Market research is the process of gathering and examining information about customers, competing options, and a market so you can make a better business decision. It can help you decide whether to develop an offer, who it should serve, how to reach customers, or what to improve.
You do not need to start with a large study. Start with a decision you need to make, gather evidence that can inform it, and keep track of what you still do not know. This guide explains the main methods and gives you a plan you can use for a new or established business.
What do you mean by market research?
Market research helps a business understand the people it wants to serve and the choices available to them. Depending on the question, it might include speaking with customers, reviewing existing industry data, comparing alternatives, or testing a proposed offer.
For example, a local retailer deciding whether to offer delivery might look at customer requests, interview shoppers, and review competing delivery options. A software business might study how teams solve a problem today before building a new feature.
The U.S. Small Business Administration's market research guide connects customer research with competitive analysis. Both can inform a decision, but neither turns an assumption into a fact on its own.
Primary research vs. secondary research
Primary research is information you collect for your own question, such as interviews, surveys, or an observation of how customers use a service. Secondary research uses information that already exists, such as public statistics, industry reports, and published company information.
| Method | Example | Useful for | Watch for |
|---|---|---|---|
| Customer interviews | Ask people about a recent experience. | Understanding what happened and why. | A few conversations can reveal useful questions, but may not represent the entire market. |
| Surveys | Ask a defined group the same questions. | Comparing responses across participants. | Results depend on who responds, how questions are worded, and how participants were selected. |
| Observation | With permission, watch how a task is performed. | Seeing steps and workarounds people may not mention. | A participant's behavior in one setting may not represent every setting. |
| Public datasets | Review relevant government data by geography or industry. | Understanding an available population or business category. | Check what the dataset counts, when it was collected, and what it excludes. |
| Competitor and alternative review | Compare official product or service information. | Understanding choices a customer may consider. | A public feature list alone does not prove which option works best for a customer. |
The distinction between newly collected and existing information is a useful starting point, not a measure of quality by itself. Business Queensland's research guidance describes both approaches and suggests choosing methods based on the information needed.
You may also hear qualitative and quantitative research. Qualitative work explores experiences, reasons, and context. Quantitative work uses numerical observations to describe or compare what was measured. A strong project can use both, provided the method fits the question.
How do you do market research for beginners?
1. Choose the decision first
Write the choice you are trying to make. “Should we test a delivery option with existing customers?” is more actionable than “Research our market.” A clear decision makes it easier to choose what to learn and when to stop gathering information.
2. Define whose experience matters
Describe the people or organizations relevant to the decision. Consider who uses the product, who decides whether to buy, and who may be affected by a change. Those roles can belong to the same person or different people.
Avoid assuming that your existing customers represent everyone you hope to reach. If you speak only with enthusiastic supporters, make that limitation visible.
3. List the questions and what you already know
Write down what you believe, the evidence behind it, and the questions that remain. For a delivery test, useful questions might include how customers shop today, when delivery would be helpful, and what alternatives they already use.
Distinguish observations from interpretations. “Three customers asked about delivery” is an observation. “Most customers will pay for delivery” is a separate claim that needs more evidence.
4. Choose a few suitable methods
Use existing sources to understand the landscape, then collect new information where it can answer your question. For a local business, that could mean reviewing public demographic data, examining local alternatives, and having a small number of customer conversations.
If you use U.S. business counts, Census Bureau Statistics of U.S. Businesses provides data for establishments with paid employees by industry, geography, and enterprise size. Check the specific dataset's coverage and date before treating a count as your potential customer base.
5. Collect information without leading the answer
Ask people about a real experience before introducing your preferred solution. “Tell me about the last time you needed this” can teach you more than “Wouldn't this feature be useful?” Ask permission before recording a conversation or retaining sensitive material.
For customer conversations, use our customer discovery question worksheet to go deeper on recent behavior, alternatives, and follow-up questions.
6. Compare evidence and decide what to do next
Look for patterns, disagreements, and gaps. Record the source and context for each important finding. If the evidence is limited or contradictory, say so. Your next decision might be to run a small test, explore a different audience, or postpone the idea.
Do not treat a positive interview as proof of demand, or a single survey response as a market-wide forecast. The objective is to make a better next decision, not manufacture certainty.
Market research example: testing a delivery idea
This example is fictional. It illustrates how a business could organize its research, not a customer result or a statistically representative study.
An independent home-goods shop is considering local delivery for larger items. The owner wants to know whether a small test is worth the operating effort.
| Research input | What the shop learns | What it does not prove |
|---|---|---|
| Existing customer requests | Several customers have asked whether large purchases can be delivered. | How many customers in the wider market would use or pay for delivery. |
| Five customer conversations | Some shoppers value convenience; others prefer to collect the same day. | That the conversations represent all current or potential shoppers. |
| Review of local alternatives | Some nearby stores describe delivery options on their websites. | Whether those services are profitable or satisfy customers. |
| Internal cost estimate | The team can estimate time and vehicle costs for a limited test. | Actual costs or demand until the test is run. |
The owner decides to test delivery for a small selection of products and track requests, fulfilled orders, and actual costs. After the test, the team can decide whether to continue, change the offer, or stop. The numbers and outcome are hypothetical, not a promise that this approach will succeed.
Copy the market research worksheet
MARKET RESEARCH PLAN
Decision to make: [what we need to decide]
Audience: [whose experience or buying decision matters]
Current hypothesis: [what we believe, not yet established]
What we already know:
- [observation] | [source and date]
Questions to answer:
- [customer, market, or alternative question]
Existing sources to review:
- [dataset, report, or official source] | [coverage and date]
New research to conduct:
- [method] | [participants or setting] | [permission and owner]
Findings:
- Observation: [what the evidence actually says]
- Interpretation: [what we think it may mean]
- Limitation or contradiction: [what remains uncertain]
Next decision: [test, revise, pause, or gather more evidence]
Owner and review date: [who will decide and when]
Is market research being replaced by AI?
AI can help organize research notes, identify questions to investigate, or summarize information you are authorized to use. It does not remove the need to check sources, speak with relevant people, or decide whether the evidence is strong enough for the question.
Treat an AI-generated summary as a starting point for review. Verify important claims against the source, preserve the difference between a participant's statement and an interpretation, and avoid treating invented examples as customer evidence.
Prompt to try with Aurora, Claude, or Codex
Help me organize the market research below for a decision.
Separate observations from interpretations and identify the source for each.
List contradictions, missing information, and possible sampling limitations.
Suggest three useful follow-up questions and one small next test.
Do not invent customer quotes, statistics, willingness to pay, or conclusions.
Decision and authorized notes: [information I am permitted to share]
Keep the people and evidence connected
Market research often starts a series of conversations with customers, partners, or experts. Keeping the relationship and follow-up context together can make it easier to return to those conversations thoughtfully.
Finta CRM brings relationship history, pipeline stages, and supported inbox context into one view. Aurora can help prepare a reviewed next step from available context. Use the research methods and data tools appropriate to your project; Finta is a place to keep relevant relationship work organized, not a substitute for a survey platform or a representative research sample.