The cluster analysis defines how many MOSAIC "clusters" or Types there should be and what the character of each cluster should look like. Once defined, the 38 Types were described by the data used in the build. These descriptions were further enhanced by the use of Roy Morgan Research data on the attitudes and preference of consumers and Cadmus data on motor vehicle preferences.

The 38 MOSAIC Types have also been assigned to 11 MOSAIC Groups to enable analysis or selection at a broader classification level.

The result is a classification that can clearly identify households or small geographic units that have similar characteristics and marketing attributes. And since every address in New Zealand can be classified, MOSAIC has the advantage of coverage, flexibility and speed, as well as modelling and mapping capabilities.

Applications for MOSAIC

MOSAIC can be used to classify customer databases (or Roy Morgan Single Source survey data), profile and rank catchment areas, map target areas, calculate market potential, determine product mix for customers and target key areas or household addresses. Targeting may take the form of prospect lists for addressed mail, unaddressed letterbox distribution, point of sale or outdoor promotion. Above the line or broadcast media can also be targeted by cross tabulating against Roy Morgan's media research data.

Accessing MOSAIC

MOSAIC analysis can be performed either by the bureau team at PMP Micromarketing or by your own in-house marketing team by purchasing a MOSAIC Directory license and, if required, analysis and data enrichment software such as Micromarketer and Kaleidoscope. Using a Directory you can append MOSAIC to your own customer name and address database and add value to your own in-house modelling and segmentation projects.

For each MOSAIC Type we have included:

• The MOSAIC Type number and name including the percentage of total New Zealand households.

• A summary description of the sort of people most likely to be found in the MOSAIC Type. The description has been based on the data used in the build of MOSAIC, as well as information on consumer attitudes and preferences from Roy Morgan Research survey data.

• A photograph that shows a typical example of a house or neighbourhood.

• A map that shows the distribution for the MOSAIC Type for one area in New Zealand. (Dots = houses in region).

• Typical Localities. Districts and Suburbs with a high concentration of the MOSAIC Type are highlighted for selected regions throughout New Zealand.

• The bar charts compare the MOSAIC Type against the national average on key demographic variables taken from the 2001 Census and other data sources used in the building of MOSAIC.

• The information contained in the bar charts has been indexed to a New Zealand base of 100. In other words, where a MOSAIC Type has a score of 50 on a particular variable (such as people aged over 65) the average percentage of adults in that MOSAIC Type who are aged over 65 is only 50% of the national average. A score of 100 shows that the proportion of those aged over 65 is equal to the national average; a score of 200 indicates a level exactly twice the national average, and so on.