Access to Cash

Machine Learning and Mobile Data Improves Aid Delivery in Togo

Featuring GiveDirectly and the Center for Effective Global Action

Game Akoko, a cash transfer recipient in Togo.
Game Akoko, a cash transfer recipient in Togo. Credit: GiveDirectly

A team of researchers with GiveDirectly and the Center for Effective Global Action, helped the Togolese government target cash transfers using a machine learning and mobile data approach.

  • Over 140,000 Togolese have received $10 million in cash through our partner GiveDirectly.
  • Machine Learning (ML) approach reduced the number of people who would have been incorrectly excluded from receiving benefits by 4–21%.
  • Machine Learning resulted in a fair cash distribution across demographics: by gender, age groups, types of households, religious backgrounds and ethnicity.