Improving national sales
conversions using machine learning
The Customer
Our client is one of Australia’s largest enterprises with head offices in each state. Vying for market share against more than one aggressive competitor, they engaged Polymathian to assist in identifying and converting potential customers.
The Problem
Given a large group of customers across the country, our job was to categorise each potential customer into a defined market segments using customer specific information (e.g. household information, credit history, marital status, etc). Our client had identified a primary market segment that was most likely to result in sales conversion.
The Challenges
- Enormous amounts of historical data from multiple sources
- Large set of attributes to analyse per data record
- Structured and unstructured data requiring a complex mix of processing methods
- Conducted at a national level, dealing with an entire population
- More than one competitor in some regions required manipulation of the classification algorithm
- Categorisation ONLY required of populations in areas of interest
The Solution
Following extensive exploratory data analysis and processing, supervised machine learning algorithms were applied to the data to achieve segmentation of the potential customers, resulting in:
- Highly successful marketing campaign which secured greater market share over competitors in a volatile market
- Ability to align marketing campaign objectives with specifically targeted audience to focus efforts on conversion and ROI
The Value
80%
KPIs
