Market Intelligence Class 8. Agenda Experimental research – Factorial designs Segmentation – Tactical – Strategic – a priori and clustering approaches.

Slides:



Advertisements
Similar presentations
( ) Why Is Segmentation Important? –CConsider a heterogeneous hypothetical market Scatter plot of ideal products of customers in the market.
Advertisements

Survey design. What is a survey?? Asking questions – questionnaires Finding out things about people Simple things – lots of people What things? What people?
UNIT C The Business of Fashion
McGraw-Hill/Irwin McGraw-Hill/Irwin Copyright © 2009 by The McGraw-Hill Companies, Inc. All rights reserved.
Statistics for the Social Sciences
Factorial Designs Passer Chapter 9
Livelihoods analysis using SPSS. Why do we analyze livelihoods?  Food security analysis aims at informing geographical and socio-economic targeting 
AEB 37 / AE 802 Marketing Research Methods Week 7
Cluster Analysis.
Chapter 17 Overview of Multivariate Analysis Methods
Chapter Seventeen Copyright © 2006 McGraw-Hill/Irwin Data Analysis: Multivariate Techniques for the Research Process.
Intro to Factorial ANOVA
By Wendiann Sethi Spring  The second stages of using SPSS is data analysis. We will review descriptive statistics and then move onto other methods.
Market Analysis and Strategy MKT 750 Dr. West. Agenda Marketing Analysis & Strategic Planning – Essential Elements (5Cs, STP, 4Ps) – Situation Analysis.
SW388R7 Data Analysis & Computers II Slide 1 Multiple Regression – Basic Relationships Purpose of multiple regression Different types of multiple regression.
Repeated Measures ANOVA Used when the research design contains one factor on which participants are measured more than twice (dependent, or within- groups.
Market Intelligence Session 7 Experimental Research.
Clustering analysis workshop Clustering analysis workshop CITM, Lab 3 18, Oct 2014 Facilitator: Hosam Al-Samarraie, PhD.
Levels of Market Segmentation
Targeting Research: Segmentation Birds of a feather flock together, i.e. people with similar characteristics tend to exhibit similar behaviors Characteristics.
Conjoint and Segmentation
© 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Review: T test vs. ANOVA When do you use T-test ?
Chapter 10 THE NATURE OF WORK GROUPS AND TEAMS. CHAPTER 10 The Nature of Work Groups and Teams Copyright © 2002 Prentice-Hall What is a Group? A set of.
A statistical method for testing whether two or more dependent variable means are equal (i.e., the probability that any differences in means across several.
Learning Objectives Copyright © 2002 South-Western/Thomson Learning Multivariate Data Analysis CHAPTER seventeen.
8 Identifying Market Segments and Targets
 Slide 1 Two-Way Independent ANOVA (GLM 3) Chapter 13.
Two-Factor ANOVA Method We are interesting in examining the effects on the response variable of two factors simultaneously. - How do price and advertising.
Data Science and Big Data Analytics Chap 4: Advanced Analytical Theory and Methods: Clustering Charles Tappert Seidenberg School of CSIS, Pace University.
One-Way Analysis of Covariance (ANCOVA)
ANCOVA. What is Analysis of Covariance? When you think of Ancova, you should think of sequential regression, because really that’s all it is Covariate(s)
Market Segmentation, Targeting, and Positioning
Review of Factorial ANOVA, correlations and reliability tests COMM Fall, 2007 Nan Yu.
Comparing Two Means Chapter 9. Experiments Simple experiments – One IV that’s categorical (two levels!) – One DV that’s interval/ratio/continuous – For.
Customer-Driven Marketing Strategy: Creating Value for Target Customers 7 Principles of Marketing.
CHAPTER 5: Marketing Information & Research Mrs. Piotrowski Principles of Marketing 1.
B121 Chapter 11 Marketing. It is concerned with exchange relationships. Transactional marketing – oriented towards single purchase Relationship Marketing.
Handout Twelve: Design & Analysis of Covariance
 Seeks to determine group membership from predictor variables ◦ Given group membership, how many people can we correctly classify?
ANCOVA.
HYPOTHESIS TESTING FOR DIFFERENCES BETWEEN MEANS AND BETWEEN PROPORTIONS.
Chapter Seventeen Copyright © 2004 John Wiley & Sons, Inc. Multivariate Data Analysis.
1 Cluster Analysis Prepared by : Prof Neha Yadav.
Analyzing Data. Learning Objectives You will learn to: – Import from excel – Add, move, recode, label, and compute variables – Perform descriptive analyses.
Ass. Prof. Dr. Özgür KÖKALAN İstanbul Sabahattin Zaim University.
Scatterplots & Correlations Chapter 4. What we are going to cover Explanatory (Independent) and Response (Dependent) variables Displaying relationships.
Market Segmentation, Targeting, and Positioning
Market Segmentation, Targeting, and Positioning Boe Dube
Business Intelligence
Bell Ringer List five reasons why you think that some new businesses have almost immediate success while others fail miserably.
Segmentation, Targeting and Positioning Strategies
Market Segmentation, Targeting, and Positioning. The STP Process Segmentation is the process of classifying customers into groups which share some common.
MKT 498 Education for Service-- snaptutorial.com.
MKT 498 Teaching Effectively-- snaptutorial.com
Indicator 1.04 – Employ marketing information to develop a marketing plan Part II.
Chapter Eight: Quantitative Methods
Interactions & Simple Effects finding the differences
Marketing Information
Hypothesis testing. Chi-square test
Indicator 1.04 – Employ marketing information to develop a marketing plan Part II.
One way ANOVA One way Analysis of Variance (ANOVA) is used to test the significance difference of mean of one dependent variable across more than two.
Experimental Design: The Basic Building Blocks
Cluster Analysis.
DESIGN OF EXPERIMENTS by R. C. Baker
Market Segmentation and Strategic Targeting
Cluster analysis Presented by Dr.Chayada Bhadrakom
 .
Chapter 7 Identifying Market Segments and Selecting Target Markets by
Chapter Ten: Designing, Conducting, Analyzing, and Interpreting Experiments with Two Groups The Psychologist as Detective, 4e by Smith/Davis.
Presentation transcript:

Market Intelligence Class 8

Agenda Experimental research – Factorial designs Segmentation – Tactical – Strategic – a priori and clustering approaches

Experiments - Factorial Designs 2 or more independent variables (manipulated and/or measured), each with two or more levels. – Type 1: 2 marketing mix variables Both variables manipulated Important for determining whether you need to coordinate marketing actions – Type 2: “tactical segmentation” (1 segment responds differently to a marketing mix variable than another segment) Segmenting variable is measured, marketing action is manipulated Important for determining whether you should segment for that particular marketing action 3

Experiments - Factorial Designs What to look for in factorial designs – Is there a main effect of A? – Is there a main effect of B? – Is there an interaction between A and B? (interaction: effect of one IV on DV depends on level of another IV) Analysis 4

Type 1: 2 marketing mix variables You manage the Oreo account for Kroger – One person on your team manages advertising, another manages in store promotions – You run an experiment to evaluate whether ads and/or end-of-aisle display increase sales You manipulate ads and display between store locations in one city

Oreo Promotion Experiment Factor A: Ads in local paper a1 = no ads a2 = ad in Thursday local paper Factor B: Display location b1 = regular shelf b2 = end aisle Store 1: a1, b1 Store 2: a1, b2 Store 3: a2, b1 Store 4: a2, b2 6 Dependent variable: expenditures per customer over subsequent 2 weeks

OREO PROMOTION EXPERIMENT Scenario 1 (EXPENDITURES/CUSTOMER/2 WKS) 7

8 Main effect of A? Main effect of B?

OREO PROMOTION EXPERIMENT Scenario 1 (EXPENDITURES/CUSTOMER/2 WKS) 9 Interaction? this diff v. this diff

OREO PROMOTION EXPERIMENT Scenario 1 (EXPENDITURES/CUSTOMER/2 WKS) 10 Interaction? this diff vs. this diff

SALES OF OREOS 11

SALES OF OREOS 12 2 main effects, no interaction

Oreo Promotion Experiment Scenario 2 (Expenditures/customer/2 wks) Main effect of A? Main effect of B?

Oreo Promotion Experiment Scenario 2 (Expenditures/customer/2 wks) Interaction? 0.50

15 “Cross-over” interaction SALES OF OREOS (Expenditures/customer/2 wks)

Oreo Promotion Experiment Scenario 3 (Expenditures/customer/2 wks)

Oreo Promotion Experiment Scenario 3 (Expenditures/customer/2 wks) Main effect of A? Main effect of B?

Oreo Promotion Experiment Scenario 3 (Expenditures/customer/2 wks) 18 Interaction? 1.30

SALES OF OREOS (Expenditures/customer/2 wks) 19 “fan effect” interaction

How to analyze in SPSS: ANOVA If both IVs are manipulated between subjects – Analyze – General linear model – Univariate – Bring DV to Dependent variable, IVs to “Fixed factors” – Options -- estimated marginal means – bring main effects and interactions to right side

Data in SPSS

Look for significance of 2 main effects and interaction

Look for means of main effects and interaction

Oreo Example No A x B interaction – Effect of changing A (Ads) is independent of level of B (Display Location). – Implies that Ad & Display decisions can be decoupled…they influence sales additively A x B interaction – Effect of changing A (ads) depends on level of B (display location), and/or vice-versa – Cannot decouple variables 25

Type 2: Tactical Segmentation Should groups be treated same or differently with respect to specific marketing decision variable? A is a controllable decision variable and B is a potential segmentation variable – Interaction means that segments respond differently to this marketing lever – Example: coupon x urban/suburban Question: does marketing mix variable have bigger effect for segment A or B? Is coupon more effective in urban or suburban neighborhoods? 26

Interactions and segmentation 27

Interactions and segmentation 28 Pending ANOVA results, coupons have a bigger effect in the suburbs

Tactical Segmentation Example - Dog Food 1 potential segmentation variable (Size of Dog) 2 decisions – Price: Hi v. Lo – Ad Theme: Love of Master v. Dog’s Active Life 29

Segmentation Example: Dog Food I (rated on 10 pt scale) Price Advertising 30

Segmentation Example: Dog Food I Price Advertising 31

32

Implications of Contrast A variable that is an excellent basis for segmentation with respect to one decision about a marketing mix element may be a poor basis for segmentation with respect to another mix element For any given mix element decision, when evaluating alternative bases for segmentation, look for ones with big differences in sensitivity to mix variable. 33

Tactical Segmentation II Example - Dog Food 1 decision: Price (hi vs. lo) 2 potential segmentation variables – Size of Dog – Income of owner 34

Segmentation Example: Dog Food II 35 How to compare? ANOVA

Segmentation Example: Dog Food II 36 How to compare? Can also look at “difference of differences”

Segmentation Example: Dog Food II 37

How to analyze in SPSS: ANOVA If both IVs are manipulated between subjects – Analyze – General linear model – Univariate – Bring DV to Dependent variable, IVs to “Fixed factors” – Options -- estimated marginal means – bring main effects and interactions to right side If 1 IV is between subjects and the other is within subjects – Analyze – General linear model – repeated measures – Name within-Ss variable and enter # levels, choose define, drag variables to right – Bring between-Ss IV to “between subject” box on right – Options – estimated marginal means – bring main effects and interactions to right side

Let’s analyze this in SPSS: same subjects rated both ads 39

Data in SPSS

Look for sign. of within-Ss main effect, between-Ss main effect, and interaction within within x between between

Look at means for main effects and interaction

Strategic v. Tactical Segmentation Tactical Segmentation – Should groups be treated same or differently with respect to specific marketing decision variable? – Evaluate interactions Strategic segmentation – What product markets to serve? – A priori vs. clustering (analytic) approaches in segmentation 43

Strategic: Choosing a Basis for Segmentation 3 Criteria for useful segments Different options or “levels” for segmenting 44

Criteria for “Useful” Segments 1.Homogeneity within segment, heterogeneity between 2.Systematic differences in behaviors 3.Marketing Mix efficiency potential – Make more money treating as segments than if we treated market as unsegmented whole 45

Potential bases for segmentation 46 Demographics/psychographics Lifestyle/Usage situation Product benefits desired Brand beliefs, perceptions, and preferences Purchase intentions Purchase behavior

Choosing how to segment When would you want to choose variables from one level vs. another? – What information is Available – Is a segmentation Actionable – Strength of correlation with Purchase Behavior 47

Potential bases for segmentation 48 Demographics/psychographics Lifestyle/Usage situation Product benefits desired Brand beliefs, perceptions, and preferences Purchase intentions Purchase behavior Less correlated with behavior More correlated with behavior

Choosing how to segment When would you want to choose variables from one level vs. another? – What information is Available – Is a segmentation Actionable – Strength of correlation with Purchase Behavior – Time frame planned for using segmentation 49

Potential bases for segmentation 50 Demographics/psychographics Lifestyle/Usage situation Product benefits desired Brand beliefs, perceptions, and preferences Purchase intentions Purchase behavior Less correlated with behavior More correlated with behavior Longer time- frame Shorter time- frame

Strategic Segmentation: A Priori A priori -- Segments chosen by the analyst before collecting data Often based on demographics – Example: Golfers, Male, 50-70, income > $50,000 – IBM initial approach for “Rubik’s Cube” segmentation 51

Strategic segmentation: Clustering Clustering-based approaches (a posteriori) – Collect data - ask battery of questions (lifestyle, benefits sought, etc) – Find natural clusters/ segments – Segments do not have labels - describe segments by their mean answers – IBM prometheus used this – Let’s try it with Kerlander soup

Kerlander Soup Raw Data Kerlander'sFisherman'sKerlander'sCapeKerlander's SubjectRegularDelightCreamyCodExtraCreamy

Clustering We want to run analyses that reveal which of the Ps cluster with other Ps and get a feel for how many segments might exist Our segmentation basis here is each subject’s taste rankings for the five brands. Do subjects segment in a systematic way based on their tastes? Use the Kerlander Soup Data for the 20 respondents.

Types of cluster analysis Hierarchical – Quantifies how far apart/close together 2 cases are, then forms groups – After running, you determine how many clusters you need/want K-means clustering – You select number of clusters – Estimates cluster means and assigns each case to cluster for which its distance to cluster mean is smallest 2 step clustering – First step, cases assigned to “preclusters” – Second step, preclusters are clustered using hierarchical algorithm.

Cluster Analysis Let’s run all 3 versions of cluster analysis What segments are revealed?

Hierarchical Clustering Will start with each individual as segment and begin combining until 1 segment is reached Look at solution at each stage and see what is clustering together and where there is a large “distance” associated with forming the next cluster

Hierarchical Clustering 58

Hierarchical Clustering Will start with each individual as segment and begin combining until 1 segment is reached Look at solution at each stage and see what is clustering together and where there is a large “distance” associated with forming the next cluster In SPSS – Make sure data are sorted by Sub # (helps for later interpretation – May need to standardize variables if on different scales – Choose Analyze – Classify – Hierarchical Cluster Pick Variables (the 5 soups that you have taste rankings for) Statistics: choose agglomeration schedule Choose Plots – Dendogram to get the chart of segments being formed

Output: Agglomeration Schedule

Output – dendogram X-axis is measure of distance – Closer you are to the left side, the smaller the distance between the objects that were combined. As you get big leaps in distance between the combined objects, you question whether you should be combining these segments Compare dendogram to original data. – Why were 10, 19, 4, 5, and 8 immediately combined into a segment? – What about 15, 16, 1, 9, and 3? – Based on this chart, how many segments?

Cluster SPSS – Kerlander Soup

Kerlander Soup Raw Data Kerlander'sFisherman'sKerlander'sCapeKerlander's SubjectRegularDelightCreamyCodExtraCreamy

Output – dendogram X-axis is measure of distance – Closer you are to the left side, the smaller the distance between the objects that were combined. As you get big leaps in distance between the combined objects, you question whether you should be combining these segments Compare dendogram to original data. – Why were 10, 19, 4, 5, and 8 immediately combined into a segment? – What about 15, 16, 1, 9, and 3? – Based on this chart, how many segments?

Cluster SPSS – Kerlander Soup If we went to two segments we would combine 10, 19, …, 14, 6 in a single segment At this stage, 10,19,4,5 and 8 will join with 13 and 17

Cluster SPSS – Kerlander Soup If we went to two segments we would combine 10, 19, …, 14, 6 in a single segment At this stage, 10,19,4,5 and 8 will join with 13 and 17

Now re-run analysis Choose Save – Single solution – 3 In data set: new variable representing cluster membership Use new variable to examine differences in segments – Descriptive statistics -- Cross-tabs – membership on row, preferences on columns

This will also be new variable in data set

Now re-run analysis Choose Save – Single solution – 3 In data set: new variable representing cluster membership Use new variable to examine differences in segments – Descriptive statistics -- Cross-tabs – membership on row, preferences on columns

K-Means Clustering Choose Analyze – Classify -- K-Means cluster – Pick Variables – Pick Number of Clusters (3 based on other cluster analysis) – Choose Save – Cluster Membership – Choose Save – Distance from Cluster Center

K-Means Cluster Analysis SPSS will save variables in your data that reflect segment membership as well as the distance an object (a taste tester here) is from the center of the cluster. – Ideally: all members close to the center of the assigned cluster. – Tradeoff: the more clusters you have, the closer each member will be to the center of the cluster, but the less interpretable / practical Check the segment assignment. Compare it to the hierarchical approach we took.

2-step cluster analysis Analyze – classify – 2 step cluster Put variables into categorical or continuous Distance measure: Euclidean Options – standardize if necessary Output – create cluster membership variable Results in this case?

Which one to use? Hierarchical – Good for small data sets – Can easily examine solutions with increasing numbers of clusters K-means clustering – If you know how many clusters you want – Good for moderately sized data sets 2 step clustering – For large data files (>1,000 cases) – Mixture of continuous and categorical variables