What Are Different Types of Attribution Models?

In the land of marketing, not all data is equal. We want data that provides insights into what influences consumers, and how to optimize our ads to get the most bang for our buck. What ad campaigns are driving the most revenue? How do we get our potential customers to respond to our call-to-action?

There are several attribution models that will help us answer those questions. But which attribution model is the right fit for your business?

First Click

As the name suggests, the “First Click” model would give 100% of the conversion credit to the first clicked ad or keyword that corresponds with that ad. This model helps build the argument that the first ad, the first click, helped move towards a sale eventually.

Last Click

With the Last Click model, this showcases the end of the customer’s journey versus the beginning. Giving credit to the final click, this model is popular with sales-driven companies, as it shows you exactly where the last touchpoint is. This is the location where people clicked last before they made the purchase (or signed up for your newsletter, or became a member, etc.). This is a popular one because it is showing the last channel that leads to revenue.

Last Non-Direct Click

Unlike the first two where the touchpoints include direct website traffic, the “Last Non-Direct Click” is referring to the channel the potential customer went through besides direct traffic. This means that someone may have clicked through email and then came back directly but it will give the credit to the email since that was the last non-direct click.

Linear

In this model, each touchpoint is given equal credit across the conversion path. This includes the first and last clicks, but also any clicks in-between. The credit is distributed evenly, so if there are 10 touches, that equals 10% credit for each. Touches could be anything from paid search, email, social media, and direct channels, —with each one getting an equal amount of the credit. The biggest pro is being able to see across the board what impacts the purchase or lead. The biggest con is that not all clicks are created equal.

Time Decay

Interestingly, unlike the Linear model where all clicks are given equal credit, the Time Decay model gives credit to clicks that are closer in time to the actual conversion. Giving a 7-day half-life, each click that is further out from the conversion gets half as much credit as the clicks that are closest to the conversion.

Position-Based

With an understanding that the first and last clicks often mean the most, the position-based model gives 40% credit to the first and last touchpoints, and the rest is distributed to the other clicks. This has also been referred to as the “U-Shaped” attribution model. This model is great if you know that the first and last touchpoints are the most important, but it might not be as beneficial if that is not entirely the case.

Data-Driven

This model is different from the others, as it uses data from your account, and previous ads, to come up with the best-individualized system for your company. This one also is dependent on a certain, often large, amount of data you’ve already accumulated in order to create a specific model for your company’s needs.

Now that you know the different types of attribution models, it’s time to decide which is right for your business. There is no right or wrong answer. Discuss this internally and debate the pros and cons of each model to find a model that will work best for your marketing team.

At lodestar marketing group, we tend to favor the Position-Based model because we prefer to look at marketing holistically. We know that it takes multiple touchpoints to ultimately convert a prospect to a lead or customer. Many times, paid campaigns will introduce your brand to new audiences, but they ultimately convert through remarketing, organic, or direct. By using the Position-Based attribution model, we can see all of the touchpoints that made a difference in the customer journey but give more credit to the first and last clicks.

If you’d like to learn more about attribution models or need help to determine which one is right for you, send us a note. We’d love to help.