fraction.
Consumer behavior marketing insights driven by machine learning

Complex Modelling of Consumer Behaviour: Inferential Statistics & Machine Learning Driven Insight and Prediction

“The T-DAB team helped us to explore new ways to analyse and extract useful insight from a large and complex dataset... for a 'lay' audience.”

David Boon | Director | Fraction

Fraction Logo

Case Study
THE CHALLENGE

A multinational manufacturer needed to improve their consumer modelling and analytics to drive their retail and marketing strategy.

Previous analytical approaches employed provided some insight but were relatively crude.

The marketing teams needed to run multiple scenarios to understand how changing consumer perceptions and targeting certain demographic groups may allow them to alter the market share of different products.

In addition, the retail teams wanted to understand what drives consumers to choose similar rival products.

THE SOLUTION

The Data Analysis Bureau (T-DAB) worked with Fraction to develop a Proof of Concept (PoC) dashboard tool to help a major FMCG company marketing team mine datasets for insights regarding drivers of brand choice by consumers. The dashboard allows a non-technical user to define the parameters they are interested in, before a machine learning algorithm mines the dataset, returning the most important features. The user can then interactive plot the data to extract further insights.

T-DAB developed an automated data mining process and leveraged machine learning algorithms to help the client better understand where to focus their resources and develop a strategy to target specific consumer groups and market sectors.

Working with multiple lines of business, the team identified an early opportunity to consolidate multiple data sources and requirements, and automate statistical processes. This allowed T-DAB to disentangle drivers of consumer behaviour that were previously hidden and illustrate interactive relationships between variables that the client could utilise in their early concept testing.

The team then used the initial development to build machine learning predictive algorithms, and produce powerful tools to interactively explore consumer relationships and test market scenarios.

THE RESULT

The multiple client teams are now able to develop and deliver more effective, targeted market campaigns to impact the market share of different products.

The cutting edge methods employed by The Data Analysis Bureau have delivered much more powerful and actionable consumer insight tools and enabled the client to understand and disrupt competitor trends.

Find out more about the project and how you can explore machine learning for consumer behaviour: