rally means dividing the sample into a small number of exclusive clusters. This is almost always done by applying some form of statistical cluster analysis. The analyst will typically try several different cluster "solutions." While there are objective statistical measures of the quality of a cluster solution, they are no more important than informed managerial judgment. Researchers oftenl offer two or three possibilities to discuss.

The "right" number of segments is subjective, but in practice companies tend to settle on four to eight segments. In a diverse, complex, and lucrative category more might be called for.

Agreeing on the number of segments, as well as the labels that will be attached to them, always requires careful study of their full profiles. While the segments may be defined with one class of basis variables (e.g. attitudes about fashion and shopping), you always need to describe the segments in terms of other variables (e.g. demographics, spending, media preferences, preferred brands) that were not part of the definition of the segments.

Application

An important deliverable from most segmentation research is a tool for classifying other consumers into the segments. This could be a formula where by asking a small number of questions you can place a "new" consumer in one of the segments. It c

What is meant by market segmentation?

Market segments consist of groups of people or organizations that are similar in terms of how they respond to a particular marketing mix or in other ways that are meaningful for marketing planning purposes.

The entire field of market segmentation is based on this idea, that consumers have differing needs. They will find value in different products. They will respond differently to marketing communications. They will gravitate toward different price points.

Because of this, businesses operate more efficiently if they can act on those differences. A business that tries to offer a single undifferentiated product with generic marketing support will always lose out to more nimble competitors who have specific targets and know how to serve them.

We often work for health plans. In the Medicare Part D world for instance, health plans are struggling to make sense out of new regulations while meeting very tight deadlines. But today’s difficulties are going to give way to a market where segmentation will be of great benefit because mass marketing will not work.

Which Segment?

It is obvious that all competitors cannot target the same segment and succeed. A common mistake is to assume that you must focus on the "heavy buyers" who, in reality, are often not the most profitable group. For example, a consulting firm may decide to forego targeting Fortune 500™ firms that are fiercely contested by the Big Four accounting/consulting firms and instead pursue middle market firms. "Sub prime" lenders and credit card companies know exactly who their prospects are, and do not bother advertising in Money magazine or funding public television shows. Part D players may find market niches among seniors with, for example, private Medigap policies and specific interests or needs that the plan can address.

For health plans, market segmentation is not common. Part of this is due to health plan cultures developed to serve large employer-based health plans and retirees. But Medicare Part D is an unprecedented public-private effort to create a consumer market. And Medicare's Part D program is just one element of a far reaching effort to use private health plans to control Medicare expenses. We believe the market understanding developed through segmentation will serve in a variety of ways.

Identifying and targeting one or a few segments instead of the entire market allows the firm to use its resources more effectively. Market segmentation means products and messages more aligned with the needs of selected consumers.

Researching Segments

Some type of systematic research is required as a foundation for market segmentation. Most often, this research entails a sample survey which should be done by research experts.

Segmentation research addresses a set of fundamental questions: 1. How many segments are there? 2. How large is each segment? 3. How do you define the segments? 4. How do you describe the segments?

Segments are typically defined by dimensions such as:

• Behavior—such as the frequency of purchasing, total spending, or the mixture of stores shopped (switching, investigating, comparing, delegating decisions), or

• Characteristics—such as geodemographics for consumers or SIC codes for businesses (health conditions, subsidy status, current coverage status), or

• Attitudes—such as financial sophistication, fashion orientation, or disposition to adopt new technologies (interest, optimism).

The variables that are used to define the segments are called the "basis." The basis is chosen in light of how the segmentation will be applied. Direct marketers, for instance, carefully dissect their databases in terms of purchase behavior. Other segments are based on consumers’ motivations and concerns. Marketer choose this because motivations and concerns are what cause consumers to act on an offer.

We don’t think that there is only one "correct" way of doing segmentation research, nor that there is a single set of segments waiting to be "discovered." While the analysis needs to employ objective statistical methods, any number of judgments, preferences and practical considerations go into the design of the process and the final determination of the segment structure. The same dataset can yield different segments.

Market segmentation spotlights opportunities for health plans to improve marketing and sales, staff training, customer service, new enrollee intake, business planning, and public relations. It will be used by health plans that want to assure they emerge from this period of dramatic Medicare market change as a coverage provider of relevance and viability. Part of this means achieving sales and marketing goals. Part of this means establishing a market position where consumers think of you as you’d like them to think of you.

Compared to other forms of marketing research, the sample sizes for segmentation studies tend to be large. The simple reason is that you need to be able to profile multiple sub-samples (the segments) with a high degree of precision, rather than just the total sample. It is unusual to use fewer than 500 and not unusual to use 2,000-3,000.

Analysis

The most complex and opaque part of segmentation research is the derivation of the segments—that literally means dividing the sample into a small number of exclusive clusters. This is almost always done by applying some form of statistical cluster analysis. The analyst will typically try several different cluster "solutions." While there are objective statistical measures of the quality of a cluster solution, they are no more important than informed managerial judgment. Researchers oftenl offer two or three possibilities to discuss.

The "right" number of segments is subjective, but in practice companies tend to settle on four to eight segments. In a diverse, complex, and lucrative category more might be called for.

Agreeing on the number of segments, as well as the labels that will be attached to them, always requires careful study of their full profiles. While the segments may be defined with one class of basis variables (e.g. attitudes about fashion and shopping), you always need to describe the segments in terms of other variables (e.g. demographics, spending, media preferences, preferred brands) that were not part of the definition of the segments.

Application

An important deliverable from most segmentation research is a tool for classifying other consumers into the segments. This could be a formula where by asking a small number of questions you can place a "new" consumer in one of the segments. It co

rofitable group. For example, a consulting firm may decide to forego targeting Fortune 500™ firms that are fiercely contested by the Big Four accounting/consulting firms and instead pursue middle market firms. "Sub prime" lenders and credit card companies know exactly who their prospects are, and do not bother advertising in Money magazine or funding public television shows. Part D players may find market niches among seniors with, for example, private Medigap policies and specific interests or needs that the plan can address.

For health plans, market segmentation is not common. Part of this is due to health plan cultures developed to serve large employer-based health plans and retirees. But Medicare Part D is an unprecedented public-private effort to create a consumer market. And Medicare's Part D program is just one element of a far reaching effort to use private health plans to control Medicare expenses. We believe the market understanding developed through segmentation will serve in a variety of ways.

Identifying and targeting one or a few segments instead of the entire market allows the firm to use its resources more effectively. Market segmentation means products and messages more aligned with the needs of selected consumers.

Researching Segments

Some type of systematic research is required as a foundation for market segmentation. Most often, this research entails a sample survey which should be done by research experts.

Segmentation research addresses a set of fundamental questions: 1. How many segments are there? 2. How large is each segment? 3. How do you define the segments? 4. How do you describe the segments?

Segments are typically defined by dimensions such as:

• Behavior—such as the frequency of purchasing, total spending, or the mixture of stores shopped (switching, investigating, comparing, delegating decisions), or

• Characteristics—such as geodemographics for consumers or SIC codes for businesses (health conditions, subsidy status, current coverage status), or

• Attitudes—such as financial sophistication, fashion orientation, or disposition to adopt new technologies (interest, optimism).

The variables that are used to define the segments are called the "basis." The basis is chosen in light of how the segmentation will be applied. Direct marketers, for instance, carefully dissect their databases in terms of purchase behavior. Other segments are based on consumers’ motivations and concerns. Marketer choose this because motivations and concerns are what cause consumers to act on an offer.

We don’t think that there is only one "correct" way of doing segmentation research, nor that there is a single set of segments waiting to be "discovered." While the analysis needs to employ objective statistical methods, any number of judgments, preferences and practical considerations go into the design of the process and the final determination of the segment structure. The same dataset can yield different segments.

Market segmentation spotlights opportunities for health plans to improve marketing and sales, staff training, customer service, new enrollee intake, business planning, and public relations. It will be used by health plans that want to assure they emerge from this period of dramatic Medicare market change as a coverage provider of relevance and viability. Part of this means achieving sales and marketing goals. Part of this means establishing a market position where consumers think of you as you’d like them to think of you.

Compared to other forms of marketing research, the sample sizes for segmentation studies tend to be large. The simple reason is that you need to be able to profile multiple sub-samples (the segments) with a high degree of precision, rather than just the total sample. It is unusual to use fewer than 500 and not unusual to use 2,000-3,000.

Analysis

The most complex and opaque part of segmentation research is the derivation of the segments—that literally means dividing the sample into a small number of exclusive clusters. This is almost always done by applying some form of statistical cluster analysis. The analyst will typically try several different cluster "solutions." While there are objective statistical measures of the quality of a cluster solution, they are no more important than informed managerial judgment. Researchers oftenl offer two or three possibilities to discuss.

The "right" number of segments is subjective, but in practice companies tend to settle on four to eight segments. In a diverse, complex, and lucrative category more might be called for.

Agreeing on the number of segments, as well as the labels that will be attached to them, always requires careful study of their full profiles. While the segments may be defined with one class of basis variables (e.g. attitudes about fashion and shopping), you always need to describe the segments in terms of other variables (e.g. demographics, spending, media preferences, preferred brands) that were not part of the definition of the segments.

Application

An important deliverable from most segmentation research is a tool for classifying other consumers into the segments. This could be a formula where by asking a small number of questions you can place a "new" consumer in one of the segments. It c

a foundation for market segmentation. Most often, this research entails a sample survey which should be done by research experts.

Segmentation research addresses a set of fundamental questions: 1. How many segments are there? 2. How large is each segment? 3. How do you define the segments? 4. How do you describe the segments?

Segments are typically defined by dimensions such as:

• Behavior—such as the frequency of purchasing, total spending, or the mixture of stores shopped (switching, investigating, comparing, delegating decisions), or

• Characteristics—such as geodemographics for consumers or SIC codes for businesses (health conditions, subsidy status, current coverage status), or

• Attitudes—such as financial sophistication, fashion orientation, or disposition to adopt new technologies (interest, optimism).

The variables that are used to define the segments are called the "basis." The basis is chosen in light of how the segmentation will be applied. Direct marketers, for instance, carefully dissect their databases in terms of purchase behavior. Other segments are based on consumers’ motivations and concerns. Marketer choose this because motivations and concerns are what cause consumers to act on an offer.

We don’t think that there is only one "correct" way of doing segmentation research, nor that there is a single set of segments waiting to be "discovered." While the analysis needs to employ objective statistical methods, any number of judgments, preferences and practical considerations go into the design of the process and the final determination of the segment structure. The same dataset can yield different segments.

Market segmentation spotlights opportunities for health plans to improve marketing and sales, staff training, customer service, new enrollee intake, business planning, and public relations. It will be used by health plans that want to assure they emerge from this period of dramatic Medicare market change as a coverage provider of relevance and viability. Part of this means achieving sales and marketing goals. Part of this means establishing a market position where consumers think of you as you’d like them to think of you.

Compared to other forms of marketing research, the sample sizes for segmentation studies tend to be large. The simple reason is that you need to be able to profile multiple sub-samples (the segments) with a high degree of precision, rather than just the total sample. It is unusual to use fewer than 500 and not unusual to use 2,000-3,000.

Analysis

The most complex and opaque part of segmentation research is the derivation of the segments—that literally means dividing the sample into a small number of exclusive clusters. This is almost always done by applying some form of statistical cluster analysis. The analyst will typically try several different cluster "solutions." While there are objective statistical measures of the quality of a cluster solution, they are no more important than informed managerial judgment. Researchers oftenl offer two or three possibilities to discuss.

The "right" number of segments is subjective, but in practice companies tend to settle on four to eight segments. In a diverse, complex, and lucrative category more might be called for.

Agreeing on the number of segments, as well as the labels that will be attached to them, always requires careful study of their full profiles. While the segments may be defined with one class of basis variables (e.g. attitudes about fashion and shopping), you always need to describe the segments in terms of other variables (e.g. demographics, spending, media preferences, preferred brands) that were not part of the definition of the segments.

Application

An important deliverable from most segmentation research is a tool for classifying other consumers into the segments. This could be a formula where by asking a small number of questions you can place a "new" consumer in one of the segments. It c

esearch, nor that there is a single set of segments waiting to be "discovered." While the analysis needs to employ objective statistical methods, any number of judgments, preferences and practical considerations go into the design of the process and the final determination of the segment structure. The same dataset can yield different segments.

Market segmentation spotlights opportunities for health plans to improve marketing and sales, staff training, customer service, new enrollee intake, business planning, and public relations. It will be used by health plans that want to assure they emerge from this period of dramatic Medicare market change as a coverage provider of relevance and viability. Part of this means achieving sales and marketing goals. Part of this means establishing a market position where consumers think of you as you’d like them to think of you.

Compared to other forms of marketing research, the sample sizes for segmentation studies tend to be large. The simple reason is that you need to be able to profile multiple sub-samples (the segments) with a high degree of precision, rather than just the total sample. It is unusual to use fewer than 500 and not unusual to use 2,000-3,000.

Analysis

The most complex and opaque part of segmentation research is the derivation of the segments—that literally means dividing the sample into a small number of exclusive clusters. This is almost always done by applying some form of statistical cluster analysis. The analyst will typically try several different cluster "solutions." While there are objective statistical measures of the quality of a cluster solution, they are no more important than informed managerial judgment. Researchers oftenl offer two or three possibilities to discuss.

The "right" number of segments is subjective, but in practice companies tend to settle on four to eight segments. In a diverse, complex, and lucrative category more might be called for.

Agreeing on the number of segments, as well as the labels that will be attached to them, always requires careful study of their full profiles. While the segments may be defined with one class of basis variables (e.g. attitudes about fashion and shopping), you always need to describe the segments in terms of other variables (e.g. demographics, spending, media preferences, preferred brands) that were not part of the definition of the segments.

Application

An important deliverable from most segmentation research is a tool for classifying other consumers into the segments. This could be a formula where by asking a small number of questions you can place a "new" consumer in one of the segments. It c

rally means dividing the sample into a small number of exclusive clusters. This is almost always done by applying some form of statistical cluster analysis. The analyst will typically try several different cluster "solutions." While there are objective statistical measures of the quality of a cluster solution, they are no more important than informed managerial judgment. Researchers oftenl offer two or three possibilities to discuss.

The "right" number of segments is subjective, but in practice companies tend to settle on four to eight segments. In a diverse, complex, and lucrative category more might be called for.

Agreeing on the number of segments, as well as the labels that will be attached to them, always requires careful study of their full profiles. While the segments may be defined with one class of basis variables (e.g. attitudes about fashion and shopping), you always need to describe the segments in terms of other variables (e.g. demographics, spending, media preferences, preferred brands) that were not part of the definition of the segments.

Application

An important deliverable from most segmentation research is a tool for classifying other consumers into the segments. This could be a formula where by asking a small number of questions you can place a "new" consumer in one of the segments. It could be a set of logical rules ["if combined account balances >$10,000 + have at least three accounts + non-mortgage debt < $20,000: assign to segment G"].

Conclusion

Market segmentation is a far-reaching strategy that can benefit players.

The underlying rationale is to use your resources more efficiently by serving consumer needs better. The mechanism is to treat the market as consisting of multiple segments with different needs, rather than as a mass market where "average" really suits no one very well.

The research requires a sample survey. It defines and describes segments from which the health plan can choose one or more targets. We believe playser in the new Medicare market are especially ready to get more "bang for the buck" by more precisely meeting segments’ needs and by using marketing tools to which that segment will respond.