Related terms


Email marketing platform with everything you need to design, send campaigns, and perform upmost efficient tracking


Attribution Models

Rules by which the different touchpoints a user passes through before a sale are recorded, and different values are attributed to them. According to the model used, each touchpoint has a variable weight in the conversion path.

Want to know the real performance of your email campaigns? With Acrelia, you have detailed and real-time statistics. Try it now!

Types of Attribution Models

You can distinguish between models that assign the total value to a single channel, for example:

  • Last click or interaction: Allocates 100% of the value to the last time the customer clicked or interacted in any way with the brand before making a purchase. Previous touchpoints are not considered, although they may have logically influenced the decision.
  • Last indirect click: Disregards direct traffic, and credit goes to the last channel prior to conversion.
  • First interaction: Grants 100% to the first touchpoint, regardless of what the last one was. This option values the channel that caused the initial impact above all else.

And attribution models that distribute the value among the touchpoints:

  • Linear: This attribution evenly distributes the total points among all touchpoints involved in the conversion.
  • Time decay: Distributes in a way that the closest touchpoints receive more credit than those farther from the conversion.
  • Position-based: The first and the last touchpoints receive more points (40% each) than the intermediate interactions (20% distributed equally).

Which one to choose?

The most commonly used attribution model is the last click, although it's not always the most suitable.

To choose the most convenient attribution, one should pay attention to the customer journey since it's the basis for measurement.

On one hand, for very short buying processes, it makes sense to use models that assign the total percentage to one step because they are almost impulse purchases. In longer processes, however, a distribution at various touchpoints may be more appropriate because it can take days to complete the conversion.

It's possible to run tests for some time to see which model best suits the characteristics of each business and customize it to serve its purpose: gather information in the best possible way to understand the performance of each channel. Additionally, it's worth noting that omnichannel impacts different sectors differently, and some provide many touchpoints before completing a purchase.