Market research methodologies: types, examples and how to choose

Market research methodologies fall into two broad splits. Primary research collects new information yourself, through interviews, surveys, observation or tests. Secondary research uses information someone else already collected, such as government statistics, industry reports or company filings. Separately, qualitative methods explain why and how something happens, while quantitative methods measure how many or how much.

Most useful research combines a few methods. The choice depends on the question you need answered, and on whether you need to understand a reason or count something.

Primary versus secondary research

Secondary research is usually cheaper and faster, so it is a sensible place to start. The SBA's market research guide recommends existing sources for general, countable questions like industry trends, demographics and household incomes, and direct research for questions specific to your business or customers. It lists free federal sources, including Census, Bureau of Labor Statistics and Bureau of Economic Analysis data.

Secondary research has two common gaps. It is rarely about your exact customers, and it tells you what happened without explaining why. Primary research fills those gaps but costs time and often money.

Qualitative versus quantitative research

Nielsen Norman Group's overview of research methods puts it plainly: qualitative methods are better suited to questions about why or how to fix a problem, and quantitative methods do a better job answering how many and how much. In qualitative work you observe or hear from people directly and can ask follow-up questions. In quantitative work the data is collected through an instrument, such as a survey or analytics tool, and analyzed mathematically.

The same article adds a second split that matters for market research: what people say versus what they do. Interviews and surveys capture what people report. Experiments, analytics and sales data mostly capture what they actually did. The two often differ.

The main methods and what each is for

Method	Type	Best for	Main limit
Public data and industry reports	Secondary, quantitative	Market size, trends, demographics, number of businesses	Rarely specific to your customers; no reasons
In-depth interviews	Primary, qualitative	Why people buy, switch or give up; how a process works	A handful of accounts cannot tell you how common something is
Expert or industry interviews	Primary, qualitative	How an industry buys, prices and operates	One person's view of other people's needs is secondhand
Focus groups	Primary, qualitative	Group reactions to a brand or concept	Group dynamics shape answers; weak for detailed behavior
Surveys	Primary, quantitative	Measuring how common a view or behavior is	Only as good as the sample and question wording
Observation and field visits	Primary, qualitative	Seeing the real process, including steps people forget to mention	Time-consuming; your presence can change behavior
Concept and usability tests	Primary, mixed	Whether people understand and can use an offer	Test conditions differ from real use
Experiments and pilots	Primary, quantitative	Whether people act: buy, click, sign up	Needs enough traffic or customers to read results
Your own sales and product data	Secondary to you, quantitative	What existing customers actually do	Says nothing about people who never became customers

NN/g describes focus groups as sessions of 3 to 12 participants and notes they can give a top-of-mind view of a brand or product concept in a group setting, but tend to be less useful for usability questions. Its description of surveys is a quantitative measure of attitudes, typically with more closed-ended than open-ended questions.

For surveys, AAPOR's best practices start by asking whether a survey is the right method at all, and whether existing data already answers the question. It also recommends pretesting questions with people similar to your respondents before sending the survey out.

A worked example: choosing methods for one decision

Suppose you run a small skincare brand and are deciding whether to launch a refill program. This is a hypothetical example.

Secondary research comes first. Look for any rules on refillable cosmetic packaging where you sell, and check what similar brands offer publicly.

Next, interview six to eight current customers about the last time they ran out of a product and what they did. You learn whether refills solve a real annoyance or only sound appealing. You might also speak with someone who has run packaging or operations for a beauty brand, since they can explain costs and returns logistics your customers cannot. The beauty and personal care product professionals directory page is one place to start.

If interviews point to a clear reason, a short survey of your customer list can estimate how many share it. Finally, a pilot with one product and a refill option on your site shows what people do when asked to pay.

Each method answers a different question. Interviews found the reason, the survey sized it within your customer base, and the pilot tested behavior.

How to pick a method

Write the question you need answered in one sentence, then ask:

Has someone already collected this? If so, use secondary research.
Do I need to understand a reason or a process? Use interviews or observation.
Do I need to know how common something is? Use a survey or existing data, with a sample that matches the people you care about.
Do I need to know whether people will act? Use a pilot, preorder or experiment.

Interviews versus surveys goes deeper on the most common choice, and combining interviews with product data covers how to use methods together.

When the method you need is a conversation with someone outside your network, Instant Expert can find people who fit a description you write. You review them, send invitations and pay only for calls that get booked.