Types of Attribution Models

Mikkel Settnes Updated by Mikkel Settnes

Attribution modeling is a framework for analysing the individual touch points of your customer journey and understanding which channels, campaigns and activities receive credit for a conversion.

There are several different attribution models but at Dreamdata we provide you with the top 6 which are: First and Last-touch, Linear, W-shaped, U-shaped and our newest Data- Driven attribution model. Each attribution model distributes a certain amount of credit (in%) of a conversion across each touchpoint.

Within Dreamdata you can switch between the models and compare their value distribution. This will help you to get an idea of what touch points to consider in your customer journey as depending on the model the amount of credit per touchpoint varies. By analysing each attribution model, you can for example get a better idea of the ROI for each channel or Paid ads.

Which attribution model should I choose?

That is a difficult question as indeed there isn't necessarily a one fits all solution. It depends on what it is that you would like to measure and analyse. In general it is important to understand that we distinguish between Position based and Data Driven attribution models.

POSITION BASED ATTRIBUTION MODELS

DATA DRIVEN ATTRIBUTION MODELS

  • First touch
  • Last touch
  • Linear
  • W-Shaped
  • U-Shaped

  • Data Driven

Rule-Based / Position-Based Models assign credit based on heuristic business rules often tied to the position of the touchpoint in the customer journey. All these are “standard” rule setups. Dreamdata also allows you to create your own complex business rules (referred to as custom attribution models), which can divide credit based on any combination of event, campaign, channel, source and/or position in the customer journey.

The rule-based models are limited to consider only a single journey at a time and follow a fixed and pre-defined rule-set.

The Data-driven model defines its own rules when looking at all your journeys at the same time.

Use Data-driven attribution if you want to attribute more weight to a touchpoint that is typically influencing the customer journeys.

Position based attribution models are a set of rules that determines how value is assigned to your customer's different touches. Each touch is assigned a certain amount of credit (%), usually depending on when it occurred in the journey.

LINEAR

FIRST TOUCH

LAST TOUCH

W SHAPED

USHAPED

Every touch is assigned an equal amount of credit. In the previous image you can see that every touchpoint received 20% of the credit.

What is the Linear model good for?

To investigate all the touch points that are worth pursuing.

Only one touch point received 100% of the credit. In our example 100% of credit has been assigned to the first touch point which was organic.

What is the First touch good for?

It helps you to understand what started the journey in order for a lead e.g. to become a MQL.

Only one touch point received 100% of the credit. In our example 100% of credit has been assigned to the last touch point which was calls.

What is the Last touch good for?

It helps to understand what was the last activity that happened before a lead became e.g. a MQL.

30% of the credit is given to both the first and last known touch. Then, 30% is also given to the touches containing conversions. The remaining 10% is divided up between the remaining known touches.

What is the W Shaped good for?

It helps to understand which touch points are more valuable and which ones, though the value is ,ow should still be considered.

40% of the credit is given to both the first and last known touch. The remaining 20% is divided up between the remaining known touches.

What is the U Shaped goof for?

It helps to understand which touch points are more valuable and which ones, though the value is ,ow should still be considered.

You may for example choose the W- Shaped model as your primary attribution model for reporting and analysis. This is a good option as it does not distribute the credit equally amongst all the touch points but highlights the ones with more importance as well as includes those that have little impact but contribute to the journey.

Generally speaking you don't need to limit yourself to one attribution model and stick with it. It is recommended to compare performance under each model to understand the importance of multiple touch points in the customer's journey.

If you're on Dreamdata's Business plan you can create custom attribution models.

Data-Driven Models replace specific business rules with a mathematical algorithm, that uses data from all the journeys to dynamically determine what touch points influenced the given Stage and therefore should receive credit.

The benefit of rule-based models are the explainability. You can look at a single customer journey and follow the rules of the model and understand the credit scoring. On the other hand, the Data-driven models consider all the journeys at the same time. This makes the model able to reason about what the journeys have in common, but at the cost that you can no longer understand the attribution fully by looking at a single journey.

To read more about the Data-Driven Models click here:

To toggle between the different attribution models, click the Attribution filter located at the top of the page. While switching between the different attribution models you will notice that the value attributed across your different channels and sources will change.

Note: The attribution models can be applied from the first interaction to the first stage of your sales/marketing funnel or on your entire funnel. To understand the result they provide, its important to know that we are always considering the touch points before a lead reached a certain stage. In our example below that would be all blue dots before the red one which is when a person "signs up" for something. We will always count from the first interaction to the stage you selected.

How did we do?

Revenue Segmentation

Contact