Marketing Research

Marketing research is the systematic study of what customers need and want, and how best to meet those needs. Its primary role is to reduce the risk of bad business decisions by clarifying interactions among stakeholders — customers, suppliers, employees. Research provides information and decision-making insight; it does not eliminate risk or guarantee correct decisions.

graph TD
    MR[Marketing Research\nStudy what customers need and want]
    MR --> PR[Research Process]
    PR --> S1[1. Study the current situation]
    S1 --> S2[2. Select a research method]
    S2 --> S3[3. Collect and analyze data]
    S3 --> S4[4. Prepare the report\nRecommendations]

    MR --> ME[Four Methods]
    ME --> OB[Observation\nWatch behaviour\nWhat — not why]
    ME --> SU[Survey\nRepresentative sample\nWhat and why]
    ME --> FG[Focus Group\n6–15 people · Moderator\nQualitative why]
    ME --> EX[Experimentation\nA/B test\nIsolate variables]

How It Appears Per Course

ADMN 201

LO4 frames marketing research as a four-step process and compares four methods on their strengths and limitations. The chapter also covers modern evolutions (crowdsourcing, data mining) and ethical concerns around electronic observation.

The Research Process

Professional research follows a structured workflow, not a “gut feeling”:

1. Study the current situation — What is the need? What do we not know?
2. Select a research method — Which method best fits the question?
3. Collect and analyze the data — Gather information from the right sources.
4. Prepare the report — Translate findings into actionable recommendations.

Skipping any of these steps — especially launching a product without collecting data — means making a decision without the information it requires.

The Four Research Methods

Method	Mechanism	Best For	Limitation
Observation	Viewing or monitoring consumer behaviour (in-store, cameras, loyalty card data)	Actual behaviour — what people do, not what they say	Does not explain why
Survey	Questioning a representative sample about purchasing attitudes and practices	Specific quantifiable answers from a population	Respondents may lie or give socially desirable answers
Focus Group	Moderator-led discussion with 6–15 people; researchers observe from behind one-way mirror	In-depth qualitative themes and opinions — the why	Small sample; not statistically representative
Experimentation	Comparing reactions of similar people under different circumstances (A/B test)	Isolating specific variables to test cause-and-effect	Expensive; harder to conduct at scale

Observation

• Can be low-tech (watching shoppers) or high-tech (cameras, loyalty cards, electronic data tracking).
• Modern form: “electronic observation” and data analytics.
• Classic example: A toy manufacturer noticed high sales of red wagons (not green) by analyzing sales records.
• Ethics concern: “Video mining” using high-definition cameras raises privacy issues. Cambridge Analytica built voter profiles through electronic observation, crossing the line into privacy violation.

Survey

• The questionnaire is designed to elicit honest answers about attitudes and practices.
• Representative sample: You cannot afford to survey everyone — so you survey a small group that accurately represents the larger population. If the sample is not representative, the data is worthless.
• Modern tool: Momentive (formerly SurveyMonkey) enables rapid online data collection.

Focus Group

• A trained moderator leads the discussion; researchers watch from behind a one-way mirror.
• Produces rich qualitative data about motivations, attitudes, and perceptions.
• Tells you why observation often cannot. Example: A focus group might reveal that customers find a label confusing — something a sales report would never show.

Experimentation

• The scientific method applied to marketing decisions.
• Example: Does a candy bar sell better with or without walnuts? Make two batches, offer them to similar groups, measure the preference.
• Real-world example: A fast food chain releasing a new sandwich only in Toronto and Vancouver first — a controlled test before national rollout.

Modern Research Methods

Crowdsourcing: Gathering information and opinions via social media and smartphone apps from a large group. Faster and cheaper than traditional focus groups, but requires filtering noise from valid data.

Data Mining: Automated analysis of massive datasets to find previously undiscovered patterns. Example: Fairmont Hotels used data mining to discover their customers preferred the Savoy in London — directly influencing an acquisition decision.

Cross-Course Connections

MarketingConcept — research is the mechanism by which the Marketing Concept learns what customers actually need
MarketSegmentation — research methods are used to identify and understand segments
ConsumerBuyingProcess — research into rational and emotional motives informs how to intervene in the buying process
Causation — experimentation is the only research method that can establish causal relationships (PHIL252)
Bias — non-representative samples and observer effects are forms of selection bias (PHIL252)

Key Points for Exam/Study

• LO4: Marketing research = study of what customers need/want to reduce decision risk
• 4 research process steps: Study situation → Select method → Collect/analyze → Report
• 4 methods: Observation (actual behaviour), Survey (representative sample), Focus Group (qualitative depth), Experimentation (A/B, isolates variables)
• Observation tells you what but not why — for the why, use Focus Groups or Surveys
• Experimentation is the only method that can establish cause-effect — other methods reveal patterns
• Representative sample: You don’t survey everyone — just a group that accurately reflects the population; biased samples produce useless data
• Crowdsourcing is a modern survey/focus group hybrid — faster, but noisier
• Research provides information, not answers — “marketing decisions are seldom perfect” even with research

Open Questions

• As electronic observation (cookies, tracking pixels) becomes universal, does it replace traditional surveys?
• How should firms handle the ethics of data mining when customers don’t know they’re being analyzed?