This summary
+380.000 other summaries
A unique study tool
A rehearsal system for this summary
Studycoaching with videos
Remember faster, study better. Scientifically proven.
PREMIUM summaries are quality controlled, selected summaries prepared for you to help you achieve your study goals faster!
Summary  Marketing Models

1.2 What is a model?

What is a model?A model is a simplified representationof the world built to help us understand the world and make predictions about it.

What is the ultimate question in assessing a model? How can it be answered?
 Is it useful?
 Best answered by comparing to another, competing model.

2.1 Introduction

What are the goals of a good segmentation scheme?
 Identifying a group of customers who are similar to each other in their preferences and purchases regarding the brand.
 Looking for segments to be different from group to group.

What does a cluster analysis algorithm do?It takes the input variables and computes a measure of the similarities between entities (e.g. customers). Then it groups together the entities that are most similar, keeping those that are more different in different clusters.

Entities can be customers or businesses.

2.2 Input variables

What is gigo?Garbage in garbage out. If we put variables in that are bad, the model won't give good results.

What types of variables can be used by marketeers?Indicators that are:
 Geographic
 Demographic
 Behavioral
 Attitudinal

What are the 5 main issues to consider when choosing which variables to include?
 Look at what data you already have inhouse. Not all may be interesting enough and there might be some missing that you don't have data for yet.
 Use free secondary data that you can find online, e.g. zip code. Avoids difficulty of asking your customers.
 Send small questionnaires to a small sample of your customers for data you are still missing.
 Preprocess the data. Checking the descriptive statistics on each variable. Useful variables have to exhibit some amount of variation (to make different clusters).
 Double check the proposed variables. Don't want to include too many variables that convey the same information because redundant variables implicitly get weighted more.

2.3 Measures of similarity

What is a measure of similarity among all the customers?Correlation coefficient. Ranges from +1 (2 customers have identical patterns) to 1 ( 2 customers have very different patterns).

What do you have to take notice of when using correlations?Correlations reflect relative patterns, not mean differences. E.g. 2 customers have bought the same 3 types of books, but 1 customer bought 2x as much. So high correlation, but 1 has a higher profile than the other.

What would you do if you don't want to subtract out the means?Compute a (Euclidean) distance between each pair of customers. Customers a, and b, k=1,2.. r attributes (r=7 book genres in this case).

Explain the measures of association method for SKU's and the simple matching coefficient (Smc).
 Measure of association method is a 2x2 matrix crossclassifying the purchases of 2 customers.
 Counts the number of products that both bought (a), that only one of them bought (b,c) and that neither of them bought (d).
 Smc is a measure of similarity and ranges from 0 to 1.
 Smc = (a+d)/(a+b+c+d).

What is the Jaccard coefficient?
 Like the simple matching coefficient (Smc), but ignores the d cell in both the numerator and the denominator. It ignores SKUs that appear in neither customers.
 J=a/(a+b+c)

2.4 Clustering algorithms

What are hierarchical clusters?Once 2 customers are put in the same segment, they are always together. Others might join the cluster, but the cluster is never broken into separate clusters in any of the model stages.

2.4.1 Hierarchical clustering models

In what 2 groups can hierarchical clustering techniques be further categorized?
 Agglomerative techniques
 Divisive techniques

What are agglomerative techniques?Every customer starts in his or her own segment, and with each iteration, the model puts together customers who are similar. Either by forming a new cluster or by adding a customer to an already existing cluster. Continues until all customers are in the same segment.

What are divisive techniques?All customers begin in one segment and each iteration breaks off the customer, customers or cluster segment that is the most different. Continues until everyone is in his or her own cluster.
Read the full summary
This summary. +380.000 other summaries. A unique study tool. A rehearsal system for this summary. Studycoaching with videos.