Research methodology fails in a specific way when the wrong type is applied to the question.

A team wants to understand why brand consideration is declining among a specific segment. They commission a quantitative tracking survey. The data confirms consideration has declined. It tells them nothing about why — because quantitative descriptive methodology measures the state of the phenomenon, not the mechanism behind it. The method was appropriate for a different question.

Understanding the types of research methodology available, what each is built on, and what questions each is genuinely suited to, is the practical foundation of research programmes that produce findings capable of informing decisions — rather than just producing findings.

Research methodology is not a single approach. It is a family of frameworks classified along four dimensions: research philosophy, logical approach, design orientation, and data type. Each dimension defines a different aspect of what the methodology can and cannot do. For how the specific design types within this classification apply to commercial brand research — and how each maps to specific commercial questions, types of market research covers the decision-led selection framework.

Classification 1: Research Philosophy — The Foundation Layer

The most fundamental classification is by research philosophy: the assumptions about the nature of knowledge and reality that the methodology operates within.

Positivism

The assumption: reality is objective, external to the observer, and can be measured and quantified. The goal of research is to produce generalisable findings through systematic observation and testing.

Associated with: quantitative methods, experimental designs, statistical analysis.

Most appropriate for: research questions that require measurement, comparison, and generalisation across defined populations. How many consumers hold this attitude? What is the brand consideration rate in this segment? Does this product configuration produce higher purchase intent than the alternative?

Interpretivism

The assumption: reality is socially constructed and understood differently by different people. Understanding it requires access to the meanings people attach to their experiences, behaviours, and contexts — not measurement of an external objective state.

Associated with: qualitative methods, depth interviews, ethnographic observation, thematic analysis.

Most appropriate for: research questions that require understanding of meaning, motivation, and context. Why do consumers associate this brand with these values? What does category participation mean to this consumer segment? How does the purchase decision feel from the inside?

Pragmatism

The assumption: neither positivism nor interpretivism is right as a fixed commitment. The research philosophy should be selected based on what the research question requires.

Associated with: mixed-method designs that combine qualitative and quantitative approaches based on what the question requires — not what any single philosophical tradition prescribes.

Most appropriate for: commercial brand research, where most questions require both measurement precision and motivational depth. Pragmatism is the most practically applicable philosophical foundation for brand-side research because it does not require choosing between measurement and understanding — it requires choosing which combination of both the specific question needs.

Classification 2: Logical Approach — The Direction of Reasoning

Within each philosophical tradition, research methodology can be classified by its logical approach: the direction of reasoning that connects theory and data.

Deductive

Direction: theory → data. Start with an established theory or hypothesis and test it against collected data.

Most appropriate for: hypothesis-testing research where the proposition is clearly enough defined to design a specific test. Does increasing brand familiarity lead to increased purchase intent? Does this specific product formulation outperform the current one? Deductive approaches are most common in positivist, quantitative programmes where the hypothesis is clear before the research begins.

Inductive

Direction: data → theory. Start with data and work toward patterns and theoretical propositions.

Most appropriate for: research where the goal is to generate new understanding rather than test a predetermined hypothesis. What is happening in this category that the brand team hasn't fully articulated yet? What motivational patterns emerge from in-depth consumer conversations? Inductive approaches are most common in interpretivist, qualitative programmes where the research is designed to surface what the team didn't know to ask.

Abductive

Direction: iterative movement between data and theory, developing the most plausible explanation for observed findings.

Most appropriate for: research where neither a fully formed hypothesis nor a blank-slate exploration is the right starting point — where there is some understanding but also significant unexplained variance. The researcher moves between observations and frameworks, revising both in response to the other. Abductive approaches are associated with pragmatist, mixed-method programmes where the goal is the best practical explanation rather than hypothesis confirmation or pure theory generation.

Classification 3: Research Design Type — What the Study Is Built to Achieve

A third classification organises methodology by design orientation — what the research is structured to produce.

Exploratory

Purpose: investigate a topic that is not well understood. Generate insight and hypotheses rather than test them.

Most appropriate when: the research team does not have sufficient understanding of the topic to design hypothesis-testing research. Exploratory methodology is most valuable at the start of a research programme — before the questions are precise enough to be measured or the relevant variables are well enough defined to be tested.

Descriptive

Purpose: characterise the current state of a phenomenon. How widespread is it? What form does it take? Who is affected?

Most appropriate when: the research question requires an accurate picture of current consumer or market reality. The majority of commercial brand research is descriptive — brand tracking, usage and attitude studies, competitive benchmarking, and consumer segmentation are all designed to describe what is, not to explain why or to test causal relationships. For the complete guide to how descriptive survey research design specifically should be structured, descriptive survey research design covers the design requirements.

Explanatory

Purpose: establish causal relationships. Not just describing that two things are associated, but explaining why one causes the other.

Most appropriate when: the research question requires causal understanding — why a change in brand communication produced a change in brand consideration, why a specific product attribute drives repeat purchase, whether a campaign caused an awareness change. Explanatory methodology requires experimental designs — controlled variation of the causal variable with measurement of the outcome. Without experimental control, correlation is the most that can be established. For how exploratory, descriptive, and explanatory designs differ and when each applies, exploratory, descriptive, and causal research design covers the selection framework.

Classification 4: Data Type — What the Research Works With

The fourth classification organises research methodology by the type of data it produces and analyses.

Qualitative

Works with non-numerical data: spoken and written language, visual content, observed behaviour. Produces rich, contextually grounded findings about meaning and motivation. Does not support statistical generalisation — qualitative findings describe the nature, depth, and texture of consumer attitudes, not their distribution or frequency across a population.

Quantitative

Works with numerical data: structured survey responses, experimental measurements, coded observational data. Produces statistically generalisable findings about frequency, distribution, and relationships. Does not capture the depth of meaning that qualitative approaches generate — numbers describe what is happening across the population without explaining what it means to the individuals experiencing it.

Mixed Method

Combines both data types within a single research programme, using each to compensate for the limitations of the other.

Qualitative provides depth and motivational insight. Quantitative provides measurement precision and population-level generalisability. The integration between the two is where the most complete consumer understanding is produced — where the motivational framework surfaced qualitatively is validated and sized quantitatively, and where the quantitative patterns surfaced by tracking are explained and contextualised qualitatively. For the complete framework on how primary data types — qualitative and quantitative — differ in what they produce and how quality should be assessed for each, primary data in research covers the full practical framework.

The Practical Selection Guide

The four classifications combine to produce a selection framework for specific research question types.

"Why do consumers behave this way / what meanings do they attach to this?" → Interpretivist, inductive, exploratory, qualitative methodology.

"How many consumers hold this attitude / what is the distribution of preferences?" → Positivist, deductive, descriptive, quantitative methodology.

"Did this specific brand action cause this specific behaviour change?" → Positivist, deductive, explanatory methodology with an experimental design.

"We need both motivational depth and measurement breadth." → Pragmatist, abductive, mixed-method methodology.

The practical application of this framework: before selecting a methodology or briefing a research supplier, identify which type the question actually requires across all four dimensions. The most common methodology selection failure is not choosing a bad method — it is choosing a familiar method that answers a similar question rather than the specific methodology the actual question requires.

Frequently Asked Questions

What are the main types of research methodology?

Research methodology is classified four ways: by research philosophy (positivism, interpretivism, pragmatism), by logical approach (deductive, inductive, abductive), by research design type (exploratory, descriptive, explanatory), and by data type (qualitative, quantitative, mixed method). Each classification captures a different dimension of what the methodology can and cannot produce.

What is the difference between positivism and interpretivism in research methodology?

Positivism assumes reality is objective and measurable, and is associated with quantitative methods designed to produce generalisable findings. Interpretivism assumes reality is socially constructed and understood differently by different people, and is associated with qualitative methods designed to surface meanings and motivations. Most commercial brand research sits closer to positivism for measurement questions and interpretivism for understanding questions.

What is the difference between deductive and inductive research?

Deductive research starts with a theory or hypothesis and tests it against data — it moves from theory to data. Inductive research starts with data and works toward patterns and theoretical propositions — it moves from data to theory. Most quantitative brand research is deductive (testing hypotheses). Most qualitative brand research is inductive (generating insight).

What is mixed-method research methodology?

A research programme that combines qualitative and quantitative data types, using each to compensate for the limitations of the other. Qualitative provides depth and motivational insight; quantitative provides measurement precision and generalisability. The combination produces more complete consumer understanding than either approach alone, at the cost of higher complexity and research investment.

When is exploratory research methodology appropriate?

When the research team does not have sufficient understanding of the topic to design hypothesis-testing or descriptive research — when the relevant variables are not yet well enough defined, the key questions are not yet clear enough to measure, or the territory is unfamiliar enough that the research needs to surface what deserves to be investigated rather than measure what is already expected.

How do you choose the right research methodology type?

By working through the four classification dimensions in order: what is the philosophical foundation the question requires, what is the logical direction (testing a hypothesis or generating one), what design type is appropriate (exploratory, descriptive, or explanatory), and what data type will produce the findings the decision requires. The most common failure is choosing a familiar method that addresses a similar question rather than the specific methodology the actual question demands.

The research methodology decision is most consequential when it is made before the first question is written — at the brief stage, where the commitment to a specific philosophical foundation and design orientation shapes everything that follows. Pulse AI Research supports the full range of research methodology types for Indian consumer categories — from qualitative exploratory studies to quantitative descriptive tracking to experimental explanatory designs.

Related reads: Research Methodology: The Complete Guide for Brand Teams | Research Design in Research Methodology: The Structural Framework | Qualitative Research Methodology: When Depth Beats Scale