The Individual or Micro level

sunandaC

New member
The Individual or Micro level

As interactions of the individual with the firm increases, the firm obtains more data about him/her. Offering individualized value adding propositions can strengthen relationship with the individual customer. For this, we need to track the cu and mine at the individual or micro level.

Some important features to note about mining at this level are:

• Micro-level mining provides specific information about a particular customer. For example, the retail store can go to the extent of finding out the preferred colours of his shirt

• A firm takes up micro level mining to build a detailed customer profile of a regular customer.

• Data mining this level might be expensive if the data mining tool has to cull out individual information from a large database. Having a separate database for profitable customer might be helpful.

• Knowledge obtained at the individual level is useful in situations where:
 The firm wants to customize its offering to the customer based on the customer’s tastes and preferences e.g. the retail store can offer discounts on the purchase of a bundle of products that the customer prefers buying together.

 The firm wants to assist the purchase of a new product based on the information it has of the last purchase. For example, if a customer has bought a suit in his visit, then the store might offer a discount on the purchase of a tie of a matching colour.

 The firm wants to take advantage of the personal events in a customer’s life (e.g. birthdays, anniversaries, birth of child etc.) to further cement the precious relationship.

 Current patterns that go against usually observed customer behavior point to interesting phenomenon. If retail customer suddenly switches brand then he/she might not be satisfied with the last purchase.

The most common operations used at this level are: -

Classification

Classification is a process that maps a given data item into one of the several predefined classes. CRM uses classification for a variety of purposes like behavior prediction, product and customer categorization.

Regression

Regression is the operation of learning a function that predicts the value of a real valued dependent variable based on values of other independent variables. Regression finds application in a CRM environment where prediction needs to be made about the behavior regarding real value added variables. Suppose the retail store collects data on the monthly visits of the customers viz. Frequency, time spent on each visit. And purchases made during each visit. If the manager has a strong intuition that total purchase is linked to frequency of visit, then this situation can be modeled by regression. This model can then be used to predict future purchases of a customer. Regression needs sufficient amount of data to be reliable and valid.

Link Analysis


Link Analysis seeks to establish relationship between items or variables in a database record to expose patterns and trends. Link analysis can also trace connections between items of record over time. The most important link analysis application in CRM, called market basket analysis, is an operation that seeks relationship between product items characterizing product affinities or buyer preferences.


The retail store collects thousands of interactions daily. A link analysis task performed on this data will point to items that are bought together e.g. bread and butter are bought together rather than bread and orange juice. Such information can be used to design store layouts, design coupons, etc.

Segmentation

Segmentation aims to identify a finite set of naturally occurring clusters or categories to describe data.

Deviation Detection

Deviation Detection (DD) focuses on discovering the most significant changes in the data from previously measured, expected or normative values. Most CRM solutions have a DD task running in parallel on a regular basis. Suppose a retailer finds out that the sales from a particular section of the store have been much less than expected. This deviation on further analysis points out to non-stocking of a popular brand.


Tools such as decision trees, rule induction, case based reasoning, visualization techniques, nearest neighbor techniques, clustering algorithms, etc are used for the above purposes.
 
The Individual or Micro level

As interactions of the individual with the firm increases, the firm obtains more data about him/her. Offering individualized value adding propositions can strengthen relationship with the individual customer. For this, we need to track the cu and mine at the individual or micro level.

Some important features to note about mining at this level are:

• Micro-level mining provides specific information about a particular customer. For example, the retail store can go to the extent of finding out the preferred colours of his shirt

• A firm takes up micro level mining to build a detailed customer profile of a regular customer.

• Data mining this level might be expensive if the data mining tool has to cull out individual information from a large database. Having a separate database for profitable customer might be helpful.

• Knowledge obtained at the individual level is useful in situations where:
 The firm wants to customize its offering to the customer based on the customer’s tastes and preferences e.g. the retail store can offer discounts on the purchase of a bundle of products that the customer prefers buying together.

 The firm wants to assist the purchase of a new product based on the information it has of the last purchase. For example, if a customer has bought a suit in his visit, then the store might offer a discount on the purchase of a tie of a matching colour.

 The firm wants to take advantage of the personal events in a customer’s life (e.g. birthdays, anniversaries, birth of child etc.) to further cement the precious relationship.

 Current patterns that go against usually observed customer behavior point to interesting phenomenon. If retail customer suddenly switches brand then he/she might not be satisfied with the last purchase.

The most common operations used at this level are: -

Classification

Classification is a process that maps a given data item into one of the several predefined classes. CRM uses classification for a variety of purposes like behavior prediction, product and customer categorization.

Regression

Regression is the operation of learning a function that predicts the value of a real valued dependent variable based on values of other independent variables. Regression finds application in a CRM environment where prediction needs to be made about the behavior regarding real value added variables. Suppose the retail store collects data on the monthly visits of the customers viz. Frequency, time spent on each visit. And purchases made during each visit. If the manager has a strong intuition that total purchase is linked to frequency of visit, then this situation can be modeled by regression. This model can then be used to predict future purchases of a customer. Regression needs sufficient amount of data to be reliable and valid.

Link Analysis


Link Analysis seeks to establish relationship between items or variables in a database record to expose patterns and trends. Link analysis can also trace connections between items of record over time. The most important link analysis application in CRM, called market basket analysis, is an operation that seeks relationship between product items characterizing product affinities or buyer preferences.


The retail store collects thousands of interactions daily. A link analysis task performed on this data will point to items that are bought together e.g. bread and butter are bought together rather than bread and orange juice. Such information can be used to design store layouts, design coupons, etc.

Segmentation

Segmentation aims to identify a finite set of naturally occurring clusters or categories to describe data.

Deviation Detection

Deviation Detection (DD) focuses on discovering the most significant changes in the data from previously measured, expected or normative values. Most CRM solutions have a DD task running in parallel on a regular basis. Suppose a retailer finds out that the sales from a particular section of the store have been much less than expected. This deviation on further analysis points out to non-stocking of a popular brand.


Tools such as decision trees, rule induction, case based reasoning, visualization techniques, nearest neighbor techniques, clustering algorithms, etc are used for the above purposes.

Well, many many thanks for your help and providing the information on The Individual or Micro level. BTW, i am also going to upload a document where you can find some useful information and can also included in your report..
 

Attachments

  • The Individual or Micro level.pdf
    130.5 KB · Views: 0
Top