Engineering B-Plan Development:

Market Research
Ram Babu Roy, RMSoEE
IIT Kharagpur
References:
1. Entrepreneurship: Successfully launching New ventures, Bruce R Barringer, R Duane
Ireland, 3rd Edition, Pearson (2010).

2. Technology Ventures: From Idea to Enterprise, 2nd Edition, Richard C Dorf and Thomas H
B (2008)

3. Conjoint and Discrete Choice Designs for Pricing Research. Vinay Kanetkar, Department
of Consumer Studies
& Various web pages on internet

Note: This study material has been prepared for the purpose of classroom discussions
only.
 Market research
• Market research is the process of gathering the
information that serves as the basis for a sound
marketing plan.
• Objective: Learn how to attract and retain customers
• Does the target segment want the perceived value that
our positioning is trying to deliver more than other
segments? If so, how can we reach this segment
efficiently?
• Use of Market research to estimate
– Market potential
– Sales forecast
• Focus group
• Conjoint Analysis
 The Market Research Process
• Define the product and its unique selling proposition. Identify
the customer segment. Develop a set of questions that will
provide the necessary data on customer preferences and
behavior.

• Collect the data using surveys, published sources, focus

groups, interviews, and other means to secure it.

• Analyze and interpret the data to determine if the product

meets the needs or wants of the customers and determine
whether they will pay the price you seek.

• Draw conclusions on the customer and their needs,

preferences and behavior.
 Conjoint Analysis
• Measures the value the market places on each feature of
the product and predicts the value of any combination of
features
• To understand customer preferences, values and choices
• Needs a tool to measure the trade-off between price and
value
• Efficient and effective approach to understand the
customer’s price and value trade-off
– Measures how buyers value components of a product/service
bundle
– Dictionary definition-- “Conjoint: Joined together, combined.”
– Marketer’s catch-phrase-- “Features CONsidered JOINTly”
• Central idea - willingness to give-up something for gaining
something else
 Conjoint Analysis
• Doesn’t force consumers to think separately
about individual attributes, and their levels in
isolation
• Asks the consumer to make judgments about
products overall value
• Uses mathematical analysis to uncover the
value system which must be behind the
preference judgments
 Breaking the Problem Down
• If we learn how consumer values the components of
a product, we are in a better position to design those
products
– to maximize societal welfare
– that improve profitability
• How to Learn What Customers/Public Want?
• One way: Ask Direct Questions about Importance
– How important is it?
• Rather than ask directly whether you prefer Power
over Fuel Economy, we present realistic tradeoff
scenarios and infer preferences from your product
choices
 Stated Importance

• Answers often have low discrimination, with most

answers falling in “very important” categories
– If they were not important, we probably wouldn’t have
included them in the research!
• Answers sometimes useful for segmenting market,
but still not very actionable
– We still don’t exactly know what product they want
 Main idea in conjoint analysis
• Construct the value system by asking about
preferences on a small subset of products
• Using the system to make predictions about
relative preferences for any products that are
plausible and could be sold in the market
place
• Expertise in this technique can give a
competitive advantage
 Concepts underlying the procedure
• Decomposition of a product into its attributes -
defining products as collections of attributes
• Valuation of the utility of each individual attribute
– importance
– most desired level (For each consumer)
• Provides a method to assess an individual’s
“value system,”
• Specifies directly or indirectly the monetary value
a consumer puts on each level of each of the
attributes
• Allows us to assess a consumer’s willingness to
trade off one feature for another
 Pricing problem in Banking Industry
Cost per Use
• Two attributes are Free $0.20
potentially important to Machine Rank 1 Rank 3
consumers: (ATMB)
– transaction mode Mode Telephone Rank 2 Rank 4
(ATB)
– cost per transaction
Internet Rank 5 Rank 6
(IBB)

Rank order the six possible combinations from most(1) to least preferred (6)
Code the best as 5 points down to the least desired alternative as zero.

Derived Value System

 Prediction from Value System

Respondent need not rank all possible products in order to be able to derive the value system
 Conjoint Analysis: Study Design
• Determine Relevant Attributes and their levels.
• Determine Product Presentation: Content and
Form
• Determine Respondent - Researcher Interaction
format
• Decide on Response Type, rating, ranking, or
choice.
• Determine Criterion for judging, liking,
purchasing or willing to pay.
• Decide on Data Analysis Technique
 Research design
• The stimuli descriptions of the product must convey
all the information that respondents feel they need
to make their decisions
• Conduct customer research project to understand
the key attributes prior to conducting the conjoint
study
• Focus on attributes that are proposed for change
and feasible to implement
• Brand name should be used in conjoint studies to
assess brand value and equity
• Consider attributes that are important to customers
and managers
 Relevant Attributes
• Must specify the attributes that influence customer decisions
• If one included an attribute that has no real importance to
most customers, the value system will reflect that.
• Very difficult to detect the absence of an important attribute
• kinds of attributes
– Physical attributes - refers to the product itself (weight or size).
– Performance benefit - refers to outcome (km per litre).
– Cost-based attributes - refers to cost of acquiring or continuing to
use (installation cost, monthly charges or fees).
– Psychological positioning - refers to user perception, (assurance)
• Inclusion of all attributes is neither practical nor economically
feasible
 Defining Attributes
• Attributes are independent aspects of a product or a
service (Brand, Price, Size, Color etc.)
• How many attributes?
• Depends on research objectives
– One rule of thumb was that no more than 6 or 7 attributes
• May cause respondents to simplify, looking only at 2-3 most important
• Attributes should be independent, mutually exclusive
– Brand, quality and product life expectancy may all measure the
same thing

• Each attribute has varying degrees, or “levels”

– Cost: $1, $2, $3

– Biodiversity Loss: 10, 50, 100

• Each level is assumed to be mutually exclusive of the

others
Rules for Formulating Attribute Levels
• Attributes are assumed to be mutually
exclusive
– Attribute: Add-on features
– Level 1= Sun roof
– Level 2= GPS system
– Level 3=DVD player
– If you define levels in this way, you cannot
determine the value of providing 2 or 3 of these
features at the same time (or none of them)
 Solutions
• 8 level Attribute: • 3 Binary Attributes:
• Features – Sunroof:
– None • None
• Sunroof
– Sunroof
– GPS System
– GPS system • None
– DVD Player • GPS
– Sunroof, GPS – DVD Player
– Sunroof, DVD • None
– GPS, DVD • DVD Player
– Sunroof, GPS, DVD
Rules for Formulating Attribute Levels
• Don’t include too many levels for any one
attribute
– The usual number is about 3-5 levels per attribute
– Make sure levels from your attributes can combine
freely with one another without resulting in utterly
impossible combinations (very unlikely combinations
OK)
• Determining Which Attributes & Levels to Include
– Talk to all stakeholders
– Focus Groups
– Search of competitors websites, sales materials
Rules for Formulating Attribute Levels
• Levels should have concrete/unambiguous
meaning
• “very expensive” vs “ costs $575”
• “weight: 5-7 kilos” vs “ weight 6 kilos”

• One description leaves meaning up to individual

interpretation, while the other does not
 Sample Size
• Products with long list of attributes and levels require
a large sample (several hundred) to understand
preferences
• By focusing on three-to-five key attributes, one can
understand the value system using a small sample
(<25).
• If the analyst is unwilling to reduce the set of
attributes to between four and seven, product
presentation must be made together with the choice
of attributes.
• Another alternative is to let the respondent choose
the top four-to-seven attributes. In such a case, each
respondent is offered a different conjoint study.
Product Presentation to the Respondent
• Partial or full profile method
– While the full list in some sense represents reality, its use
may render the rating or choice task too complex and
confusing for consumers
– using partial profiles, the researcher gets a better
understanding of the desired level and relative importance
• Presentation form
– Descriptions based entirely on words provide better
predictive validity than pictorial presentation
– Prototype and computer animation can improve the
predictive validity
– Attributes based on experience and / or emotion, can only be
described in a limited way using words
 Respondent - Researcher Interaction
Format
• Personal interview mode - expensive
• Computer assisted interview modes
– Allow respondents to design and create their own
list of attributes and attribute levels
• Mixed mode data collection to reduce costs
– Three broad activities, initial contact, responding to
instrument and communicating responses
– telephone, mail, and facsimile - usually cheaper
– problems relating to sampling, response rate, and
response quality
 Response Type
• Format in which respondents are asked to
express their judgments
• Ratings or ranks (“paired comparison” or a top-
to-bottom )or choices
• Ranking - Sort the product descriptions into three
piles, for example: Like very much, Like
moderately and Like little or not at all. Then, rank
within the piles
• Situation involving more than two alternative -
“discrete choice experiments”
 Response Type

Paired comparison: Which car would you consider buying?

Choice tree: implementation of the opt-out option

 Response Type
• Traditionally, ranking methods were preferred
because providing a quantitative measurement of
the “degree of liking” or “degree of intention to
buy” was felt to strain the capabilities of
respondents.
• Full rankings often take four to five times as much
time as one would to perform rating tasks
• Choice of method is largely situation-specific and
relates to the form that a respondent is able to
provide more reliable and / or valid responses
 Criterion
• Standard to be used in the judgments
– preference
– likelihood / intention to purchase
• Which do you prefer? versus Which are you more likely
to buy?

• Choice depends largely on whether the focus of the

study is market share or unit sales (market size)
• Intention to purchase is necessary to gauge the likely
market size
 Analyzing the Outputs
• Each individual provides a set of judgments and
his / her value system
• There is no assumption that all consumers have
the same value system
• Three major types of analysis are:
– Aggregate analysis of attribute importance and
desirability.
– Segmentation analysis.
– Competitive scenario simulations to predict sales
levels
 Interpreting Conjoint Utilities

• You cannot compare one level from one attribute

with one level from another attribute, since conjoint
utilities are scaled to an arbitrary constant within
each attribute (often zero-centered)
• You CAN compare differences between two levels of
one attribute versus two levels of another attribute
• Difference among the scores for levels of a given
attribute gives a rough measure of that attribute’s
importance
 Aggregate Analysis
Average part worths of each attribute level Relative importance (rough indicator)

• Relative importance is proportional to the range

covered by the levels
• Percentages are dependent on the specific levels of
the attribute used in the study
• Good estimate of importance only if the variable
levels specified cover the range of relevant options
 Segmentation Analysis
• Cluster analysis can be used to produce “benefit
segments”
• Group respondent such that segments have similar
within a segment and different across segments
• Look at predefined group of customers based on some
prior knowledge about them
– current versus prospective customers
– heavy versus light volume buyers
• Understanding the variation of attribute importance
and desired levels across consumers is crucial for target
market selection
Example: Notebook Computer

Similarly, products B and C would have value of 3.4 and 2.5 respectively
 Choice Predictions
• Three major rules
• “first choice” – will buy the product that has the
highest value
– A product’s market share is simply the percentage of
consumers for whom that product “wins”
• “share of preference”- gives a probability
estimate that a consumer will buy a brand
– consumers do not always buy their most preferred
alternative
• Share of attraction
– necessary to compute share of exponentials
(attractions or utilities) for negative values
 Example of Share Calculations
• The market share for a product is the average
purchase probability across all customers in the
study
• Share of attraction approach is a compromise
between the two extremes, most preferred or
share of preference (value) approaches
 Scenario Simulations
• Predicting market shares or unit sales in various
scenarios
• Given the value system of a consumer and a
description of alternative products, one can
calculate the value of alternative products.
 Scenario Simulations
• The scenario simulation capability is a powerful tool
• Helps assess product introduction strategy
• Can obtain not only a market share estimate, but also an
indication of which competitive products will be hurt
most.
• This is achieved by first simulating the scenario of
– only the current competitive products being available
– then the environment of current competitive products plus
the prospective new product
• The scenario simulation also could be used to determine
product profitability.
 Accuracy of Conjoint Analysis
• Fit between observed ranks and predicted ranks capture
individual differences very well
• Satisfactory accuracy in many applications
• Important in an individual situation to be able to check
the validity of the findings before implementing actions
• The three primary checks are:
– common-sense test, or face validity (based on prior,
strongly-held belief)
• If parameter estimates vary across individuals in a reasonable way
– holdout prediction
• small number of the original products rated by the respondent are
“held out” from the calculation of the value system and validated
– actual vs. predicted market share (ultimate test)
 Necessary Assumptions
• Product as a bundle of attributes
• Researcher must know important attributes
• Respondents can reasonably rate products
• Attributes should be actionable
• Choice shares and price premiums are
indicators
– manager must weigh risks associated changing
prices in terms of profitability, market share and
achieving sustainable competitive advantage