What are the different attribution models and why are they important

When it comes to the measurement of digital marketing activity, most people will focus on the Last Click revenue of a channel or campaign. This is a very important metric to measure and should always be considered as one of the main KPI’s. However, by only looking at Last Click, there is the potential that as a marketer you are missing out on the bigger picture and could be hurting the long-term growth of the brand and missing out on revenue long term.

Alternatively, considering the full customer journey and all the possible touchpoints that a person could potentially have with a brand before they make the final purchase could lead to a very different marketing strategy.

A great example of this is email, which is a great driver of revenue from a Last Click point of view, therefore, a lot of resources and time gets focused on this channel. However, growing the CRM database is the hard part and is often overlooked as a KPI when evaluating the effectiveness of channels like social, display and non-brand search for both organic and paid results.

What are the different attribution models?

Example customer journey: A customer sees an ad on Facebook and clicks through to the site to browse the products. She then searches for similar products online and sees an organic search result and clicks on the website again. After a few days of consideration, she comes back directly on a brand search and makes a purchase.

Last Click – based on this model in the scenario above, all revenue and credit for the transaction will be given to the organic brand search result.

First Click – using a First Click model, all the revenue and the transaction will be attributed to the Facebook ad, as this is where the customer first interacted with the brand.

Linear Model – Using this model, each touchpoint with the brand will gain a share of the revenue. So in the example above 33% will go to the social ad, 33% to the non-brand organic result and 33% to the direct brand search.

Position-Based Model – Using this attribution model, 40% of the credit is assigned to the first and last interaction, with the remaining 20% credit evenly distributed between the middle interactions.

Where to find this data

In Google Analytics it is now possible to view these different attribution models with the Model Comparison Tool.

From here you can see how each model compares to each other.

This means you can see where the strength of each channel is and how they feed into each other. Very often channels are seen in isolation and dismissed as ineffective because they are not viewed in the right context. The data below is an example of First Click Vs Last Click and as you can see, almost reverses which channels perform best.

Direct often takes a lot of credit for revenue when looked at from a Last Click point of view, but feeds off the work of other channels. Whereas Display and Organic search are often an underestimated part of the marketing mix.

Making the most of this data

As with all data, it is only useful if you can gain actionable insights from it. By having a better view of the full user journey as a marketer, you can make more informed long-term strategic decisions that will help grow your brand long term. It is often the case where a business is trying to cut costs that the channel that contributes least to revenue directly will be cut first. However, if you have the data to prove that that channel contributes significantly to growing the customer base and bringing in revenue in the medium to long term then you will be in a much better place to drive long term growth.