brand perceptual map

graphical mapping

a solution to the data clutter

principal component analysis (PCA)

A mathematical transformation to convert a set of high-dimensional data (possibly correlated variables) into a set of linearly uncorrelated principal components

principal component analysis objective

PCA reduces larger number of variables to a smaller number of dimensions (dimension

reduction). So, there is no guarantee that the dimensions are interpretable in case of PCA

unsupervised learning method pca

looks into the greatest sources of variation of data

pca forms

orthogonal linear combinations of the original predictors such that

• the first component accounts from the largest variance in the data;

• the second component accounts for the second largest variance in the data

• can be done by matrix decomposition of a data covariance (correlation) matrix

• can identify these with eigenvalues and eigenvectors of the covariance matrix

assessing competition

firms should assess competitors with respect to objective attributes, as well as subjective perceptions from the voice/minds of consumers

firms should monitor

• share of market(objective) competitor’s market share - sales data

• top of mind awareness (subjective): “name the first company that comes to mind in this industry"

• share of mind (subjective)- “name the company from whom you would prefer to buy the product

typical data sets

competitor x attribute matrix

• customers comparative ratings in attributes across competitors

perceptual map is a

visual representation of how target customers view the competing alternatives in a Euclidean space

perceptual map characteristics

• The pair-wise distances between product alternatives directly indicate how close or far apart the products are in the minds of customers

• The axes of the map are a special set of vectors suggesting the underlying dimensions

(e.g., Principal Component1 or 2) that best characterize how customers differentiate between alternatives

car example perceptual map

From this perceptual map: 1. we can see how each car brand is

comparatively positioned in peoples’ mind.

2. This is helpful for developing marketing strategy for differentiating

from other competing car brands

brand positioning map PCA

location coordinates for each brands

directions for each attribute vector

differentation

the creation of tangible or intangible differences on one or two key dimensions between a focal product and its main competitors

positioning

the set of strategies that firms develop and implement to ensure that the differences occupy a distinct an important position in the minds of consumers

good position strategy requires

• An understanding of the dimensions along which the consumer perceives the product

• identifying the brand position compared with other competitors brands

• Considering a firm’s strengths and available resources, the firm can find the optimal positioning location and develop corresponding strategy.

• knowing how competitors brands are perceived on these dimensions

multidimensional scaling mds

• refers to class of mathematical procedures (algorithm) for representing perceptions and preferences of respondents spatially (i.e., in two-dimensional map).

• This is basically projecting high dimensional data to two dimensional space using “distances” between observations – identifying coordinates in 2-dimensional space...

• Note, we are interested in “Perceptual” and not necessarily Engineering attributes – focusing consumer’s perceptions..

• Data type examples:

Similarity

Attributes

data types/collection for MDS

• can collect primary or secondary sources

1 .Collect pairwise (dis)similarities of competitive brands in a designated product

class from consumers.

2. Collect ratings for perceptions or beliefs of how much each brand has of

specific attributes from consumers – then we can compute distances.

3. Record objective attribute information from secondary sources (e.g., package

labels).

Then, construct brand maps via Multidimensional Scaling (MDS)

description of mds

• Input matrix of dissimilarities (e.g., distance) between pairs of items and outputs a coordinate matrix

• Computing distances between brands

• Searching coordinates in 2 dimensional-map to minimize gap/differences between observed distances and estimated distances in MDS space