NEW PRODUCT MANAGEMENT

Chapter 7

Analytical Attribute Approaches:

Trade-Off Analysis and Qualitative Techniques

*

Trade-Off (Conjoint) Analysis
Put the determinant attributes together in combinations or sets.
Respondents rank these sets in order of preference.
Conjoint analysis finds the optimal levels of each attribute.

*

Conjoint Analysis Input: Salsa Example
Figure 7.2

*

Thickness

Spiciness

Color

Actual Ranking*

Ranking as Estimated by Model

Regular

Mild

Red

4

4

Regular

Mild

Green

3

3

Regular

Medium-Hot

Red

10

10

Regular

Medium-Hot

Green

6

8

Regular

Extra-Hot

Red

15

16

Regular

Extra-Hot

Green

16

15

Thick

Mild

Red

2

2

Thick

Mild

Green

1

1

Thick

Medium-Hot

Red

8

6

Thick

Medium-Hot

Green

5

5

Thick

Extra-Hot

Red

13

13

Thick

Extra-Hot

Green

11

11

Extra-Thick

Mild

Red

7

7

Extra-Thick

Mild

Green

9

9

Extra-Thick

Medium-Hot

Red

14

14

Extra-Thick

Medium-Hot

Green

12

12

Extra-Thick

Extra-Hot

Red

17

18

Extra-Thick

Extra-Hot

Green

18

17

* 1 = most preferred, 18 = least preferred.

Regular Thick Ex-Thick

UTILITY
2

1

0

-1

-2
Mild Medium-Hot Ex-Hot
Red Green
Thickness Spiciness Color
0.161 0.913 -1.074 1.667 0.105 -1.774 -0.161 0.161
Conjoint Analysis: Graphical Output
Figure 7.3

*

Conjoint Analysis:

Relative Importance of Attributes
0 20 40 60 80 100 %
Spiciness
Thickness
Color
59.8%
34.6%
5.6%
Figure 7.3
(cont’d.)

*

Is Conjoint The Right Method?
Must be able to break the product down into discrete attributes.
Customer must combine these features rationally when evaluating brands (do you buy a Smart Car because it’s practical and gets good mileage, or because it’s cool?)
Complex products may have too many attributes to consider – be sure the important ones are selected.
Purchase occasion not considered.
Joint decisions not considered (does the husband or wife make the car purchase decision?)
Assumes some familiarity with the product category; may be less useful for new-to-the-world products.
Firm must be able to act on the results!

Conjoint Analysis for More Complex Problems
The above example was a full-profile conjoint analysis (all possible combinations were included).
For larger problems, a fractional factorial design may be used (not every combination is included, but a subset of the combinations in which each level and each attribute is used at least a few times). Rankings and results are similar to the full-profile analysis.
More advanced methods are needed for very complex problems, and also to handle interactions among attributes.

*

Commercially Available Conjoint Methods
Advanced conjoint analysis methods available commercially include:
Adaptive conjoint analysis (respondent judges the importances of the attributes, then responds to options that focus on the most important attributes and levels).
Choice-based conjoint analysis (the respondent is shown several choices in combination and asked which is preferred)
Both of these procedures reduce the number of attributes to which the respondent is exposed.

*

Virtual Prototypes in Concept Testing
Information Acceleration (IA): applied by GM to test new electric car concepts.
Uses virtual buying environment that simulates the information available in a purchase situation.
Uses surrogate travel technology to allow the customer to virtually walk around the showroom and look at computer-generated car prototypes.

Some Qualitative Attribute Analysis Techniques
Dimensional Analysis
Checklists
Relationships Analysis
Analogy
There are many others.

*

A Dimensional Attribute List
Weight
Rust resistance
Length
Color
Water resistance
Materials
Style
Durability
Shock resistance
Heat tolerance

Explosiveness
Flammability
Aroma
Translucence
Buoyancy
Hangability
Rechargeability
Flexibility
Malleability
Compressibility

Figure 7.4

*

An Idea Stimulator Checklist

for Industrial Products
Can we change the physical/chemical properties of the material?
Are each of the functions really necessary?
Can we construct a new model of this?
Can we change the form of power to make it work better?
Can standard components be substituted?
What if the order of the process were changed?
How might it be made more compact?
What if it were heat-treated/hardened/cured/plated?
Who else could use this operation or its output?
Has every step been computerized as much as possible?

Figure 7.5

*

Templates for Creativity
Attribute Dependency: Find a functional dependency between two attributes. Ex.: color of ink on coffee cup is sensitive to heat and can reveal message if coffee is too hot.
Replacement: Remove a component and replace with one from another environment. Ex.: antenna is replaced by headphone cord on Walkman.
Displacement: Remove a component and its function to change the product. Ex.: Removing floppy drives resulted in ultra-thin PCs.
Component Control: Find a new connection between a component internal to the product and one external to the product. Ex.: Toothpastes with whiteners, suntan lotions with skin moisturizers.

Source: Jacob Goldenberg and David Mazursky, Creativity in Product Innovation, Cambridge University Press, 2002.
Figure 7.6

*

Relationships Analysis
Force combinations of dimensions (features, functions, and benefits) together.
Techniques:
Two-dimensional matrix
Multidimensional (morphological) matrix
Two-dimensional example: person/animal insured and event insured against.
Household cleaning products example used six dimensions:
Instrument used, ingredients used, objects cleaned, type of container, substances removed, texture or form of cleaner

*

Another Form of Dimensional Analysis
Two key dimensions for winning new product ideas:
Utility lever: How the product will affect the customer’s life (such as simplicity, fun/image, environmental friendliness, reduced risk, convenience, and productivity).
Buyer’s experience cycle: The stage when/where the product will affect the customer (purchase, delivery, use, supplements, maintenance, disposal).

Source: W. C. Kim and R. Mauborgne, “Knowing a Winning Idea When You See One,” Harvard Review, Sept.-Oct. 2000, pp. 129-138.
Figure 7.8

*

Morphological Matrix:

New Coffee Maker
Heating:
Heating element in pot
Open flame under pot
Microwave unit
Adding Coffee:
By spoon
Built-in measuring cap
Automatic feed
Filtering:
Filtering paper
Porous ceramic filter
Centrifuge method
Keeping Coffee Warm:
Thermal insulating technology
Warming unit in put
External heat source
Pouring Coffee:
Valve under pot
Pump in lid of pot
Espresso-like jets

*

Analogy
Airplane food systems: analogous to preparing, serving and consuming food at home, suggesting adaptations of food processes and furniture.
Amusement park lines: park designers were inspired by watching cattle being herded.
Bicycle: analogous to driving a car (steering, moving, turning) but key differences suggesting new types of bicycles (e.g., electric bikes).

Criteria for Selecting Analogous Products
Analogy should be vivid and have a life of its own.
It should be full of concrete images.
It should be a happening – a process of change or activity.
It should be a well-known activity and easy to visualize and describe.

Other Methods:

Lateral Search Techniques
Free association
Stereotype activity
Lateral thinking — avoidance
Creative stimuli words
Studying “big winners”
Use of the ridiculous
Forced relationships

*

Lateral Thinking — Avoidance
Keep an idea from dominating thinking as it always has in the past by asking avoiding questions.
Ask “Is there another way of looking at this?”
Ask “Why?”
Focus on an aspect of the problem other than the “logical” one.
List all possible alternatives to every aspect of the analysis.
Break apart aspects (concepts) of the problem, or combine them to create even more concepts.

*

Some Creative Stimuli Words
Guest stars
Alphabet
Truth
Outer space
Charity
His and hers
Style
Nation
Family
Videotape

Photography
Testimonials
Decorate
Fantasy
Hobbies
Holidays
Weather
Calendar
Push button
Snob appeal

*

Use of the Ridiculous
How can you join two wires together?
Hold them with your teeth.
Use chewing gum.
Can you think of others?
Do any of these ridiculous ideas suggest a not-so-ridiculous solution?

*

Thickness

Spiciness

Color

Actual

Ranking*

Ranking as

Estimated

by Model

Regular

Mild

Red

4

4

Regular

Mild

Green

3

3

Regular

Medium-Hot

Red

10

10

Regular

Medium-Hot

Green

6

8

Regular

Extra-Hot

Red

15

16

Regular

Extra-Hot

Green

16

15

Thick

Mild

Red

2

2

Thick

Mild

Green

1

1

Thick

Medium-Hot

Red

8

6

Thick

Medium-Hot

Green

5

5

Thick

Extra-Hot

Red

13

13

Thick

Extra-Hot

Green

11

11

Extra-Thick

Mild

Red

7

7

Extra-Thick

Mild

Green

9

9

Extra-Thick

Medium-Hot

Red

14

14

Extra-Thick

Medium-Hot

Green

12

12

Extra-Thick

Extra-Hot

Red

17

18

Extra-Thick

Extra-Hot

Green

18

17

* 1 = most preferred, 18 = least preferred.