Attribution Model for Digital Marketing Analytics 3

Attribution models in digital marketing

If you liked the content, please share the knowledge :)
  • Yum

Let’s learn Attribution model to understand how marketing channels contribute to goal conversions in businesses.

Have you ever wondered how marketers allocate budgets in different marketing channels for their company? They must follow some marketing models to decide how different marketing channels are contributing at different touch points on goal conversion. One can check the contribution of all the marketing channels.

Here is the deal.

For example, my website site got one conversion, but the person visited my site multiple time before fulfilling my goal.

Multi Channel Funnel Attribution model

 

The user first visited my website from Social Media, then from Email, then Organic search and lastly from Paid search which led him to conversion.

Now, the question is who is contributing towards the conversion. Maybe all or maybe either. There are companies who follow different models and attribute different values to each step.

The different pre-defined attribution models are as follows:

  • First click model
  • Last click model
  • Linear model
  • Position based model
  • Time decay model

That’s not all, you can define your own model as well. I will take these pre-defined models one by one.

Before doing that I will create few situation of goal conversion.

Multi channel funnels Digital Marketing

I got four conversions with the above multi-channel funnels. Now, let’s see how they contribute to conversions for different models and what is the success of each channel overall.

 

First Click Attribution Model:

In this case the first channel gets the entire credit (100%), whereas rest gets zero.

 

Contribution
Social Media Organic Search Email Paid Search Referral
Goal 1 100 0 0 0 0
Goal 2 0 100 0 0 0
Goal 3 0 0 100 0 0
Goal 4 0 100 0 0 0
TOTAL 100 200 100 0 0

 

So, finally the deal is;

Organic has got the maximum contribution and the best performing channel if First Click attribution is considered. Whereas Social Media and Email has same value with Paid search and referral are worst with zero credit.

But, the story may change if we go by some other attribution model.

 

Last Click Attribution Model:

In this situation the last channel (or the channel closest to the goal) gets the entire credit (100%) and rest others get zero.

Multi channel funnels Digital Marketing

Contribution
Social Media Organic Search Email Paid Search Referral
Goal 1 0 0 0 0 100
Goal 2 0 0 0 100 0
Goal 3 0 100 0 0 0
Goal 4 100 0 0 0 0
TOTAL 100 100 0 100 100

 

In this case; Social Media, Organic Search, Paid Search and Referral are contributing equally to the conversions and Email fails in contributing.

 

Linear Attribution Model:

This might seem interesting as everyone gets equal credit. So if the conversion completion happens in five steps, each channel will get 100/5 = 20% credit.

i.e. Contribution of each channel in a multi-channel funnel = 100/no. of touch points.

Multi channel funnels Digital Marketing

Contribution

Social Media

Organic Search Email Paid Search

Referral

Goal 1

25 25 25 0

25

Goal 2

25 25 25 25

0

Goal 3

0 25 25 25

0

Goal 4

20+20 20 20 20

0

TOTAL

90 95 95 70

25

 

So if we consider the Linear Model, Organic Search and Email seem to perform the best and the referral being the worst.

 

Position Based Attribution Model:

This model is somehow very close to what marketers look for, it gives credit to both branding and helping in goal fulfillment. The first click model we generally use for branding campaign and the last click model is preferred for sale or acquisition.

But, in case of Position Based Attribution model the first channel and the last channel gets 40% credit each and the rest 20% is equally distributed to the rest of the intermediate channels and they are considered for retention.

For example, if there are 5 steps; first channels gets 40%, last channel gets 40%, and second, third, fourth channel gets 20%/3 = 6.6% each.

Multi channel funnels Digital Marketing

Contribution

Social Media

Organic Search Email Paid Search

Referral

Goal 1

40 10 0 10

40

Goal 2

10 40 10 40

0

Goal 3

0 40 40 20

0

Goal 4

6.6+40 40 6.6 6.6

0

TOTAL

96.6 130 56.6 76.6

40

 

So, the clear winner is Organic search leading by a huge margin in this case.

 

Lastly,

 

The Time Decay Model:

In this case, the credit keeps on decreasing with the distance from the goal and also the value reduces if the time difference between two channels is large. It follows the radioactive decay method. The exact value of each touch point will be discussed in Time Decay blog.

In that blog, I will describe the entire attribution credit with the mathematical model in details.

Time Decay Attribution Model in Digital Marketing Analytics

The channel closest to the goal gets the highest credit (value will be defined in the other blog) and since the time gap between the green channel and yellow channel is high in case 2 hence the contribution value is small. The ratio of the contribution depends on the distance from the goal and also the time lag.

This is considered to be the best model yet, as the credit of channel months old will have credit almost close to zero as in real life the effect of the channel fades away.

It’s the job of the digital marketer to choose between the attribution model for their business type which suits their goals and objective. This also helps them allocate the budget for the respective channel.


If you liked the content, please share the knowledge :)
  • Yum

Leave a comment

Your email address will not be published. Required fields are marked *

3 thoughts on “Attribution Model for Digital Marketing Analytics