Is email marketing worth it for Shopify?

Usually yes, if you judge it on the orders it causes rather than the orders it touches. Sending costs little, so email rarely loses money on fees. Where it can lose money is on discount codes handed to customers who were coming back anyway. Hold back a random slice of your list to see what it really adds.

The usual answer is a cheerful yes, often with a big return-per-euro figure attached. Those figures divide the revenue email touched by what the sends cost. And the sends really are cheap. Shopify Messaging includes 10,000 free emails a month on its Basic, Grow, Advanced and Plus plans, then charges $1 USD per 1,000 more. Its abandoned checkout automations are always free. Klaviyo prices its email plans on active profiles and emails sent, and any profile it can email counts, subscribed or not.

So fees are rarely the problem. Credit is.

What one store's data shows

One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts.

None of these rows is an email row, and that is the first lesson. A channel table shows where credit landed. It can't show what an email caused.

Direct holds 57.7% of revenue in that store, in all three views on the Channels sheet. In GA4, a click on an untagged email link can end up there. So can a reader who types your address after reading. The export can't say how much of that 57.7% an email prompted: maybe none, maybe plenty.

One-touch journeys carry 79.5% of that store's revenue on the Journeys sheet, and those buyers took 0.5 days to buy. A loyal customer who clicks your newsletter and orders that afternoon looks exactly like that. With one touch, every model gives the email the whole order. Whether the customer needed the email is a question no path can answer.

The slower journeys are where email can do honest work. In that store's export, journeys with 2 to 3 touches took 12.5 days to buy and held 12.2% of revenue. A reminder that lands mid-decision can change the outcome. One that lands on a decision already made only collects the credit.

The last row is the yardstick. In that store's Break-even sheet, a 40% margin gives a break-even ROAS of 2.5x, which is 1 divided by 0.40. That is arithmetic, not a return anyone earned: it is the bar every cost has to clear. For email, the costs are mostly fees and discounts. For illustration, a €10 code needs €25 of sales the email caused, at that margin.

What the export can't show is any email cost, your list size or which quick buyers were regulars. That takes your email tool, Shopify and a holdout.

Why does the usual answer mislead?

It treats one tool's credit as the whole story, and two common Shopify email tools disagree about the same send.

• Klaviyo counts generously. By default it uses a last touch model with 5-day windows for both email clicks and email opens. It also counts a conversion when Apple Mail Privacy Protection opens an email automatically, unless you change the setting. A customer who never read the email can still land in its revenue.
• Shopify Messaging counts strictly. Its campaign reports use the last non-direct click by default. Shopify's help says a customer who reads the email and then visits another way isn't reflected in those analytics.

So the same send can look like a hero in one report and a passenger in the other. Neither number says what the email caused.

Email can cause real sales, though. In 70 randomized experiments at a large online ticket resale platform, emailed discount offers raised average spending by 37.2% during the promotion window. The researchers found that 90% of those gains did not come through redeeming the offer. That is one platform's result, not yours. It still says two useful things: the email works like an ad, and a discount-code report misses most of what it does.

What can an email report not tell you?

• Whether the order needed the email. Only a group that got nothing can show that. Klaviyo's built-in global holdout needs at least 400,000 profiles, and Klaviyo recommends running one for 3 months. A store with a smaller list builds one by hand, inside one flow.
• What the discount cost. A code report lists the orders that used the code, not the ones that would have happened anyway. A code handed to a customer who was already buying is margin given back.
• What sending too often costs. Unsubscribes and spam complaints shrink the list you can reach next month, and no revenue column shows that loss.

What to do this week

1. Set email's last-click share against its claim. In Shopify, go to Analytics > Reports, filter the Category by Sales and open Total sales by referrer. Read the Email row for last month, then your email tool's revenue for the same dates. Pass: the two sit close, so email's credit rests on clicks you can see. Fail: the email tool claims far more, so treat its figure as a ceiling.
2. Price your email codes. In the same Reports list, open Sales by discount codes and find the codes only your emails carry. Pass: their discount amount is small next to the sales on those orders. Fail: it is a big slice, so next time send half the list the same email without the code, and compare.
3. See whether Klaviyo will hold out for you. In Klaviyo, go to Analytics > Experiments and open the Global Experiments tab. Pass: it opens, so set a holdout of about 5% of marketable profiles for 3 months, as Klaviyo suggests. Fail: Klaviyo says no access means you lack the 400,000 profiles needed, so split one flow at random instead.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Sources, 1 October 2026: Shopify Messaging email pricing (Shopify Help Center); How Klaviyo billing works (Klaviyo Help Center); Understanding Klaviyo message attribution (Klaviyo Help Center); Understanding message conversion tracking (Klaviyo Help Center); Shopify Messaging email analytics (Shopify Help Center); Do Targeted Discount Offers Serve as Advertising? Evidence from 70 Field Experiments (Management Science, via Stanford Graduate School of Business); Getting started with global holdout groups (Klaviyo Help Center); Sales reports (Shopify Help Center)

Related answers

• All Budget & channel mix questions
• How to measure email marketing ROI on Shopify, step by step
• Email capture and attribution: what a welcome flow adds
• Email marketing, in the glossary
• Incrementality testing, explained