, such as "hero_banner," "product_grid," or "footer_link."

utm_term: Less commonly used for email, but can be applied to A/B test variations or specific keywords if relevant.

Keep the naming strict: "klaviyo" in one campaign and "Klaviyo" in another splits your reports. Automatic tagging removes most manual errors, but check the settings once, the medium above all.

Klaviyo's Native Attribution Model and Reporting

Klaviyo reports its own attributed revenue. By default it uses a last-touch model: an order goes to the most recent message a customer opened or clicked, as long as the order falls inside that message's window.

For new accounts, the default attribution window is 5 days for email clicks and 5 days for email opens. So an order placed four days after someone merely opened an email counts as email revenue, even if an ad or a search brought them back. Apple Mail Privacy Protection can register opens nobody made; Klaviyo lets you remove those opens from attribution, and it excludes bot clicks by default for new accounts. Shorten the open window, or stop counting opens, if email revenue looks too good.

The Campaigns and Flows dashboards show open rate, click rate, placed order rate and revenue, and the analytics reports break them down by campaign, flow and message. The revenue there is "attributed revenue": what Klaviyo credits to its own messages under its model. It can't tell you whether the order would have happened without the email.

Integrating with Google Analytics for a Broader View

GA4 gives you the second number. With the UTM parameters above, it sees email next to paid ads, organic search and social, under the same rules for every channel, so you can compare email's share with the others.

In GA4, Reports > Acquisition > Traffic acquisition shows email next to your other channels, under Email when the medium is "email". Explorations let you build path reports and compare GA4's attribution models for the Email channel.

Check that Shopify sends GA4 a purchase event with revenue and items: without it, GA4 has nothing to attribute. GA4's reports now offer three attribution models, data-driven and two kinds of last click; first click, linear, time decay and position-based were removed in November 2023. Setting data-driven next to last click shows how much of email's credit depends on the model.

Common Attribution Models Compared

Each model shares out credit differently, so the same email can look essential or marginal depending on which one you read:

Klaviyo's default is last touch, and GA4 now offers only data-driven and last click; the other models in the table survive as ideas, not as GA4 options. Klaviyo's numbers on their own favour the flows that sit closest to the purchase, like abandoned cart, over the ones that start the relationship. GA4's data-driven attribution learns from your own conversion paths: a different split of credit, not a test of what email caused.

Advanced Reporting and Custom Segments

Both tools go further. In Klaviyo, you can create custom segments based on email engagement, website activity and purchase history: for example, customers who opened a specific email but didn't buy, or those who bought after both an email and a paid ad. Use them for re-engagement or for further analysis.

Klaviyo's Custom Reports combine metrics like Placed Order Value with Email Clicked events, filtered by campaign or flow. You could compare the average order value (AOV) of customers who bought after your welcome series with those who bought after a promotional send.

In GA4, Explorations do the detailed work. A path exploration shows the touchpoints before a purchase, including where email sits, and a funnel exploration shows where email visitors drop out. The attribution models report, under Advertising, sets data-driven next to last click for the Email channel, so you can see how much of email's credit depends on the model.

Overcoming Attribution Challenges: The Multi-Touch Reality

A path to purchase often has several touches: a social ad, a product page, a welcome email, a search visit, an abandoned cart, a reminder email, then the order. Crediting the sale to the last email ignores everything before it, which is the flaw of any single-touch model.

Cross-device journeys add another gap: an email opened on a phone, a browse on a tablet, a purchase on a laptop. Unless cross-device tracking ties the devices together, for example through a login, they look like separate journeys, and the channels that start them get less credit than they earned.

Privacy rules such as GDPR and CCPA, and browser limits such as Apple's Intelligent Tracking Prevention, make it harder still to follow one person across sites and devices. That pushes measurement toward first-party data, collected directly and with consent, and toward methods that work on aggregated data rather than individual journeys.

The Real Problem: Correlation vs. Causation in Attribution

UTMs and model comparisons tell you more about what happened. They're still correlation: even data-driven models find touchpoints associated with a purchase, not whether an email caused it. Nothing on either dashboard flags the difference, and the platform that sends your emails is also the one grading them.

Consider an abandoned cart email. A customer abandons a cart, gets the reminder and buys, and a last-click attribution model credits the email. But was the customer coming back anyway, with the email a reminder at most? Without a test, you can't tell, and the flow's revenue will look better than what it adds.

Most attribution tools work the same way: they track journeys, apply a model (last click, multi-touch or media mix modeling) and show which channels were present before a purchase. That's useful for patterns. It isn't the independent effect of each touchpoint, unless the tool runs a controlled test.

Or take a flash sale: you send the email and sales spike, and attribution credits the email. But the sale itself may have done the work, with the email only announcing an offer customers wanted anyway, or a paid campaign running the same week may have its own return on ad spend claim on the same orders. Credit can't untangle that; only a comparison against customers who didn't get the email can.

Beyond Correlation: Uncovering Causal Impact

The cleanest way to see what email causes is a holdout, causal inference at its simplest: keep a random slice of your list out of a campaign or flow, then compare what the two groups buy. The difference is what the email added; the rest would have happened anyway.

For illustration: 1,000 customers abandon a cart; 900 get the reminder and 100 are held back. If 12% of the reminded group buys and 9% of the held-back group buys anyway, the email added about three orders per hundred customers, not the twelve per hundred Klaviyo can credit it with.

Three checks before you judge an email flow by Klaviyo's revenue:

1. Put the number in proportion. Set Klaviyo's attributed revenue for a month next to your Shopify total and GA4's Email revenue for the same month. Much of the difference comes from opens and five-day windows, which GA4 doesn't count.
2. Tighten the rules and see what's left. Remove Apple Mail Privacy Protection opens from attribution, or shorten the open window, and watch how much attributed revenue survives (Klaviyo recalculates past data when you change these settings, which can take up to 36 hours). What disappears was credit, not sales.
3. Hold out a slice. Keep 10% of recipients out of your biggest flow for a few weeks and compare what each group buys. That's what the flow causes.

Rule of thumb: use Klaviyo's revenue to compare one email with another, and a holdout to decide what email is worth.

FAQ

Q1: What is the main difference between Klaviyo's attribution and Google Analytics attribution? A1: Klaviyo credits its own messages: by default, the last email opened or clicked in the five days before an order. GA4 credits every channel under the same rules, with data-driven or last-click models, and counts a visit as Email only when the link is tagged. Opens and five-day windows are why Klaviyo's number usually runs higher.

Q2: How do UTM parameters help with email marketing attribution? A2: They tell GA4 that a visit came from an email. Set utm_medium to "email" so GA4 files those visits under its Email channel, and use utm_campaign and utm_content to tell campaigns, flows and links apart.

Q3: What is a good attribution window for email marketing? A3: Match it to how long your customers take to buy: shorter for impulse buys, longer for considered purchases. Klaviyo's default is 5 days for both email clicks and email opens, and whether to count opens at all is the bigger choice, since an open is weak evidence that the email did anything.

Q4: Can I use both Klaviyo's internal attribution and Google Analytics attribution? A4: Yes. Klaviyo's reports are quick and email-specific; GA4 puts email next to every other channel. Where they disagree, the gap usually comes from opens and window rules, and neither tells you what email caused.

Q5: Why is causal attribution important for email marketing? A5: Because email reaches many people who were going to buy anyway. Causal attribution asks whether an email caused a purchase or merely came before it, so you stop funding flows that only take credit, and keep the ones that add orders.