Fractional attribution

What is fractional attribution?

Fractional attribution is when credit is given to multiple sources for an app install. You’re probably already familiar with single source attribution methods such as first and last click attribution, which give 100% credit to one source. Fractional attribution is an alternative approach, acknowledging more than one advertisement’s contribution to an app install.

Each time an ad influences the user, this is called a touchpoint. With fractional attribution, the impact of each ad is measured. By tracking every touchpoint that influenced the user, that measurement can be used to attribute partial credit to any number of publishers, from the very first interaction to the user’s conversion. This is called the user’s journey to install.

How does fractional attribution work?

As the term suggests, fractional attribution means dividing credit into fractions. This can happen several ways, including curve models, equal weighting and various multi-touch methods.

Different types of fractional attribution

There are various multi-touch attribution models, all of which fall under the umbrella term of ‘fractional attribution.’ Finding the right attribution model depends on how you think credit should be divided. Here are a few examples of different fractional attribution models:

• Linear: This model gives all interactions the exact same credit for the conversion. There is no difference between the assigned weights, just even totals calculated by dividing the whole value by the total number of touchpoints along the path to purchase.
• Time decay: This model gives more conversion credit to interactions that happen closer to the conversion event.
• U-shaped: This type focuses on two key milestones, while also acknowledging the middle touchpoints between them.
• W-shaped: similar to U-shaped, but the model covers additional key touchpoints evenly, distributing the rest of the credit to between-stage moments.
• Custom: If a business has already created extensive touchpoint tracking, then they will also have the capability (and, likely, the need) to adjust the weighting of their attribution model to fit their individual reporting needs.