Beginnings urban essence unattached multi cultures

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Rugged Rural Style Latino Nuevo Struggling City Centers College Town Communities Metro Beginnings Urban Essence Unattached Multi-cultures Academic Influences African-American Neighborhoods Urban Diversity New Generation Activities Getting By Varying Lifestyles Military Family Life Major University Towns Gray Perspectives Page 2 Selected behaviors for Mosaic USA target group ranked by potential Site selection Mosaic USA can help pinpoint the best sites and trade areas for maximizing sales potential. Site analysis for Mosaic USA target group by census tract within 20-minute drive time How is Mosaic USA built? Over the last 20 years, Experian has built more than 40 segmentation services worldwide. This gives us valuable insight into the best sources of data and methodologies to build truly innovative segmentation. In building Mosaic USA, the following approaches were taken: • Identification of the most appropriate data sources as inputs • A sophisticated proprietary approach to clustering, unique to Experian • Extensive analysis to assist in validation and interpretation of the segmentation Data components quantitative data More than 300 data variables have been used to build Mosaic USA, including more than 70 household-level elements from Experian’s INSOURCE Database. These have been selected as inputs to the classification on the basis of their volume, quality, consistency and sustainability. To be input into the classification, the data must meet one or more of the following criteria: SM • Allows identification and description of consumer segments that are not necessarily distinguished solely by the use of census data • Ensures accuracy of the Mosaic USA code by either household or neighborhood • Is updated regularly to ensure change is monitored • Improves discrimination and allows for the identification of a wide range of consumer behaviors Mosaic USA was developed using consumer demographic information sourced from Experian’s wealth of data assets, including INSOURCE, which provides coverage for more than 110 million households and 215 million individuals; demographic estimates and projections from Applied Geographic Services; and the U.S. Census. All of this information is updated regularly and used to replenish our view of the classification each year. Clustering Mosaic USA is designed to identify groupings of consumer behavior for households and neighborhoods. The methodology used is unique to Experian and has been refined during many years of creating classifications using data from different sources and different levels of geography. • Household size The first step is to gather data for all residents and households in the country. This data then is combined with information from other higher levels of geography, including census and postal information. All the input variables go through a selection process, where they are tested for discrimination, robustness and their correlation to other variables. • Employment Once the final list of variables is selected, a set of input weights is applied as part of the clustering process. The result is a list of variables that have differing importance to the clustering methodology, depending on how well they discriminate at differing levels of geography. • Dwelling type This “bottom-up” approach enables us to maximize the effectiveness of each input variable depending on its relative importance to the classification and its ability to discriminate. It allows for the optimization of data and creates a classification that is truly best of breed. • Income • Marital status • Presence of children Socioeconomic • Commute • Education • Language spoken • Industry • Occupation • Social status Property characteristics • Amenities • Housing value • Length of residence • Rent • Year built Location • Population...
View Full Document

This note was uploaded on 02/07/2014 for the course MIS 304 taught by Professor Mejias during the Spring '07 term at University of Arizona- Tucson.

Ask a homework question - tutors are online