Skip to main content

Procter & Gamble - Case Study

Procter & Gamble - Case Study

Machine learning for improving products

The challenge

Procter & Gamble (P&G) own world-renowned personal care, health and hygiene brands. Their brands include Pampers, Herbal Essences, Olay, Ariel and Gillette.

P&G wanted to move towards ‘measured behaviour’. They were keen to learn customer’s habits through direct observation.

This can offer more accurate predictions on what people think about their products.

But they struggled to find systems to enable them to collect data effectively through direct observation.

P&G decided to develop a machine learning platform in-house.

They needed a steer on how to bring it to life and equip employees with the skills to do so, which is where NICD came in.

The Catalyst

Sorry, you need JavaScript to view this video

Resolution

NICD worked with P&G to create new predictive systems for capturing and processing customer insights.

They helped them to better understand skills needed to develop such systems in-house.

P&G wanted to know how they could generate even more data to allow it to:

  • develop new product capabilities
  • upskill their employees in the process

P&G needed someone with the knowledge to drive the project forward in a collaborative way.

NICD helped P&G work through the challenges to meet the aim of gathering more data.

P&G has now advanced their knowledge of measured behaviour systems.

They use machine learning to make informed decisions about what consumers think. This can be used to improve their products.

This project has changed the way P&G approach the development of new data models.

P&G learned from NICD is that their principle-based approach is a welcome step up. They're now applying that knowledge to a number of projects.

Solution

Procter & Gamble:

  • adopted a measured behaviour system, which:
    • uses machine learning to make informed decisions about what consumers think about products
    • can be used to improve products
  • regarded NICD as a ‘critical friend’ during the project to help challenge the feasibility of the machine learning platform required  
  • learned how to develop similar systems in the future and the skills required to do so