Attribution in Digital Marketing – How It Works

How It Works (and Why You Should Care)

In this guide, I’ll walk you through the concept of attribution and why it’s critical for online business owners and digital marketers. Attribution is basically the method we use to decide which channel deserves credit for a conversion. Sounds simple, right? The truth is, once you start digging into the data, it can get a bit tricky.

For the examples, I’ll focus on e-commerce, using source/medium as the dimension and “completed purchase” as the conversion event. But keep in mind that attribution exists on multiple levels. My advice: regularly look at your own numbers and question them. You’ll learn the most by analyzing your own data over and over, and soon enough, the insights will start clicking into place.

What Is Attribution, Exactly?

Attribution is, put simply, the judge that decides which channel deserves the “cred” for a sale. It often takes several visits to a website before a customer feels comfortable enough to make a purchase, and days—or even weeks—can pass between the first click and the actual buy. If you’re driving traffic from multiple sources (Google, Meta, TikTok, email, etc.), it’s helpful to understand as much of that journey as possible.

To determine which channel gets the credit, you’ll look at, among other things:

• Attribution event – What counts? Does the user have to click on an ad, or is seeing it enough?

• Attribution window – How many days can pass between the initial “event” and the purchase to still give credit?

• Attribution model – How does your analytics tool decide which channel deserves the credit?

Why Should You Care About Attribution?

When a single sale involves multiple channels, the point of an attribution model is to fairly distribute that sale’s value among the channels involved, based on your chosen settings. But what’s fair is a hotly debated topic, and it varies from one business to another. That’s why it’s so important that you sit down, discuss, and decide how you want to handle attribution for your business.

It’s also very common (and smart) to look at multiple models. By comparing different views, you can get a more well-rounded picture of which channels actually perform best.

However, it’s not all sunshine and rainbows—attribution does have its pitfalls, as we’ll see later.

A Quick Story to Illustrate How Attribution Works

Meet Stefan. One Monday evening, he’s relaxing on the couch, scrolling through his Facebook feed. He comes across an ad from an e-commerce store selling tools. A few products are listed underneath, including a cordless screwdriver—just what he needs, now that he’s a new homeowner.

Intrigued, he clicks through to learn more. The product page highlights the screwdriver’s many benefits, and now he’s really invested. Being the savvy shopper he is, he goes to Google to compare models and prices. After clicking on a few different sites and spending some time exploring, his workout session calls. Before leaving, however, he quickly signs up for the store’s newsletter (right there on the screwdriver page).

Fast forward a few days—weekend is approaching, and it’s time to continue the home renovations. Turning screws by hand is exhausting. During a coffee break, Stefan checks his phone and sees a notification: there’s a newsletter from the same tool shop, complete with recommended products and—best of all—a 10% discount code. Perfect timing!

He clicks on the screwdriver again and makes the purchase immediately.

Summary: Who Deserves the “Credit”?

• Facebook initially sparked his interest. Without that ad, would he have decided to buy a screwdriver right now?

• Google helped him compare and contrast. Would he have purchased via a Google Ad if his workout hadn’t interrupted him?

• Email finally sealed the deal, giving him that extra push (and discount) to complete the purchase.

So, which channel gets the honor of driving the sale? That’s exactly where your attribution model steps in, determining:

• Do we require an actual click for attribution, or is a mere ad impression enough?

• Is it still considered valid if five days pass?

• Should we split the purchase across all three channels? (e.g., Facebook gets 0.3, Google 0.3, and Email 0.4?)

You can tweak these settings based on your attribution preferences. Many e-commerce stores evaluate multiple models simultaneously to see the full picture.

Attribution’s Biggest Challenges

Now that we’ve explored how attribution works and why it’s valuable, here’s the reality check: How reliable is this data, really?

• Multiple devices: If Stefan uses his phone at one point and his laptop at another, how does your analytics tool know it’s the same person?

• ITP and clearing: Apple’s ITP (Intelligent Tracking Prevention) and similar privacy measures wipe data regularly, meaning your analytics may “forget” returning visitors.

• Cookies and consent: If someone declines cookies, your analytics tool may not be able to link their journeys across multiple channels.

• Incognito mode: If someone always surfs incognito, that also disrupts tracking.

All these factors ensure that your data will never fully mirror real-life behavior. We simply can’t track every touchpoint with total accuracy.

So What Do We Do?

The short answer? There’s no one-size-fits-all solution. But you can get close to the truth with creative thinking and by comparing multiple models. How you go about it depends on your specific business, and it’s a big topic—one I’ll cover in more detail in a future post.

Thanks for Reading!

I hope this gave you a deeper understanding of what attribution is and why setting it up thoughtfully is so important. There are plenty of questions left, but by analyzing, challenging, and testing various models, you’ll gradually gain a clearer picture of what truly drives sales for your e-commerce business.

Got more questions? Feel free to reach out—I’d be more than happy to talk attribution!