Rosella       Machine Intelligence & Data Mining

Customer Profiling

Statistical customer profiling along with demographic, geographic, and psychographic variables is a widely practiced technique in business. For example, it can be used to determine premium rates in insurance industry. Or it can be used to identify top customers who provide highest revenues or to identify customers who may be on risk of losing in customer retention. Better understanding through profiling can lead to better marketing plans. Typically, profiling is performed with the following variables;

  • Demographic profiling variables describe characteristics of customers and include age, gender, race, education, occupation, income, religion, marital status, family size, children, home ownership, socioeconomic status, and so on.
  • Geographic variables include various classification of geographic areas, for example, zip code, state, country, region, climate, population, and other geographical census data. Note that this information can come from national census data.
  • Psychographic profiling variables describe customer's life style, personality, values, attitudes, and so on.
  • Behavioral variables include product usage rate and end, brand royalty, benefit sought, decision making units, ready-to-buy stage, and so on. This information can be extremely useful for marketing purposes.
  • RFM: Recency, frequency and monetary values.
Combinational Factor Analysis and Combinatorial Blowout!

Notice the following two important facts on customer data. This has important implication in selecting your customer profiling methods and techniques!

  • Profiling information can consist of many variables (or dozens of them).
  • Majority of them are categorical variables (or non-numeric variables or nominal variables).

In conventional customer profiling methods, analysts use visualization and statistical reporting tools. These tools can work on only a few variables at a time. When applied to data with many variables like customer information, the numbers of combinations to be examined will grow combinatorially to the numbers of variables. Therefore, thorough systematic accurate analysis of such data by manual means is all but impossible. General practice is to examine only variable combinations what experts think promising. However, intuition can omit important trends and patterns emerging. Customer profiles developed with incomplete analysis can result in wrong business decisions. Better tools are needed for timely thorough systematic analysis!

Hotspot Profiling Analysis

Customer profiling is to characterize features of special customer groups. Hotspot Profiling Analysis can search profiles of special customer groups systematically using Artificial Intelligence techniques. It generates accurate profiles based on search and incremental learning techniques. Hotspot search can be based on various measurement criteria depending on the types of target variables;

  • Categorical variables: probability, Laplace, goodness of fit, entropy, etc.
  • Numerical variables: average, total, harmonic ratio, etc.

The following examples show where hotspot analysis can be used in customer profiling;

  • A company wishes to find out profiles of most valued customers who have bought most in average values and provide most financial contribution to the company.
  • An insurance company wishes to find out all the most (or least) important factors of policy claims. Discussed further below.
  • A department store wishes to find out demographic and geographic profiles of customer groups that responded best in the previous catalog mailing campaigns and wish to use the information to boost overall sales.
  • A utility provider (telco, gas, electricity, etc.) wishes to find out profiles of customers who defected to rival providers and use it to develop customer retention strategies.
  • A market research firm wants to identify all the most and least important factors (or demographic properties) of certain product sales.
  • A healthcare fund wants to know profiles of fraudulent healthcare claims and providers (fraud detection).
  • A health club wishes to know profiles of customers who failed to renew in last 12 months.

Generalized Customer Profiling

Customer profiling often associates this general method. This information provides overall patterns about customers and can play vital role in developing overall marketing strategies.

[Example 1] A company that sells boutique products may develop customer profiles as follows. Intuitively, most customers are females (91%). Majority of customers work as office worker (78%). Note that office workers have more needs to use beauty products! And so on. This information can be used in selecting magazines or TV and Radio programs for advertisement.

Gender=Female 91%
Vocation=Office worker 78%
Education=High school 67%
Age=20s 36%
Age=30s 31%

Focus Group Customer Profiling

Developing profiles of special focused customer groups are much harder than that of general customer profiles, especially when many variables are involved. Hotspot analysis can develop focused customer group profiles very effectively with accuracy. The examples that follow explain how StarPorbe hotspot analysis can be used in customer profiling.

Insurance Industry Examples

Customer profiling is very important in determining premium rates. Typically, insurers collect every information available. However, analyzing thoroughly is not feasible since the number of variables is normally large. The following two examples demonstrate how hotspot analysis can be used in profiling risky insurance policies out of dozens of customer variables;

[Example 2] An insurance company keeps health insurance or life insurance records in its database: gender, age, education, smoking, drinking, sun activity, height, weight (=obesity level), claim payment, etc., as well as contact information. The company wishes to know which health insurance groups are at the highest risk, i.e., have the highest claim ratio. The following is a possible output of hotspot analysis;

Insurance customer profiling.
Credit/Finance Industry Examples

Customer profiling is very important in reducing risk of loans, credits and finances. Credit always poses risk of being defaulted. Profiles of risky credits and loans can be very useful in developing risk management plans.

[Example 3] A financing firm (or bank) keeps loan records on motor vehicle purchase in its database including default information: gender, age, education, occupation, income; vehicle type, manufacturer, model, year make, price, loan amount, default, default amount, etc. The firm wishes to know which types of loans for motor vehicle purchases are at the highest risk, i.e., highest default ratio by probability;

Finance customer profiling.

For more information, please read Hotspot Analysis, Insurance Risk Analysis and Credit Risk Analysis.

For information about customer profiling software, please read Data Mining Software. Software download is available from the page.