Short answer
Market research is usually classified in three ways: by source (primary or secondary), by data type (qualitative or quantitative), and by research design (exploratory, descriptive or causal). A single study can sit in all three groups at once. For example, a customer interview is primary, qualitative and often exploratory. The best choice depends on the question you need answered, not on which method is most popular.
Market research and market survey: how they differ
Market research is the broad discipline of gathering and analyzing information about customers, competitors and markets. A market survey is one tool within it: a structured questionnaire sent to a sample of people. Other tools include interviews, focus groups, observation and analysis of existing data. So a market survey is always a form of market research, but market research is not limited to surveys.
Primary and secondary research
- Primary research collects new information for your specific question. Common methods include surveys, interviews, focus groups and observation. You control the questions and the sample, but it usually takes more time and budget.
- Secondary research uses information that already exists, such as government statistics, industry reports, academic studies and a company’s own sales records. It is usually faster and cheaper, but the data was not collected with your question in mind.
Many projects start with secondary research to learn what is already known, then use primary research to fill the gaps. Secondary market research examples shows what this looks like in practice.
Qualitative and quantitative research
- Qualitative research works with non-numerical information, such as words, images and behaviors, to understand why people think or act as they do. Interviews, focus groups and observational studies are common qualitative methods. Focus groups and in-depth interviews are good examples.
- Quantitative research works with numbers to measure how many, how much or how often. Surveys with closed questions and analysis of transaction data are typical examples. Many of these methods now run online, and online market research methods explains the options.
The two approaches often work best together. Qualitative work can suggest what to measure, and quantitative work can show how widespread a pattern is.
Exploratory, descriptive and causal research
- Exploratory research is used when the problem is loosely defined. It helps you scope a new market, spot possible opportunities or find the right questions to ask. Interviews and focus groups are often used here.
- Descriptive research paints a clear picture of an existing market, customer group or behavior. It answers questions such as who buys, how often, and what they prefer. Surveys, panels and customer data analysis are typical methods. Consumer panels are one example.
- Causal research tests whether one thing causes another, for example whether a price change or a new ad leads to more sales. It usually involves changing one variable while holding others steady, such as a controlled experiment or a split test. Causal designs need careful planning, because results can be misleading when the groups being compared differ in other ways.
Exploratory research tends to come first, descriptive research tells you what is happening, and causal research tests what drives it.
Common methods and where they fit
| Method | Usual data type | Usual design | Typical use |
|---|---|---|---|
| Customer survey | Quantitative (may include open text) | Descriptive | Measure attitudes, preferences and satisfaction across many people |
| In-depth interview | Qualitative | Exploratory | Understand motivations and decisions in detail |
| Focus group | Qualitative | Exploratory | React to concepts, packaging or messages in a group |
| Observation | Qualitative or quantitative | Descriptive | See what people actually do |
| Consumer panel | Quantitative | Descriptive | Track purchases or behavior over time |
| Secondary data review | Either | Exploratory or descriptive | Size a market and find existing benchmarks |
| Split test or experiment | Quantitative | Causal | Compare two versions to see which performs better |
Each row is a typical pairing, not a rule. An interview can be used for descriptive questions, and a survey can include exploratory open-ended items.
Specific research areas
Beyond the broad categories, research is often grouped by business question:
- Market segmentation divides a broad audience into groups with shared needs or behaviors.
- Brand research looks at awareness, perception and loyalty.
- Product research tests features, prototypes and fit with customer needs.
- Pricing research explores how customers judge value and what they are willing to pay.
- Competitive analysis compares rivals’ offers, positioning and strengths.
Each area can use any of the methods above. The question determines the method, not the reverse.
How to choose a research type
- Write the decision you need to make. A clear decision keeps the study focused.
- Check what already exists. Start with secondary sources so you do not repeat work.
- Decide what kind of answer you need. Use qualitative methods to explore why, quantitative methods to measure how many, and causal designs to test what works.
- Pick a sample that fits the question. Existing customers, potential customers and broader market samples answer different questions.
- Plan how you will act on the results. Research is only useful if it changes a decision.
For a practical guide to the online side of these steps, see online market research methods. For how researchers frame their questions before choosing a method, see marketing research questions and hypotheses.
What is primary research?
Primary research is any research where you collect the data yourself for the question at hand. A survey you run on your own customers, a set of interviews with buyers, or a focus group you organize are all primary research. It gives you the most relevant data, but it also takes the most planning and resources. The alternative, using data that others have collected, is secondary research.
A related term is observational research, which records behavior directly and is often used when what people do matters more than what they say.