Cluster analysis is a statistical procedure that identifies homogeneous groups or clusters of individuals. In marketing research, cluster analysis is often used to determine whether there are distinct groups of customers with different needs, preferences, perceptions, product usage, and purchasing behavior. By understanding how customers differ, management can develop products and/or marketing strategies that are tailored to each group’s individual needs.
K-means analysis is a clustering technique that can be used to create customer segments. This type of cluster analysis uses a procedure in which individuals are assigned and reassigned to "clusters" or "segments" repeatedly until each individual is assigned to a final segment. Each final segment is comprised of individuals who are more similar to other people within that segment than those in other segments. This method implicitly minimizes the diversity within each segment -- and thus, in this case, produces distinct segments with homogenous needs, preferences, etc.
Factor analysis is a statistical technique that is used to identify the structure within a set of variables. By examining the association among variables, factor analytic techniques produce a smaller set of variables or factors that represent the underlying dimensions of the original set of variables. Each factor is not a single, directly measurable entity, but rather a construct that is derived from the measurement of the original set of variables. This technique is often used for the purpose of data reduction -- that is, reducing a large number of variables to a smaller set of factors greatly simplifies the description and understanding of large sets of data.
Consider an example of a data set that includes preference ratings for a large number of specific foods such as bread, apples, blueberry pie, cereal, spare ribs, green beans, watermelon, chocolate mousse, pasta, ham, asparagus, artichoke, and roast beef. A factor analysis of this data set would distill this large amount of information by identifying the dimensions that underlay the individual ratings.
More specifically, dimensions are identified for groups of variables that are highly intercorrelated with each other, but are not highly correlated with variables outside of that group. In this data set, for example, we might expect a fairly high association among preferences for cereal, pasta and bread -- that is, in general, the higher an individual's preference for bread is, the higher we might expect his preference would be for pasta and cereal. However, knowing an individual's preference rating for cereal, pasta, and bread would probably not be useful in predicting that individual's preference rating for any other food on the list. If this were the case, these three variables would be represented by a factor that might be called the "starch" factor. Cereal, bread, and pasta would then be said to "load" on the "starch" factor. Besides the "starch" factor, other factors likely to emerge in this example include "fresh fruits," "deserts," "green vegetables," and "beef."
When a small number of factors account for most of the variance in the original set of data, we can "explain" the original set of variables in terms of a smaller set of factors without losing important information. Simplifying large sets of market research data in this way provides a conceptual clarity that facilitates one's comprehension of the data and its strategic marketing implications.
Neural networks are an alternative to traditional statistical techniques for prediction, classification, segmentation, and time series analysis. A primary advantage of neural networks is that they can find non-linear relationships in the data. They do not depend upon the same assumptions (i.e., multivariate normal distributions, equal variance-covariance matrices, etc.) as conventional techniques. Since neural networks are non-linear, they can find patterns of any form -- linear, logarithmic, exponential, sigmoidal, sinusoidal.
Neural networks rely on a validation dataset to avoid the problem of “overtraining” a model that will not extrapolate well to similar data. They have generalization ability -- they can correctly process data that only broadly resemble the data on which a model is built.
Neural networks “learn” patterns in the data and use an iterative process to create models. They start with randomly generated weights and compare these to known predicted outputs. The weights are then adjusted and compared again. This adjustment process continues until the network produces output sufficiently close to the actual output.
Neural networks can take two forms:
- Supervised networks - Similar to regression analysis or discriminant analysis. The input dataset consists of records with known outcomes (i.e., the targeted respondent purchases product or does not purchase product). The neural net produces predicted output values that it compares with the known outcomes. The trained model can be saved and used to predict purchase behavior of future respondents.
- Unsupervised networks - Similar to cluster analysis or factor analysis. The input dataset does not have known outcomes. The neural net will group input patterns together. Useful in identifying similar segments in a market.
Logistic regression is a technique that identifies variables that are important for distinguishing between two groups of individuals. It should be mentioned that other techniques such as regression and discriminant analysis can also be used to predict a dichotomous dependent variable. However, these techniques depend on certain assumptions (i.e., multivariate normality of the independent variables and equal variance-covariance matrices for the two groups). When these assumptions are violated (which they typically are in studies of this type), the results of the analysis may be less than accurate. Logistic regression does not depend on these assumptions.
Discriminant analysis is a technique that is used to identify variables that are important for distinguishing among groups of individuals. For example, it might be useful to identify those variables that determine whether or not, for a given product, an individual would be most likely to choose to deal with a (1) bank, (2) finance company, or (3) credit union. Variables to be used in such an analysis would depend on your particular interests, but might include demographics, past loan/credit experience, likely reasons for loans, and needs/benefits.
If desired, discriminant analysis can also be used to develop a procedure for predicting group membership (bank vs. finance company vs. credit union) for individuals not included in the analysis. This type of procedure might be useful in selecting potential customers for a targeted marketing campaign.
It provides a graphic representation of the market structure using categorical variables, e.g., brands and product usage. Maps provide a graphic representation of the relationship between two categorical variables, e.g., brand and product usage. The map illustrates:
- Perceived similarity/competitiveness among products in your category (represented by proximity on the perceptual map)
- Strengths and weaknesses of your brand and key competitors
- Market gaps (spaces on the map) representing potential opportunities to provide a product with attributes that no current product is perceived as delivering well
Other techniques include meta-analysis, geo-coding and cluster coding, time-series forecasting models, data-base mining, exploratory data analysis techniques, conjoint and hypothetical choice modeling, network analysis, LISREL models, business outcome modeling, validation and reliability measurement, CRM, employee relationship management, log-linear models, logit, probit, tobit, Chaid, OLS, SES, Chi squared, process and impact evaluations, T.U.R.F., and SWOT.
Retaining Customers and Maximizing Profitability
- These studies developed and refined a Customer Satisfaction Measurement System for B2B customers and consumers. The research provides key information regarding customers' wants, needs and expectations, as well as perceptions of performance within a systematic customer satisfaction framework.
- Service-measurement studies are used to evaluate the kind and quality of customer service interactions in order to position this particular kind of customer service interaction within the broader framework of how best to market customer satisfaction. Client-specific methodologies have been established that update databases with usable information about the quality of customer service rendered, which, in turn, can be used to evaluate performance.
- These competitive positioning studies examined basic attitudes toward the company on a variety of customer satisfaction measures. The basic purpose was to strategically position the company as the preferred provider.
- Given the increasing need to accurately measure corporate image and customer satisfaction, a time-series methodology was created to measure communications and public relations efforts. By taking advantage of specific time measurements, a built-in capability is established to examine seasonal fluctuations in overall corporate image and customer satisfaction, as well as more accurately measure the impact of specific communications efforts.
Differentiating the Marketplace and Target Marketing
- Directly involved with the development and implementation of market research databases to support and direct comprehensive marketing plans
- Scoring databases, lifetime value, opportunity analysis
- Qualified lead programs
Managing Brand and Customer Expectations
- Brand work—awareness, value
- Measuring brand—personality, brand image
- Repositioning brand
- These projects were designed to investigate basic attitudes customers have about the company, as well as the evaluations customers make of proposed programs that might be offered by a company to demonstrate its concern and caring for the needs and problems of its customers. Tracking surveys measuring the effectiveness of communications efforts developed from the research generally followed these studies.
- Developed communications programs designed to enhance overall corporate image. This experience includes the development of programs and messages based upon market segmentation and tracking surveys that measure the impact of the communication program.
- Explored major environmental issues, such as recycling, acid rain, nuclear waste, emergency evacuation zone planning, and hydrocarbon remediation. This experience includes the development of behavioral predictive models to predict participation levels, as well as ongoing tracking of emerging environmental issues.
- PR work
- Often, managers have too much data - not too little. The key is to put together the right type of information.
- Data mining
Scoping the Market and Predicting the Future
These planning studies are designed to generate extensive information on saturation and usage that, in turn, can be used for supply side planning and forecasting. The data generated by these studies have been used by planners and forecasters to accurately depict the present marketplace, which provides successful interventions to maximize participation. Also included is the development of robust statistical models to more precisely monitor and forecast future demand.
Understanding Process and Evaluating Impact
- Process evaluations are used to maximize participation, as well as to minimize the lost opportunities in the market. The objectives of these evaluations were to determine the interrelationships among new decision-makers; to discover at what stage in the decision-making process, to whom, and how a company should direct its efforts; to review what methods are available for consumers or vendors to self-identify themselves as potential program participants; and to evaluate what signals are readily available that are indicative of customer entry into the market. These goals were accomplished using a combination of in-person and telephone interviews.
- Impact evaluations examine the statistically defensible changes in consumer behavior.
- Market tests
- Experimental designs
Delivering Value--Great Employees
- Employee surveys
- Enriched the ability of upper management to communicate corporate goals by establishing objective measurements of employee satisfaction. This experience includes the development of indices to measure overall customer satisfaction, as well as to evaluate employee performance.
New Product Development
These studies, often employing some form of hypothetical choice modeling, are used to design new goods and services to maximize market penetration and profitability.
Often, the most important marketing issue is how and to whom to market goods and services.
- Mystery shopping
- SWOT analysis based on end-users perspectives