Segmenting for Success

What is Segmentation?

Segmenting the market involves identifying and quantifying groups of individuals (decision makers) that are similar to one another but different from individuals in another group. Differences can be based on attitudes, behaviours, benefits sought etc. and can be linked to other factors such as demographics, location, brand use, communications methods etc.

Why Segment?

With segmentation of the market you can take advantage of differences between groups for target marketing and sales activities to focus on certain product/service attributes and develop your business. It lets you:

• understand similarities between groups in order to develop overarching strategies that can be used across segments.
• understand unique differences between segments to target your resources.
• quantify the size of each segment to assist in making decisions about which segments to target and the appropriate level of resources to use in targeting.
• predict segment membership post-hoc for all individuals based on a sub-set of simple questions used in all prospect and customer contact activities
• maximize ROI from sales, marketing and product/service design activities.

Example: Product, Service and Offer Attributes

Companies spend an enormous amount of money on:

• Brand Strategies (positioning, advertising, promotion, direct marketing, social media, sales activities etc.)
• Distribution Strategies (dealer and distributor activities and programs).
• Company and product line strategies (product line programs, company positioning, PR etc.)

Problem:

• Prospects don’t react the same way to each strategy
• Resources are wasted on customers that would have bought anyway

Solution: Target strategies to selected segments that maximize sales or profitability.

Segmentation Methods

Traditional Method: Rating scales in questionnaire passed through a typical cluster analysis such as k-means. Problem: Scale use bias and non-optimal cluster solutions, less predictable.

iFusion Methods: Rating scale standardized and passed through numerous cluster techniques and optimal cluster technique chosen OR choice questions passed through Latent Class Analysis and/or Choice Based Conjoint/Hierarchical Bayes (CBC/HB) to eliminate scale bias.