The Ecomm Analyst

You track attribution on Shopify by layering three things: Shopify’s own order and UTM data, a server-side event stream so your ad platforms and analytics count conversions the same way, and a dedicated attribution tool that stitches individual touchpoints into one view. Shopify’s native reports tell you what sold. The other two layers are what tell you why. Most stores that feel like their numbers never agree are missing one of the last two.

I set up measurement on stores in the roughly $1M to $20M range for a living, and the sequence below is the one I actually run, not the one a vendor deck would sell you.

What can Shopify track on its own?

Shopify captures the order, the customer, and the referring source for each session under Analytics, plus a “Sessions by referrer” and a marketing report that assigns sales on a last-click basis. If you tag your campaigns with UTMs, those flow into the same reports. That is genuinely useful for a store running one or two channels. It answers “where did today’s orders come from” at a surface level, and it is free with your plan.

The catch is that it only ever credits the last click before purchase, and it only sees sessions that reached your site with tracking intact. Anything upstream of that final click is invisible to it.

Why isn’t the native tracking enough once you run ads?

Two reasons. First, signal loss. Since iOS 14 and the steady erosion of third-party cookies, a large share of the journey between a first ad impression and the final purchase simply does not survive to Shopify’s session data. Second, double counting. Meta, Google, and TikTok each claim credit for the same order in their own dashboards, so if you add up what the platforms report, you get a number well above what Shopify actually banked. I have written before about why your Shopify revenue and ad-platform numbers never agree, and the short version is that nobody is lying, they are just each counting a different thing.

How do I set up tracking that holds up?

The layer that fixes the most for the least effort is server-side tracking. Instead of relying only on a browser pixel that ad blockers and Safari can drop, you send conversion events from your server (or a server-side container) to the ad platforms and your analytics. That recovers a meaningful chunk of the events browsers lose and makes the numbers steadier week to week. On Shopify, this is usually done through a server-side tracking app or a tool that ships its own first-party pixel.

Before any of that, get your UTMs clean. A disciplined, consistent tagging scheme on every paid and email link does more for attribution quality than most paid tools, and it costs nothing. Decide your naming convention once, document it, and audit it quarterly.

Which tools do the stitching?

Once you are spending real money across more than two channels, a dedicated attribution platform pulls the pixel data, the server-side events, and your Shopify orders into a single model so you can compare channels on the same basis. The tools I rank for Shopify mostly share the same core: a first-party pixel, multi-touch models, and a blended revenue view sitting next to platform-reported numbers.

ThoughtMetric, which sponsors this blog, is one I reach for on Shopify because it installs quickly, joins order-level Shopify data to ad clicks, and starts around $99 a month rather than four figures. Triple Whale and Polar Analytics cover similar ground with different reporting emphases. None of these replaces judgment. They give you a defensible number to argue from instead of three platform dashboards that each flatter themselves.

What order should I do this in?

• Fix UTMs and adopt one tagging convention across every link.
• Turn on server-side tracking so browser loss stops eating your conversions.
• Watch blended metrics (total revenue against total ad spend) as your ground truth.
• Add a dedicated attribution tool once you run three or more paid channels and native reports stop being enough.
• Sanity-check the tool against a cheap geo holdout now and then, because no model is right by default.

One habit that saves more headaches than any tool: keep watching blended revenue against blended spend the whole time, even after you install a fancy platform. The dedicated tool is there to explain the movements inside that blended number, not to replace it. When a model tells you a channel is a hero but your blended efficiency did not move when you scaled it, believe the blended number and distrust the model. Attribution on Shopify is not one setup you finish. It is a stack you maintain, and the maintenance (clean tags, working server-side events, an occasional holdout test) matters more than which logo sits on top.

Frequently asked questions

Does Shopify have attribution built in?

Yes. Shopify’s Analytics section includes last-click marketing attribution and sessions by referrer at no extra cost. It is fine for one or two channels and starts to mislead once you buy across several.

Do I need a paid tool to track attribution on Shopify?

Not at first. Clean UTMs, server-side tracking, and a blended revenue view get you a long way. A paid tool earns its cost when you are running multiple paid channels and need to compare them on the same basis.

What is the single biggest fix for bad attribution?

Server-side tracking, followed closely by consistent UTM tagging. Both address the root problem, which is that events are being lost or labeled inconsistently before any model ever sees them.