M S

asked on

# How to do a linear decay in excel?

I would like to implement a line decay model in excel.

Here is how it is. This is going to be used in attribution of weights or credits to different touchpoints for a user who visits a particular website through multiple channel.

So, if there are 5 interactions. Then, let's say that the total credit it just distributed to all the interactions (5) equally. This would mean that each gets 20% of the credit for that conversion (sale).

But let's say that the 4th interaction was of a certain type (direct visit) and I need to take that credit and distribute it to the previous 3 interactions in a (reverse) linear decay model.

This means that the 3rd interaction would get the highest and then, the 2nd would get lower and the 1st would get the lowest.

This type of model should be applied for any number of interactions and for any position (of the direct visit).

How do I do this?

Here is how it is. This is going to be used in attribution of weights or credits to different touchpoints for a user who visits a particular website through multiple channel.

So, if there are 5 interactions. Then, let's say that the total credit it just distributed to all the interactions (5) equally. This would mean that each gets 20% of the credit for that conversion (sale).

But let's say that the 4th interaction was of a certain type (direct visit) and I need to take that credit and distribute it to the previous 3 interactions in a (reverse) linear decay model.

This means that the 3rd interaction would get the highest and then, the 2nd would get lower and the 1st would get the lowest.

This type of model should be applied for any number of interactions and for any position (of the direct visit).

How do I do this?

ASKER

ok. Let's just say that there are 5 interactions.

The 4 interactions that came before the 5th are all channels. The last (5th) one is a direct interaction.

The weights are distributed equally - so 20% to each interaction.

The weight of the last direct interaction (20%) is taken and distributed to the ones before it.

For ex:

Interaction 1:

20%

Interaction 2:

20%

Interaction 3:

20%

Interaction 4:

20%

Interaction 5:

20%

Then, The final result would be:

Interaction 1:

21.5%

Interaction 2:

23.5%

Interaction 3:

25%

Interaction 4:

30%

Interaction 5:

20%

We got there by doing this:

Interaction 1:

20%

+ 1.5%

Interaction 2:

20%

+ 3.5%

Interaction 3:

20%

+ 5%

Interaction 4:

20%

+ 10%

Interaction 5:

20%

- 20%

The 4 interactions that came before the 5th are all channels. The last (5th) one is a direct interaction.

The weights are distributed equally - so 20% to each interaction.

The weight of the last direct interaction (20%) is taken and distributed to the ones before it.

For ex:

Interaction 1:

20%

Interaction 2:

20%

Interaction 3:

20%

Interaction 4:

20%

Interaction 5:

20%

Then, The final result would be:

Interaction 1:

21.5%

Interaction 2:

23.5%

Interaction 3:

25%

Interaction 4:

30%

Interaction 5:

20%

We got there by doing this:

Interaction 1:

20%

+ 1.5%

Interaction 2:

20%

+ 3.5%

Interaction 3:

20%

+ 5%

Interaction 4:

20%

+ 10%

Interaction 5:

20%

- 20%

I've requested that this question be deleted for the following reason:

The question has either no comments or not enough useful information to be called an "answer".

The question has either no comments or not enough useful information to be called an "answer".

ASKER CERTIFIED SOLUTION

membership

Create an account to see this answer

Signing up is free. No credit card required.

Flyster