Lecture 1 – The marketing research process
• This course is not about:
o Only learning 4Ps theory in a vague way.
o Very heavy math formulas only.
• You will use data and statistics, but always linked to real marketing problems.
Why marketing research matters (real-life examples)
Tropicana packaging redesign
In 2009, Tropicana changed its orange juice packaging. After the redesign:
• Sales dropped by 20%.
• Revenue decreased by 30 million dollars.
• They had just spent 35 million dollars on advertising.
Later, they said they underestimated the emotional bond customers had with the old
packaging. Which wasn’t something that came out in their research. A month later, they
changed back the design.
Shows that bad or incomplete research is dangerous.
Coca-Cola “New Coke”
Coca-Cola replaced the 99-year-old formula with New Coke. Goals:
• Regain market share.
• Make the brand feel fresh and modern to new audiences.
They did about 200,000 taste tests with consumers (spent over $1 million). Results show
that people preferred New Coke in tests (53% vs. 47%).
After launch, people were very angry: protests, calls, lawsuits.
After 79 days, Coca-Cola admitted the mistake and brought back “Coca-Cola Classic”.
Lesson: even large, expensive research can be wrong if it doesn’t capture real
feelings, context and brand meaning.
Emotions: Kmart Australia “1000 Mums”
• Kmart wanted to understand emotions around low prices.
• Research showed that some consumers feel anxiety and shame about needing low
prices, but there is also pride in living within one’s means.
• The “1000 Mums” campaign tried to:
o Make people feel proud to live within their budget.
o Reduce shame about shopping at low-price stores.
• The campaign was based on academic research (Roberts et al., 2015).
• Results after the campaign (2011-2013):
o 20% increase in total annual visits.
o 42% more items sold.
• Key idea: research helps you find the right emotional story, not just rational facts.
,Research, data and algorithms in the real world: Netflix artwork personalization
• Netflix uses recommendation systems and A/B testing.
• Even the poster image (artwork) you see for a show like Stranger
Things is personalized.
The algorithm chooses from several images, based on what you
usually watch. This uses big amounts of data and experiments to
increase clicks and viewing time.
Importance of marketing research
• 50-95% of new products fail.
• About 50% of advertising has no effect.
Old quote: “Half the money I spend on advertising is wasted; the trouble is I don’t
know which half.”
• 85% of price promotions do not pay off.
Green idea: without good research, managers really don’t know what works.
Goals of marketing research & why you should care
Marketing research helps to guide and improve decision making in a marketing plan (for
example STP: segmentation, targeting, positioning). Goals:
1. Trace problems: find out what is going wrong.
2. Understand changes: why are sales, attitudes, markets changing?
3. Predict outcomes: what will happen if we change something (price, design, ad)?
Marketing research
Marketing research = the systematic and objective identification, collection, analysis and
dissemination of information for the purpose of improving decision making related to the
identification of problems and opportunities in marketing.
• Systematic and objective → not random or biased.
• Focus: information that helps decisions.
• Green comment: it may sound boring, but it becomes interesting when you connect it
to real managerial problems.
Systematic
You follow a structured method, not intuition or guesswork.
Example study: carefully set up variables like packaging colour and food type and
measure purchase intention → that’s a scientific approach.
Objective
You rely on evidence and scientific numbers, not personal opinion.
Why marketing research?
Main purpose: solve business problems, there are many different approaches, examples:
• Segmentation, Targeting, Positioning (STP) → most difficult one.
• 4 Ps: product, price, promotion, place.
, • Customer Relationship Management (CRM).
• Market-based asset theory of brand competition (marked as the “most difficult
one”).
What happens in an Internet minute
Data ≠ actionable insights
Data alone is not enough; you must turn data into insights and decisions.
Examples of insights:
• Netflix recommendations screen: suggests movies based on your past behaviour.
• Research can create brand positioning maps to show how brands are positioned in
the consumer’s mind.
Research can take many forms
Just a dew examples:
• Qualitative vs. quantitative.
• Explorative vs. descriptive vs causal.
• Secondary vs. primary data.
• Survey vs. transaction data.
• Ad-hoc vs. continuous vs panel data.
• B2B vs. consumer research.
• Applied vs. scientific.
• Valid vs. not valid.
, Marketing research process (6 steps)
1. Problem definition
2. Development of an approach
3. Research design formulation
4. Fieldwork / data collection
5. Data preparation & analysis
6. Report preparation
1) Problem definition
Problem definition is the most important decision. You look at:
• Symptoms (lower profits, lower market share).
• Goals (expansion, new market, growth).
To define the problem you:
• Clarify the MDP.
• Consider purpose, background and information needed.
• Write the research problem and questions.
• Make sure management and researcher agree.
• Check that collecting information is worth it (has value).
Problem definition for the earlier cases
• Tropicana: regain market share, modernize the brand.
• New Coke: regain market share, re-invigorate brand for different audiences.
• Kmart: regain market share, re-juvenate the brand.
Harley-Davidson example
• Management decision problem: Should Harley produce more?
• Marketing research problem: Are Harley’s customers loyal buyers?
• Research questions:
o Who are the customers?
o Are there different customer segments?
o How do they feel about their Harleys?
o Are they loyal?
2) Development of an approach
• Objective / theoretical framework:
o Which variables (constructs) are relevant?
o Based on academic literature and existing findings.
• Model:
o Variables and how they are related.
o Can be verbal, graphical or mathematical.
• Hypotheses:
o Tentative statements about relationships (“higher pride → higher store visit
likelihood”).
o They are possible answers to the research questions.
o They must be testable with data.
• This course is not about:
o Only learning 4Ps theory in a vague way.
o Very heavy math formulas only.
• You will use data and statistics, but always linked to real marketing problems.
Why marketing research matters (real-life examples)
Tropicana packaging redesign
In 2009, Tropicana changed its orange juice packaging. After the redesign:
• Sales dropped by 20%.
• Revenue decreased by 30 million dollars.
• They had just spent 35 million dollars on advertising.
Later, they said they underestimated the emotional bond customers had with the old
packaging. Which wasn’t something that came out in their research. A month later, they
changed back the design.
Shows that bad or incomplete research is dangerous.
Coca-Cola “New Coke”
Coca-Cola replaced the 99-year-old formula with New Coke. Goals:
• Regain market share.
• Make the brand feel fresh and modern to new audiences.
They did about 200,000 taste tests with consumers (spent over $1 million). Results show
that people preferred New Coke in tests (53% vs. 47%).
After launch, people were very angry: protests, calls, lawsuits.
After 79 days, Coca-Cola admitted the mistake and brought back “Coca-Cola Classic”.
Lesson: even large, expensive research can be wrong if it doesn’t capture real
feelings, context and brand meaning.
Emotions: Kmart Australia “1000 Mums”
• Kmart wanted to understand emotions around low prices.
• Research showed that some consumers feel anxiety and shame about needing low
prices, but there is also pride in living within one’s means.
• The “1000 Mums” campaign tried to:
o Make people feel proud to live within their budget.
o Reduce shame about shopping at low-price stores.
• The campaign was based on academic research (Roberts et al., 2015).
• Results after the campaign (2011-2013):
o 20% increase in total annual visits.
o 42% more items sold.
• Key idea: research helps you find the right emotional story, not just rational facts.
,Research, data and algorithms in the real world: Netflix artwork personalization
• Netflix uses recommendation systems and A/B testing.
• Even the poster image (artwork) you see for a show like Stranger
Things is personalized.
The algorithm chooses from several images, based on what you
usually watch. This uses big amounts of data and experiments to
increase clicks and viewing time.
Importance of marketing research
• 50-95% of new products fail.
• About 50% of advertising has no effect.
Old quote: “Half the money I spend on advertising is wasted; the trouble is I don’t
know which half.”
• 85% of price promotions do not pay off.
Green idea: without good research, managers really don’t know what works.
Goals of marketing research & why you should care
Marketing research helps to guide and improve decision making in a marketing plan (for
example STP: segmentation, targeting, positioning). Goals:
1. Trace problems: find out what is going wrong.
2. Understand changes: why are sales, attitudes, markets changing?
3. Predict outcomes: what will happen if we change something (price, design, ad)?
Marketing research
Marketing research = the systematic and objective identification, collection, analysis and
dissemination of information for the purpose of improving decision making related to the
identification of problems and opportunities in marketing.
• Systematic and objective → not random or biased.
• Focus: information that helps decisions.
• Green comment: it may sound boring, but it becomes interesting when you connect it
to real managerial problems.
Systematic
You follow a structured method, not intuition or guesswork.
Example study: carefully set up variables like packaging colour and food type and
measure purchase intention → that’s a scientific approach.
Objective
You rely on evidence and scientific numbers, not personal opinion.
Why marketing research?
Main purpose: solve business problems, there are many different approaches, examples:
• Segmentation, Targeting, Positioning (STP) → most difficult one.
• 4 Ps: product, price, promotion, place.
, • Customer Relationship Management (CRM).
• Market-based asset theory of brand competition (marked as the “most difficult
one”).
What happens in an Internet minute
Data ≠ actionable insights
Data alone is not enough; you must turn data into insights and decisions.
Examples of insights:
• Netflix recommendations screen: suggests movies based on your past behaviour.
• Research can create brand positioning maps to show how brands are positioned in
the consumer’s mind.
Research can take many forms
Just a dew examples:
• Qualitative vs. quantitative.
• Explorative vs. descriptive vs causal.
• Secondary vs. primary data.
• Survey vs. transaction data.
• Ad-hoc vs. continuous vs panel data.
• B2B vs. consumer research.
• Applied vs. scientific.
• Valid vs. not valid.
, Marketing research process (6 steps)
1. Problem definition
2. Development of an approach
3. Research design formulation
4. Fieldwork / data collection
5. Data preparation & analysis
6. Report preparation
1) Problem definition
Problem definition is the most important decision. You look at:
• Symptoms (lower profits, lower market share).
• Goals (expansion, new market, growth).
To define the problem you:
• Clarify the MDP.
• Consider purpose, background and information needed.
• Write the research problem and questions.
• Make sure management and researcher agree.
• Check that collecting information is worth it (has value).
Problem definition for the earlier cases
• Tropicana: regain market share, modernize the brand.
• New Coke: regain market share, re-invigorate brand for different audiences.
• Kmart: regain market share, re-juvenate the brand.
Harley-Davidson example
• Management decision problem: Should Harley produce more?
• Marketing research problem: Are Harley’s customers loyal buyers?
• Research questions:
o Who are the customers?
o Are there different customer segments?
o How do they feel about their Harleys?
o Are they loyal?
2) Development of an approach
• Objective / theoretical framework:
o Which variables (constructs) are relevant?
o Based on academic literature and existing findings.
• Model:
o Variables and how they are related.
o Can be verbal, graphical or mathematical.
• Hypotheses:
o Tentative statements about relationships (“higher pride → higher store visit
likelihood”).
o They are possible answers to the research questions.
o They must be testable with data.
