Use of Statistical Techniques in Market

Research

An overview

What is market research?

Essentially analysing peoples relationships to brands/products/policy

FMCG/Services/social policy Two components:

• Qualitative

• Quantitative

• Both do the ‘same’ thing

• build models

Broad brush strokes

Who do you want to talk to? Has someone else/you talked to the

above? Why do you want to talk to them? How will this information be used?

Examples

Poll survey – eg. Colmar Brunton Poll Market segmentation

• Can I simplify thinking about my customers into easier to understand segments

Customer satisfaction studies• What should I change (and how) my

customers experience of this service

Examples

Advertising effectiveness

Which products should I launch• At what price/ what features

Colmar Brunton Poll

Sampling – talk to eligible voters Sample needs to be representative Needs to be done in 1-4 days CATI interviewing used Question wording very important

One News Colmar Brunton poll

RELEASED: Sunday 5th November 2008 POLL CONDUCTED: Evenings of 1st-5th October 2007 inclusive SAMPLE SIZE: N = 1010, Eligible VotersSAMPLE SELECTION: Random nationwide selection using a type of stratified sampling

to ensure the sample includes the correct proportion of people in urban and rural areas.

SAMPLE ERROR: Based on the total sample of 1010 Eligible Voters, the maximum sampling error estimated is plus or minus 2.8%, expressed at the 95% confidence level.

METHOD: Conducted by CATI (Computer Assisted Telephone Interviewing).WEIGHTING: The data has been weighted to Department of Statistics Population

Estimates to ensure it is representative of the population in terms of age, gender, and household size.

REPORTED FIGURES: Reported bases are weighted. For Party Support, percentages have been rounded up or down to whole numbers, except those less than 3.5% which are reported to 1 decimal place. For all other figures percentages have been rounded up or down to whole numbers except those less than 1% which are reported to 1 decimal place.

METHODOLOGY The party vote question has been asked unprompted as at February 1997.

One News Colmar Brunton poll

One News Colmar Brunton poll

Party Support“Under MMP you get two votes.

One is for a political party and is called a party vote.The other is for your local M.P. and is called an electorate vote.”

Party Vote*

“Firstly thinking about the Party Vote which is for a political party.

Which political party would you vote for?”IF DON’T KNOW –

“Which one would you be most likely to vote for?”

Segmenting your market

Segmentation Identify segmentation bases and segment the market. Develop profiles of resulting segments.

Targeting Evaluate attractiveness of each segment. Select target segments.

Positioning Identify possible positioning concepts for each target segment. Select, develop, and communicate the chosen concept.

U.S. SME Lifestyle Segmentation Example

Segment Definition

Avg line of credit balance ($000)

Projected avg borrowings next 12

month ($000)Top Borrowing Needs

Bolting start-ups Less than 3 years old; sales over $1 million

90 225 Working capital management

Young laggards 4-10 years old; no or flat sales growth

125 50 Facility expansion

Mature cash cows 10+ years old; moderate sales growth

175 140 Equipment financing

Experienced expanders

10+ years old; high sales growth

675 200 Equipment financing; facility expansion

Segmenting your market

Techniques used (of many)• Factor analysis to reduce dimensions

• Cluster analyses to create ‘segments• Fastclus algorithm very popular

• A difficult process

• Needs much direction form experienced ‘segmentors’

• Discrimination techniques used to place future observations into segments

Customer Satisfaction Studies

Thinking firstly about the service you received from (top secret). I am going to read you some statements about this service and as I read you each statement, please give your opinion using a five-point scale where 1 is extremely dissatisfied and 5 extremely satisfied

(read, rotate (start at x). write in (one digit) per statement)

How satisfied or dissatisfied are you with:.

... everything being kept straightforward

... being kept in touch while the claim was being processed

... the general manner and attitude of the staff you dealt with

... your claim being dealt with promptly

... being treated fairly

Importance Index by Means

prompt

fairly

kept in touch

straightforward

manner/attitude

0.2

0.3

0.4

0.5

0.6

0.7

3.9 4 4.1 4.2 4.3 4.4 4.5 4.6

performance (means)

Imp

ort

ance

ind

exPresentation of results

Strengths

Maintain or divertSecondary drivers

Concern

Satisfaction studies

How do you measure importance?? Correlation fine but misses relationships

between Xs MLR bad as assumes X’s uncorrelated PCR used – regression on principal

components etc… Other solutions include CART

New product/pricing studies

What new features need to be included/added

What price are most customers prepared to pay?

How do my competitors’ prices affect my share of the market

Heineken Steinlager Stella Artois

Other premium beers

Export Gold

Please indicate which of the following you would buy if these were the prices per dozen:

$22.95

Other mainstream beers

$18.95

$20.95

$14.95

19.95 $13.95

SCENARIO 1SCENARIO 1

Amstel

$18.95

New product/pricing studies

Techniques used: Experiments designs

• Main effects/interactions

Choice modelling• Multinomial logistic regression

• Survival analysis

New product/pricing studies

Actual Tarps

How advertising is modelled...How advertising is modelled...

New

5-10% decay in recall/week

How advertising is modelled...

Give TARPS a memory = Adstock• Via exponential smoothing

Linear model on Adstock vs ECT intercept = long term memory slope = return /unit Adstock Once happy can use for scheduling…

Major points

Constrained optimisation

The problems are both stimulating and demanding

Interpretation is everything

Useful website

Check out our website to find out more about the Market Research Society of New Zealand

www.mrsnz.org.nz