Market Research Questions: Connect Wording to Decisions

Market research questions help you learn about the people, choices, and conditions behind a business decision. A useful question has a clear purpose, reaches someone who can answer, and leaves room for an answer you did not expect.

A long list of questions cannot make up for a vague decision.

Start with what the decision maker needs to know. Then separate what earlier sources already tell you from what you still need to ask. The examples below show that process for a fictional lunch service. They are drafts to review and test, rather than a ready-to-field survey.

Separate the decision from respondent questions

Keep three levels distinct:

• Business decision: what you may do.
• Research question: what you need to learn.
• Respondent question: an item a person can answer about that gap.

Mixing these levels can turn a survey into a request for approval of your plan. For example, a decision might be whether to trial weekday lunch pickup near an office district. The research question asks how workers currently get lunch and what limits their choices.

One respondent question asks where they got lunch on their most recent day working in that district. That item supplies a piece of evidence; it cannot settle the entire launch decision.

Qualtrics's market question guide starts with goals before question types. Turn your own goal into specific research objectives before borrowing items from a question bank.

Also check whether asking people is the right way to fill the gap. The excerpt below raises that method-choice question before any wording advice.

AAPOR's March 2022 guide, page 3, puts method choice first. Existing records or another method may answer some parts of the decision.

Define who can answer each question

Write the target audience in concrete terms. For the lunch example, people who work in the district on weekdays differ from residents who rarely visit it. Existing customers, people who tried a rival, and people who bring food from home may also have different experiences to report.

Check that each item fits the people receiving it. Ask a non-user what they currently do before asking why they stopped using a service they may never have tried.

Route questions about a recent order only to people who placed one. A screening question should establish eligibility without teaching people which answer gets them into the study.

Attest's practitioner guide distinguishes audience profiling from behavior within a category. Use that distinction to choose relevant questions, rather than collect every personal detail a template offers.

For business buyers, the person using a service may differ from the person approving its budget. If the question concerns a completed deal, win-loss analysis helps examine that choice through buyer accounts. Prospective questions and completed-deal accounts support different claims.

Question examples for different decisions

Choose examples by the gap they address. The SurveyMonkey question bank groups items across customer, product, and market topics. The following teaching examples add a purpose and a limit so you can adapt them without treating the bank as one survey.

Current behavior

Ask, “On your most recent workday in this district, where did you get lunch?” Follow up on the time, route, or choice that matters to the decision. A recent occasion gives people something concrete to recall, though memory and unusual days still need checking.

Needs and constraints

Ask, “What, if anything, made getting lunch difficult that day?” The phrase “if anything” allows the person to report no difficulty.

Do not assume that everyone has a problem your offer can solve. For a deeper account of the situation and desired progress, use jobs-to-be-done research.

Alternatives

Ask, “What other ways of getting lunch did you consider that day?” Include doing nothing or bringing food from home when those are plausible. A competitor list that includes only paid services can hide the choice you most need to understand. Attest's category examples help frame this broader scope.

Concept response

Show the same clear description of a proposed service before asking what is easy or hard to understand. Then ask which part, if any, would fit their routine. A favorable answer describes a reaction to that stated offer, not proof that the person will use it.

If you ask people to rate the concept between two words, check both anchors. The semantic differential scale guide shows how to review the word pair, rated object and midpoint before borrowing a published item for your market study.

Price and conditions

Ask what they paid for the relevant recent purchase, where appropriate, or compare responses to a clearly priced offer. State what the price includes.

“Would you buy?” without a price, location, or service conditions can mean different things to different people. Qualtrics's product and price examples are starting points that still need those details.

Channels and awareness

Ask where they looked for lunch options before showing a list of brands. If unaided awareness matters, let them name options first. Showing names changes what they can recall. Formbricks's category guide provides channel examples, but you must adapt the choices to your market.

A worked decision-to-question map

Consider a fictional team deciding whether to trial lunch pickup in one office district. Its planning packet has a prior survey P, a draft concept C, and an interview memo M.

The documents and page locators below are invented for teaching; they contain no real customer findings.

The map shows how the team could revise four questions. Read the last column before borrowing the wording: a cleaner item still has an evidence gap to resolve.

Table 1: These four revisions belong to a fictional planning exercise. They do not show that any question has passed a pretest or that the service has demand.

The first row changes the assumed story. “Better delivery” treats delivery as the relevant option for everyone and assumes dissatisfaction.

Asking about a recent lunch opens space for other routines. The prior survey can inform the planning gap, but its delivery-only scope cannot describe all workers.

The last row also changes the claim you can make. A response to a priced concept may guide further investigation. It does not tell you how many people will buy, at what rate, or for how long. Keep that limit beside the question rather than hide it in a report footnote.

Check wording and response options

Review the stem

Read each item as a person who knows nothing about your project. Remove praise such as “convenient” unless the question explicitly measures reactions to that claim. Split price and service when someone could have different views of each.

Define the place and time frame, and avoid asking people to reconstruct details they are unlikely to remember. Pew's wording guidance explains how language and combined questions can change what people report.

The effect is more than a stylistic concern. In a January 2019 survey, Pew randomly assigned people to questions about available “jobs” or “good jobs.” The share reporting plenty available was 60% for jobs and 48% for good jobs.

Pew's original account shows a wording effect in that study; its size does not predict the effect of a lunch question.

Check the choices and order

For closed questions, read the offered answers as carefully as the stem. Someone who brought lunch needs an option that fits. Frequency ranges should not overlap at their boundaries.

For open questions, plan how you will read the full answers before fieldwork. The qualitative survey analysis guide shows how to retain each question, response and exception when interpreting written answers.

Decide whether respondents choose one answer or several, and consider “other,” “none,” or “don't know” where each has a real role. AAPOR's response-choice guidance calls for choices that cover reasonable answers without overlapping.

Review the sequence too. Asking about named services before a broad question about lunch choices can bring those services to mind. Pew's question-order guidance describes that risk. Keep the broad inquiry ahead of details when that fits the purpose, and test the full sequence.

Test the questions before fieldwork

A desk review can catch assumptions, but you also need to learn how people understand the question. Ask people similar to the planned audience to explain what the draft item means to them and how they chose an answer. This is the thought process that AAPOR recommends checking through pretesting.

Keep a revision record. If “workday here” means a day in the office to one person and any weekday to another, resolve the wording before collection. Then test the revised version. Do not write “validated” beside an item merely because a colleague found it clear.

A procedure pilot checks a different problem: whether recruitment, routing, survey delivery, and data handling work together. AAPOR also recommends that check.

Clear wording cannot repair a sample that misses an important part of the audience. Review those wider market research challenges before deciding what the results will support.

The researcher remains responsible for access, consent, respondent burden, collection, and analysis. No fixed number of questions or test participants guarantees a sound study.

Review source gaps in Atlas

Atlas can help compare selected prior research and draft questions before fieldwork. For the teaching example, the task is to check what P, C, and M support and where new evidence is needed. Keep their fictional status explicit if you use them to practice.

1. Add permitted PDFs through Add a source and Upload files. Check processing and open each source to confirm its content is readable. Include the decision brief and relevant earlier research rather than an unrelated full archive.
2. In chat, mention the selected sources with @. Ask: “For the weekday lunch-pickup decision, compare the prior survey, concept and interview memo. For each information gap, propose one neutral question, name who can answer, cite the source behind the gap, and state what remains unknown. Do not infer demand from purchase intent.”
3. Open the numbered citations and read the surrounding passages. If a suggestion says the prior survey establishes all workers' routines, inspect P page 2. In this fictional packet it covered delivery users only. Correct the scope and retain non-users as a group to reach; do not turn a narrow source into a market claim.
4. Review the final wording and answer choices yourself. Create a note through New and Note, save the checked questions with source locators and unresolved gaps, and wait for Saved. Carry that note into your study review.

Treat the answer as a proposal to check. Atlas cannot recruit respondents, field the survey, verify responses, or certify your sample and question design. Citation inspection checks a source claim; it does not replace a respondent pretest.

Keep a checked question handoff

The final handoff should state the decision, eligible audience, exact wording, answer choices, and any routing rules. Include the reason for each item, its prior-source support, the checks completed, and the gaps still open. Keep the version you will field so later comparisons use the right wording.

Attach the questions to a market research plan that names collection and analysis responsibilities. Drop an item if you cannot explain how its answer helps the decision or what you will do with it. A concise, checked set gives the next researcher a clearer starting point than an unreviewed bank.