Brand Mapping

How To Run Brand Mapping

Tip:
It's easy to get up to speed with MM4XL in a matter of seconds:

1. Select the MM4XL tool you want to work with from the toolbar.
2. Click the Example button on the form. This opens an example sheet with data for the tool you are using.
3. With your mouse, select the data sets as explained in the example sheet, then click the OK button to print the report.

It is easy to run Brand Mapping. Choose Brand Mapping under Strategic Tools in the menu bar. Alternatively, hit the third button from the left on the floating toolbar.

In the first window, select one of the three option buttons corresponding to the analysis you want to run. If you choose Contingency table, Brand Mapping jumps to its final window, the control panel. In other cases an intermediate window appears before the control panel.

The Supplementary data option requires a number in each of the two drop-down boxes, placed in the lower right corner. The input corresponds to the last columns and rows of the input table. In cases where there are no supplementary points, simply enter zero. If you input a zero in both boxes, Brand Mapping runs a plain contingency table analysis. If you are not familiar with the concept of supplementary points (also called passive) read the chapter How to interpret Brand Mapping.

The Missing data intermediate window appears when option 3 is chosen. This window is simply a short explanation and no user input is required here.

The last window, or control panel, tells Brand Mapping where to get the data and how to fine-tune both output and the analysis itself.

The picture in the lower right corner of the control panel shows an example of the output report. Make an accurate selection in this window and your output will be much easier to read and interpret. There are four framed areas for user input.

1^st - The data.

In the first frame, The data, place the cursor in the Data table edit box and select with your mouse the region where the data is stored. The first row and first column of this area must be text labels, so blank cells and numerical values are not allowed. In the Output range edit box, select only the start cell, where you want Brand Mapping to begin printing the output report. Both of these fields must be completed with a range, in order for the analysis to run. Change text in the third edit box to assign a title to your map, this defaults to Brand Mapping.

2^nd- The bubble size.

The second frame option The bubble size determines which values are used to compute the bubble diameter. By default, Brand Mapping uses the mass value. Alternatively, the Third axis coordinates or a custom address may be chosen.

The Third axis option uses coordinate values when setting the bubble diameters, useful for displaying broader data variance on a single map. However, this visualization may be biased by negative values, as the bubble size is computed treating the input values as absolute values.

The Select an address option allows users to set the bubble diameter using a custom value stored on the sheet. This is a very useful option to enrich the map with external information that may make the interpretation easier.

The values are stored on the sheet as in the above example. The column values are stored first, and than the row values. In our example, the range B1:B9 is entered in the edit box, either manually or using the mouse, and Brand Mapping uses these values to set the bubble diameters.

Note:
When the number of supplementary rows is larger than the number of active rows, the order of the labels must be reversed, so the custom row labels come first and the column labels next.

3^rd - The output.

By default, Brand Mapping draws a map, and prints contributions and squared cosines to factorial axes. Click the respective check box to deactivate one or both options. Coordinates and mass values are also printed by default and cannot be deactivated.

When working with large data sets, it may help to reduce the number of axes printed in the output report. Use the option Max number of axes (0 = all) to define how many axes you want to show. By default Brand Mapping limits the number to 4, as suggested by Benzcri (1992). Type a zero (0) to display all axes.

4^th - The last step.

This last quadrant is only active when the option Plain contingency table has been selected in the first window. By default, Brand Mapping uses the Compute unequal mass option, which runs the analysis using the raw data, exactly as input by the user. The option Set column total = 100% transforms the original values as percentages, so that the sum of each column (profile) is 100%. The same happens to rows if the option Set row total = 100% is chosen. This is a useful option to reduce the weight in orientating the map of columns and rows, which account for a large portion of the whole data variance. Use this option, for example, to reduce the effect of leaders with large market shares, or those with large market segments (see the chapter How to interpret Brand Mapping for more details).

The What's this buttons provide short descriptions relating to single functions.

The data input

Brand Mapping is very flexible and can handle almost any type of data table. However, the GIGO theorem holds: Garbage In, Garbage Out. Do not forget it!

The input to Brand Mapping is a rectangular table of figures and labels, a contingency table, similar to the one below.

Each column and row of the input table is a unique profile and Brand Mapping displays it on the map as a bubble. Using the table above, the map output would have eight column bubbles and ten row bubbles.

Most often in marketing, the figures are preference data gathered from surveys, both ad-hoc or syndicated. We recommend extensive use of Brand Mapping with the latter studies, in order to get the most benefit from these expensive data, which often lays unused on a dusty shelf in a dark room. For example, the pharmaceutical industry has a lot of data available about usage of drugs. Arrange this data in a table with the product name as the column label, the name of the treated disease as a row label, and fill out each cell with the number of prescriptions (preference) for each drug.

Also percentages, ratings, e.g. on a scale 1-10, (Greenacre, 1984), and binary data, 1 or 0, (Hoffmann and Franke, 1986) can be used as input to Brand Mapping.

Tip:
Do not be afraid to use very large data tables. Brand Mapping is actually a segmentation tool and is of great help to find out latent structures in complex data sets. When working with large tables, one may need to repeat the analysis several times, focusing more and more on the truly relevant portion of data.

Brand Mapping's advanced features allow users to work with even more complex data tables, such as a contingency matrix with both active and passive points. A passive point is a bubble displayed on the map, without actually contributing to the orientation of the map itself. It can be seen as a "what if" point. This useful feature makes Brand Mapping extremely flexible and well suited to the analysis of dynamic data sets. In the output report, passive data items are printed with bold labels.

A dynamic data set, also longitudinal, is made up of cross-sectional data, both longitudinal and not. In other words, take a survey today and gather the data, then repeat the same survey a while later and gather a second set of data. This is what we call dynamic data: a set of two or more observations from the same universe. The table below, for example, is made of two different surveys. The data in the gray area was gathered in 2001, and data in the lower yellow one in 2002.

Section A, the gray area (as usual with labels) can be input to the Plain contingency table. Sections A and C conjointly may be input to the Supplementary data option with four passive rows. Sections A and B conjointly may also be input to the Supplementary data option with three passive columns. Finally, all four quadrants together may be used with the Estimate missing data option (using the last option, quadrant B must have blank cells). The four-quadrant matrix is useful to handle the entrance in the market of new competitors, and it is a feature unique to Brand Mapping. These are the three basic input tables to Brand Mapping.

The main question one should answer before starting Brand Mapping is: Should my map only display active points, or are there useful passive points (also called supplementary points)? In other words, do I choose the first option in the opening windows or do I choose one of the other two options?