our approach to predictive analytics Simplifying the data mining process and creating solutions

Our analytic services are scalable. We tailor analytics to solve your problems within your budget! We offer a variety of custom predictive analytic services to help you gain customer and market insights, and identify the best opportunities and channels to engage with consumers.

Our core strengths in predictive analytics

CUSTOMER-AQUISITION-ICON-01

Customer Acquisition

SALES-FORECAST-ICON-02

Sales Forecast

TRADE-AREA-ICON-03

Trade Area Analysis

CUSTOMER-RETENTION-ICON-04

Customer Retention

TERRITORY-MAPPING-05

Fraud Detection

UP-SELL-ICON-06

Territory Mapping

RISK-MODELLING-ICON-07

Up-Sell & Cross-Sell
Campaigns

RISK-MODELLING-ICON-08

Risk Modeling & Management

THEMATIC-MAPPING-ICON-09

Thematic & Heat Mapping

SERVICE-BUNDLING-ICON-10

Service Bundling & Market Basket Analysis

CUSTOMER-SEGMENTATION-ICON-11

Customer Segmentation & Profile Analysis

In a consumer society, we are driven by the behaviour of mainstream consumers, but marketers want to focus on the leaders or tastemakers. If the leaders appreciate and value a product or service, their behaviours and attitudes will carry a message to different segments of the broader population. Identifying segments with most value or potential and growing them are critical for sustainable business growth. We apply both Machine Learning and traditional statistical approaches to optimize the segmentation process.

BALLOON-MAN

CUSTOMER SEGMENTATION

Manifold’s Customer Segmentation is an adaptive clustering technique, based on a combination of comprehensive and precise demographic, expenditure and consumer behaviour data products with your customer data. It enables you to identify, retain and grow the most valuable customer segments and tap into segments with highest potentials for your products and services.

Customer Segmentation is a process to divide customers into mutually exclusive groups so that:

Customers within a group are as similar as possible;

Customers in other groups area as different as possible.

Differentiating customers’ needs and values provides you with the ability to:

Prioritize your efforts and gain the most with your valuable customers.

Tailor your company’s products and services using your customers’ lifestyle clusters.

Leverage the power of one-to-one marketing and the coverage of mass marketing by targeting customer segments with tailored tactics.

CUSTOMER SEGMENTATION PROCESS

1

Cleanse and consolidate customer information and transactional data to the customer level.

ARROW1
3

Perform factor analysis or Multiple Correspondence Analysis (MCA) to reduce dimension of database to handful driving factors.

ARROW2
ARROW3
5

Validate, refine and profile clusters.

ARROW4
2

Attach relevant 6-digit postal code level geo-demographic, spending, lifestyle and behaviour data.

4

Apply an adaptive K-means algorithm for numeric data and a divisive hierarchical algorithm for categorical data to group customers into natural clusters.

1
2
3
4
5

Cleanse and consolidate customer information and transactional data to the customer level.

Attach relevant 6-digit postal code level geo-demographic, spending, lifestyle and behaviour data.

Perform factor analysis or Multiple Correspondence Analysis (MCA) to reduce dimension of database to handful driving factors.

Apply an adaptive K-means algorithm for numeric data and a divisive hierarchical algorithm for categorical data to group customers into natural clusters.

Validate, refine and profile clusters.