Multi-Touch Attribution. Find out which channels contribute most to user conversion. This repo extends the implementation in https://github.com/eeghor/mta by applying so-called Simple Probabilistic Attribution Model by Shao and Li at the user/lead-path level, as opposed to just aggregate level attribution breackdown as found in that implementation. See mta/shao_user_level_output.py for this addition.
The package comes with the same test data set as an R package called ChannelAttribution - there are 10,000 rows containing customer journeys across 12 channels: alpha, beta, delta, epsilon, eta, gamma, iota, kappa, lambda, mi, theta and zeta.
These are conversion aggregations by path. Suppose there’s a path (customer journey)
a > b > c
with total_conversions equal to 2 and total_null equal to 5. This means that we recorded 2 consumer journeys
a > b > c > (conversion)
and 5 customer journeys
a > b > c > (null)
There’s an option to generate timestamp data if you want to use the Additive Hazard model (the only model that explicitly incorporates exposure times).
- Shao and Li (2011) - Data-driven Multi-touch Attribution Models pdf