Tuesday, September 21, 1999

1
Tuesday, September 21, 1999

• Marketing information systems
• Announcements
• email
• Tapes
• Cases
• Teams
• Ill be out Thursday through Friday AM
• Proquest
• Case MSA

2
The Market Driven Organization

• Gathers information
• Disseminates information
• Uses information

3
Example Celestial Seasonings

• Taste tests on site
• Weekly panels (3 sites/weekend)
• Focus groups in 11 cities
• Mall intercepts
• National samples
• Customer service
• Tracy Jones

4
Market Research Process

• 1. Determine final uses of information
• 2. Determine final report format
• 3. Specify necessary analysis
• 4. Determine data requirements
• 5. Scan available secondary data sources
• 6. Design study
• 7. Implement field work
• 8. Analyze and report

5
Potential Problems in Research Use

• Confusing managerial and statistical significance
• Confusion relationships and causality
• The use of inappropriate data
• Overreliance on quantitative data
• Pressure to generate desired solutions

6
Ethical Issues The User

• Issuing bid requests for free advice
• Poor use of information
• Making false promises
• Access to information

7
Market Assessment
8
Uses of Conjoint Analysis

• Product design
• Market segmentation
• Forecasting shares of product concepts
• Pricing

9
Conjoint is Suitable When . . .

• We must make tradeoffs between attributes and
  benefits in the product
• We can decompose the product in ways that are
  meaningful for customers and product design
• It is possible to describe the product bundles
  realistically

10
Product Design Conjoint Analysis

• Derive utility values for attributes and
  attribute options based on customers stated
  overall preferences for different bundles of
  attributes. Example Memory and Price bundles.

PriceMemory 1,000 1,500 2,000 32
Mb 4 2 1 64 Mb 7 5 3 128 Mb 9 8 6 9 Most
preferred 1 Least preferred
11
Simplified Utility Calculation
Price Part- Memory 1,000 1,500 2,000 Wor
th 32 Mb 4 2 1 7/3 2.3 64 Mb 7 5 3 15/3 5.0
128 Mb 9 8 6 23/3 7.7 20/3 15/3 10/3 Part-Worth
6.7 5.0 3.3 9 Most preferred 1
Least preferred
12
Utility for this Customer

• Example
• 128 Mb vs. 64 Mb 7.7 5.0 2.7 units
• 1,000 vs. 1,500 6.7 5.0 1.7 units
• So D 64 Mb is worth more than 500 to this
  customer.

13
Alternative Pairwise Comparisons of Full Profiles
For example

• PII 233
• 64MB
• 4.3 G HD
• 24X CD
• 1979

• PII 233
• 64MB
• 4.3 HD
• DVD
• 2299

14
Designing the Conjoint Study

• Determine relevant attributes
• Determine attribute levels
• Determine attribute combinations
• Choose stimulus representations
• Choose response type
• Choose data analysis technique

15
Market Share Forecast

• We can estimate market shares by estimating
  utility for different product offerings and
  calculating the percentages of preference for
  each product in the study

16
The Bass Diffusion Model

• When will a customer adopt a new product or
  technology?
• Useful when
• The product has been recently introduced
• The product has not yet been introduced but there
  are reasonable parallels

17
Assumptions of the Basic Bass Model

• Diffusion process is binary
• Constant number of maximum potential buyers
• All potential buyers will eventually purchase the
  product
• No repeat purchases or replacement purchases
• The impact or word of mouth is independent of
  adoption time
• Innovation is considered independent of
  substitutes
• The marketing strategies supporting the
  innovation are not explicitly included

18
The Bass Diffusion Model

• St p Remaining q Adopters
  Potential Remaining Potential
• Innovation Imitation
  Effect Effect

where St sales at time t p coefficient
of innovation q coefficient of imitation
Adopters S0 S1
St1 Remaining Total Potential
Adopters Potential
19
Examples of Bass Model Parameters
Innovation Imitation Product/ parameter
parameter Technology (p) (q) BW
TV 0.028 0.25 Color TV 0.005 0.84 Air
conditioners 0.010 0.42 Clothes
dryers 0.017 0.36 Water softeners 0.018 0.30 Recor
d players 0.025 0.65 Cellular telephones 0.004 1.7
6 Steam irons 0.029 0.33 Motels 0.007 0.36 McDona
lds fast food 0.018 0.54 Hybrid
corn 0.039 1.01 Electric blankets 0.006 0.24 A
study by Sultan, Farley, and Lehmann in 1990
suggests an average value of 0.03 for p and an
average value of 0.38 for q.
Source Lilien and Rangaswamy
20
Specification of the Model
21
Product Factors Affecting the Rate of Diffusion

• High relative advantage over existing products
• High degree of compatibility with existing
  approaches
• Low complexity
• Can be tried on a limited basis
• Benefits are observable

22
Market Factors Affecting the Rate of Diffusion

• Type of innovation adoption decision
• Communication channels used
• Nature of links among market participants
• Nature and effect of promotional efforts

23
Caveat

• Do customers have the ability to articulate
  preferences?
• Market research is probably not helpful when a
  new technology is not tied to familiar
  applications
• e.g., the personal computer, internet access

24
Observation Can Overcome . . .

• Customers who dont know possible applications
• Unreliability of self reporting
• Interruption/removal from natural use
• Giving expected answers

25
Empathic Design (Leonard and Rayport)

• Gathering, analyzing, and applying information
  gleaned from field observations
• Requires creative interdisciplinary analysis

26
Learning from Observation

• Triggers of use
• Interactions with the users environment
• User customization
• Intangible product attributes
• Unarticulated user needs

27
The Empathic Design Process

• Observation
• Capturing data
• Reflection and analysis
• Brainstorming for solutions
• Developing prototypes of possible solutions

28
The Challenge

• Linking technology with needs to develop solutions