https://github.com/google/fractribution

https://github.com/google/fractribution

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/google/fractribution
• Owner: google
• License: apache-2.0
• Created: 2020-08-21T05:59:21.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2022-04-07T05:54:05.000Z (over 2 years ago)
• Last Synced: 2024-05-14T00:11:14.283Z (6 months ago)
• Language: Python
• Size: 982 KB
• Stars: 47
• Watchers: 8
• Forks: 12
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-marketing-machine-learning - fractribution

README

        
# Fractribution code base

## Attribution Overview

In a marketing context, Attribution involves identifying the set of user actions

("events" or "touchpoints") that contribute in some manner to a desired outcome,

and then assigning a value to each of these events. Users might click on

multiple ads before converting. This can make it challenging to assign proper

credit to the different marketing channels. For example, should all the credit

go to the last ad the user saw, or the first ad? Should all the ads share in the

credit equally? Or should some other rule be used to determine how to distribute

credit?

## Data-driven attribution (DDA)

DDA attempts to algorithmically work out a fair weighting of credit among

marketing channels. For example, a particular display ad might not convert

immediately, but users who click the display ad might be much more likely to

convert later on. In this case, the display ad should get credit, even though

it may not be the first or last ad on a user's path to conversion.

## Fractribution Package

Google Marketing Platform products already support DDA. This Fractribution

package is a DDA algorithm that generates **user-level fractional attribution

values** for each conversion. The advantage of user-level attribution is that

the attribution values can later be joined with custom user-level data (e.g.

transaction value, lifetime value etc). This can be useful when regulation or

data policy prevents ecommerce/revenue events from being shared with the Google

Marketing Platform.

Please see Fractribution_Slides.pdf file in this directory for more background

on use cases and details on the DDA algorithm.

## Using Fractribution

For more instructions, including a tutorial for running Fractribution over

sample GA360 data from the Google Merchandise Store, from see py/README.md.

## Directory structure

```bash

fractribution

├── README.md

├── py

├──── README.md

├──── main.py

├──── fractribution.py

├──── templates/

├──── Dockerfile

└──── start.py

```

Disclaimer: This is not an officially supported Google product.