Attribution

What is attribution and why is it good to know about?

In the old days, it was hard to measure why people bought something. With the internet, it was different. We could measure that people started in one place (e.g. Google) and ended up buying. It's good to know whether your customers come from Google, Google Ads, Facebook or other websites, for example, so you know where to focus your marketing. If it turns out that none of the visitors from Facebook buy anything, it's basically pointless to do Facebook marketing.

So it's quite important to find out where the visitors who become customers come from.

But it has become more muddy today compared to 2003. That's why the concept of attribution has become important.

Attribution became a hot topic in 2019. Attribution has always been a concept in online marketing, but something happened in 2019 that forced people to think more about it. In this article you can read about what attribution is, what happened in 2019 and why it has become more complex today than it was in 2003.

What is attribution?

Attribution is the term used to describe the concept of looking at which traffic source should get the "credit" for a conversion on your website (it doesn't necessarily have to be a webshop).

Let's look at an example. A customer searches for "dog food" on Google. The customer clicks on your Google Ads ad and comes to your online store to buy dog food. It's pretty obvious that your Google Ads can take credit for creating the sale. They say that the sale is attributed to Google Ads. That's the easy example.

But let's try to make it a bit more complex. Let's assume that the user doesn't buy anything. But he clicked on your Google Ads ad and it cost you 6.40. In principle, a waste of money. But you have set up Google Remarketing. This means that the user is now followed by your ads when he clicks around on pages with Google Remarketing ads. After two days, he clicks on one of your ads (it costs you 3.20) and ends up buying. Now the question is: Should the Google Ads ad or the Google Remarketing ad get the "credit" for creating that sale? Or should they share the credit? After all, you've spent $9.60 to create the sale.

Some would say it's Google Ads. Others remarketing and still others will say "they share the attribution".

It can get even more complex. Imagine the user searches and clicks on your organic result. He doesn't buy anything. But since you have a Facebook pixel on your page, you can pursue him with ads on Facebook and Instagram. And you do. So one day he decides to click from one of your Facebook ads to your page. He still doesn't buy, but now he signs up for your newsletter. You send out a newsletter a month later and describe a new product. He decides to Google the product and sees your Google Ads ad and clicks on it. And here he buys. The big question is "should the sale be attributed to Google Ads, to email marketing, to Facebook ads or to SEO?"

That's the whole essence of attribution!

But it's about to get even harder

Unfortunately, our dear user doesn't just stick to one computer. Many users today have multiple devices. They have a laptop at home, then they have a work computer, a mobile and a tablet. When they switch from their laptop to their mobile, we can "stalk them" with ads on Facebook, for example. But if we use traditional "cookie-based tracking" (i.e. we keep track of where they come from and what they do using a cookie), we no longer have a handle on where they came from originally. So that's why there is a huge number of unknowns in attribution. The user can easily start the journey on their work computer and end it on their iPad at home.

Why did attribution become hot in 2019?

In 2019, online marketers started talking about attribution in earnest. They did so because a new way of tracking sales was introduced. Traditionally, this had been done by dropping a cookie on people's computers and then keeping track of how users entered the site. This worked fine in 2003. But then came the smartphone. Then the tablet. And then we were in trouble. Because they didn't share cookies, so if you started your journey on your computer and ended up converting on your mobile, well, we didn't know how you started the "journey". And it got even worse when Facebook made it possible to stalk the customer from desktop to mobile. In all likelihood, this increased the amount of users who started the journey on a computer and ended it on mobile - and vice versa. We got really big shadow numbers.

Facebook to the rescue....

But in from the sidelines stepped Mark Zuckerberg. He said "I know where your visitors are coming from" and he was right. Facebook had it all figured out. Especially in a country like Denmark, where up to 3 million Danes have a Facebook profile (and surprisingly often are logged into it). Instead of throwing a cookie about where the visitor came from, Facebook could throw that information directly into the database back in Palo Alto. This meant that they could register that Mr. Petersen from Ebeltoft came to your website from Google. And four days later, the same Mr. Petersen saw an Instagram ad. And a week later he signed up for the newsletter and a month later he converted. Facebook knew it all and was ready to share this information with you.

The old tracking method was cookie-based. Facebook's new one was human-based.

Therefore, we are currently in a situation where many companies are finding a middle ground between cookie-based tracking and human-based tracking.

Attribution models

And we need to look at even more. Because while our attribution is likely to become more accurate (at least for people who have a Facebook account and are logged into it), there are different models for how to calculate attribution.

Last click attribution

Imagine a user searches for you on Google. Clicks through to your page via the organic result. Buys nothing. He does the same thing again the next day. And two days later, he searches again. But this time he clicks on your Google Ads and ends up buying something.

Now you have to decide whether the two visits generated by SEO or the one generated by Google Ads should get the credit for the sale.

If you use the "Last click attribution" model, as Google Analytics normally does, then Google Ads gets the credit as it is the last click before the conversion that your user made.

First click attribution

Then there's the reverse model. It's called First click and is not as widely used. But here you say "it's the first touchpoint that should get the credit for the sale". In the case above, it would be your SEO work that gets the credit for capturing a new customer.

Linear / Equal attribution

This is the communist method. This attribution model says "there were three touchpoints to create this conversion. Two from SEO and one from Google Ads. They share with a third each".

Which attribution model should you choose?

It's hard to say anything concrete. The most important thing is that you have given some thought to why you choose the one you do. And that you don't jump around between them so that you use one model one year and another the next. You can change models, but do yourself a favor and think carefully about which one you want to use and why, and then stick with it for many years to come.

We know online marketing in Shoporama

We've been working with online marketing ourselves for decades. As the only shop system in the country, we have spoken multiple times at conferences such as Marketingcamp, SEOday, Shopcamp, Digital Marketing, E-commerce Manager, Ecommerce Day, Web Analytics Wednesday and many more.