Chapter 4 Notes: Marketing Research & Analytics

Marketing Research vs Analytics

• Marketing research: plan, collect, and analyze data to inform decisions.

• Marketing analytics: optimize collected data to improve marketing decisions; forward-looking insights.

Core Concepts

• Distinct but connected: research asks questions; analytics extracts insights to guide strategy.

• Big picture goal: turn data into actionable marketing decisions.

The Marketing Research Process

• Define the problem or research gap: clarify what you’re trying to fix or discover.

• Set research questions and plan data collection.

• Decide data sources:

  • Secondary data: existing data; pros: affordable/fast; cons: not tailored.

  • Primary data: new data collected for this study; pros: tailored; cons: more costly/time-consuming.

• Collect data (primary or secondary) and analyze it to draw conclusions.

• Present findings and recommendations to decision-makers; justify budget and actions.

Data Types and Sources

• Secondary data: existing, often affordable; not tailored to your exact question.

• Primary data: collected for this study; split into:

  • Quantitative (surveys, scales, numeric measures)

  • Qualitative (focus groups, interviews, observation, ethnography)

• Common data sources for analytics:

  • Internal website analytics (clicks, pages, funnels)

  • Social media monitoring

  • Online research communities

  • Government datasets (often dear to purchase but rich in demographics)

  • Supplier and partner inputs

Focus Groups and Primary Data Collection

• Focus groups: 6–10 participants; moderated discussions; aims to uncover motivations and deeper insights.

• Pros: rich qualitative data, uncover hidden motivations.

• Cons: observer interference, small samples, group dynamics can bias results; moderator quality is crucial.

• Variations: multiple moderators, hidden observers, or one-way mirrors.

Primary Data Collection Methods

• Surveys:

  • Types: online, mail, intercept (mall screens)

  • Question types: open-ended, closed-ended, scaled (e.g., 1–10, strongly agree–strongly disagree)

  • Online surveys: common but often low completion rates (typical rate around $3\%$)

• Observational/ethnographic methods: online communities, in-person observation, etc.

• Experiments: use to test causality and measure impact of changes.

• Data analysis after collection to derive insights and next steps.

Big Data and Technology in Marketing

• Big data: extremely large and complex data sets that exceed traditional tools’ ability to store/process easily.

• Volume example: about $2.5\times 10^{18}$ bytes of data generated every day.

• Netflix example: data-driven recommendations influence about $80\%$ of content watched.

• Big data analytics vs why: analytics can reveal patterns but may not explain why; qualitative methods explain why.

• Mixed methods: combine quantitative and qualitative approaches to get both numbers and explanations.

Data Organization and Mining

• Data organization challenges: structure data for analysis; data warehouses store large-scale data.

• Data mining: explore data to find patterns and relationships.

• Structured vs unstructured data:

  • Structured: clearly defined fields (e.g., customer list, transactions).

  • Unstructured: text, video, social posts; harder to analyze without processing.

Use of Data in Marketing Strategy

• Strategic applications: product development, promotions/advertising, audience targeting, personalization, and customer retention.

• Data can guide budgeting and resource allocation; executives often want data-driven justification.

Ethics, Consent, and Privacy

• Consent concerns and data sharing issues are prominent.

• Public social media data raises ethical questions; researchers may operate under differing consent norms depending on context.

• Data breaches and spam are practical risks; careful data governance is essential.

Technology Trends in Marketing Research

• Online focus groups: remote, interactive group discussions; some studies pay participants.

• Online research communities: continuous, qualitative data sources.

• Mobile marketing research: collecting data via smartphones.

• Social media analytics: leveraging platforms to study consumer conversations.

Data Visualization and Interpretation (What’s Important)

• Data organization into themes: qualitative analysis often uses thematic coding.

• Data-driven decisions should link back to concrete implications and actions for the business.

In-Class Experiment: Pepsi vs Coke Case (Illustrative Biases)

• Setup: taste test labeled with letters m and q; participants indicate preference.

• Pepsi case:

  • In a Dallas–Fort Worth test, labeling and taste led to more favorable Pepsi responses; sales impact observed in the region.

  • Broadly adopted as a promotional approach (taste tests to drive perception).

• Coke response: argued bias due to packaging/letter labeling (M vs Q) and conducted large-scale tests.

  • Historical bias: many people prefer the letter M over Q; studies show about $78\%$ prefer M, $22\%$ prefer Q.

• Classic contrast: taste vs branding can override preference; Coke later introduced New Coke (brand misalignment led to failure) and then reverted to Coke Classic.

• Takeaway: branding, packaging, and perceived identity can trump taste in consumer decisions; importance of aligning with brand image.

• Classroom note: the exercise illustrates how data can be manipulated or biased by presentation and sampling.

Quick Exam Prep and Schedule

• Next week:

  • Tuesday: exam prep and details in class (essential; not easily found online).

  • Thursday: first exam.

• Come prepared; what is discussed in class on Tuesday is intended to help you perform better on Thursday.