Data Mining? The Foundational Elements of Data

¶ … data mining? The foundational elements of data mining are multidisciplinary in nature, encompassing analytics, computer science, database systems integration and management, statistics and artificial intelligence. Often these technologies are used to create a single system of record used for analysis and advanced queries by the enterprises who build them. Data mining is often included in business intelligence (BI) suites and the analytics layer of an enterprise-wide computing system, as each application needs to gain access to the metrics and key performance indicators (KPIs) (Peacock, 1998).

The use of data mining has become more pervasive in marketing, sales and service as organizations strive to gain insights from the terabytes of data they have accumulated over years and in some cases decades of operation. Data mining can provide marketers with greater insights into the preferences, needs and wants of customers, in addition to potential new product or service ideas based on a careful analysis of the accumulated data on customer bases (Koh, Kin, 2002).

Data mining also creates a highly effective platform for completing simulations of potential pricing and service strategies, and can serve as a very effective source of product line enhancement ideas (Peacock, 1998). The potential also exists to use data mining as a means to create a highly effective segmentation model based on previous customer purchases and the patterns of services purchased as well. Data mining's greatest potential however is being seen in the precise aligning of pricing and the potential to generate greater profits over time. 2.

What is the Value Map? The foundational elements of a value map is the charting of price and benefits of a given product. This two-dimensional grid is often used for determining the price-quality relationship of a given series of products or services. Advanced forms of value maps can also provide insights into the price elasticity of products, and how a per unit change in a given pricing schedule will impact demand over the long-term.

Marketers also rely on the value map to determine the perceived quality of their products relative to competitors and also substitutes. The market share changes that can occur over time with differences in perceived cost and quality are also tracked in this two-dimensional model. There is no single, optimal point for a product to occupy in the matrix as differentiation and market position varies significantly by product, its relative competitive position and value delivered.

For more inelastically-priced goods having a higher perceived price and higher perceived quality, and therefore continually stay at a price premium that ensures long-term profitability. As products encounter greater competition the pressure to move into an economic-driven positioning strategy of mid-to-low cost with perceived quality at a lower point begins to force commoditization into markets. These dynamics occur slowly in markets with highly elastic product demand and very quickly in highly inelastic markets.

The greater the elasticity of a product in a given market, the higher the probability of being able to resist the commoditization that occurs in markets comprised of products that lack differentiation over time. 3. What are the approaches to market segmentation? There are four major types of market segmentation used today. The most common are demographic and geographic with behavioral and psychographic being most often used for consumer products (Tuma, Decker, Scholz, 2011).

Geographic segmentation is often defined by city, state or region of a country, and often is cross-referenced by income levels and other demographic data. It is common to find marketers segment their markets geographically when demand varies by location of the country. Sales of antifreeze during the winter months is an example of how geographic segmentation would be used for selling this product, with the primary focus being the colder climates in the Untied States and Europe.

One of the most prevalently used segmentation criterion that consumer products companies rely on is demographic variables. These include age, gender, family size, income, occupation, education, religion, race and nationality (Craft, 2001). Given the increased availability of analytics for gaining insights into customers' preferences, psychographically-derived segmentation is increasingly being used. These segmentation criterion include social class, lifestyle type, personality type and other factors that underscore the way customer's define themselves as parts of a group (Tuma, Decker, Scholz, 2011).

In addition to all of these segmentation criterion, marketers are also using analytics to determine behavioral patterns of product usage, brand loyalty, type of user, brand preference by age and income segment, and the preferred channels to learn about new products and services as well. Data mining techniques in banking are being used for defining behavioral segmentation by evaluating and finding trends in banking activity over time (Koh, Kin, 2002).

This is providing banks with greater insights into how best to segment their customer's and also which services to offer to increase sales. 4. When do you need to do market research and when you don't need to do market research? When a potential business strategy is being slowed down by several unknown factors about a given customer segment, market, or potential product acceptance level, market research needs to be done to better understand the dynamics of the market of interest.

Market research is also necessary for mitigating risks of new ventures including the offering entirely new products or services,.