What Is a Market Research Survey? Types and Examples
What a market research survey is, the main types and 15 questions you can adapt. Follow a four-question example to connect answers with a clear decision.
A market research survey asks a defined audience the same questions about needs, opinions, understanding or reported behavior. Closed questions produce answers you can count; open questions help explain them. This guide covers survey types, 15 questions you can adapt and a four-question example that keeps each result tied to its answer count.
Book a demo to explore simulated answers to your offer and questions. See our comparisons with human research and ask for validation that fits the audience, wording and result your decision needs.
What a survey can tell you
Closed questions give comparable categories and ratings. Report the counts, spread of answers and how many people answered each question. Open questions help explain those numbers. Who answers, how they were recruited and whether responses are human or simulated determine which claims you can make.
Study choice | What to specify |
|---|---|
Purpose | Learn what a defined audience needs, understands or prefers using consistent questions. |
Design | Define eligibility, recruitment, the exact offer, question order and valid answering bases. |
Question order | Collect unaided understanding before ratings or answer choices; report usefulness separately. |
Evidence limit | Identify human or modeled responses. A stated buying answer is not an observed purchase. |
Choose a survey type
Choose questions for a specific business decision. Check who will answer, what they will see and whether their responses can change that decision.
Exploratory surveys describe needs and occasions. Concept and message surveys compare reactions to complete offers. Tracking surveys repeat the same measures over time. Saying you would buy something is different from making a purchase.
Example: 15 questions for a grocery-delivery survey
This hypothetical example follows a grocery-delivery team choosing wording for a two-hour window. Changes close the evening before delivery. Six teaching records show why a usefulness rating needs to be read alongside the deadline each buyer understood.
One buyer gives a high usefulness rating while thinking changes are allowed during the slot. Approving the wording from that rating would repeat the wrong promise. The survey should show the writer what to change and retest before promoting the offer.
Use these 15 editable templates as a menu, not a requirement to ask every question. Choose items that affect your decision, test them first and save the answer formats and routing. They are proposed questions. The six-record example later uses the four-question core, not this expanded menu.
For the offer questions, show this fixed card: “Select a two-hour delivery window at checkout. Changes are allowed until the evening before delivery.” Record which card version each person saw. Ask items 6–8 without prompts before the offer ratings and any choices or prompts that disclose the deadline; ask the recent-experience items before showing the offer.
Editable question template | Response format and use |
|---|---|
1. Have you received a grocery delivery in the past 30 days? | Single choice: Yes; No; Prefer not to answer. Continue Yes only. Establish adult and service-area eligibility separately before this item. |
2. When did you most recently receive a grocery delivery? | Date response, with Cannot recall and Prefer not to answer available. Retain those states separately from a missing answer. |
3. How did you choose the delivery time for that most recent order? | Open text. Ask about the actual recent episode before presenting the new offer. |
4. What, if anything, made the timing of that delivery difficult for your household? | Open text. Nothing was difficult is a valid answer; do not assume a problem occurred. |
5. Which alternatives, if any, did you seriously consider instead of receiving that grocery delivery? | Select all that apply: Collect groceries from a store; Shop in person; Use another delivery service; Ask someone else to shop; Delay the order; Another option, please describe; No alternative considered; Cannot recall. Keep the last two exclusive. |
6. Under this offer, what would you be able to choose at checkout? | Open text after exposure to the fixed card. Collect before any answer choices or explanation of the offer. |
7. When can you last change the delivery window? | Open text after the same card. Ask before supplying cutoff choices, corrective feedback or the deadline-fit item below. |
8. What do you understand a two-hour delivery window to mean for this order? | Open text before explanation. Keep the original answer so a timing misunderstanding is distinct from dislike of the arrangement. |
9. How useful would this arrangement be for your household? | Single choice: 1 Not at all useful; 2 Slightly useful; 3 Moderately useful; 4 Very useful; 5 Extremely useful; Cannot judge. Keep Cannot judge outside substantive ratings. |
10. How well would choosing a two-hour window at checkout fit your household's usual schedule? | Single choice: Very poorly; Somewhat poorly; Neither well nor poorly; Somewhat well; Very well; Cannot judge. Use only if schedule fit changes a decision. |
11. How well would an evening-before deadline for changes fit your household's schedule? | Single choice: Very poorly; Somewhat poorly; Neither well nor poorly; Somewhat well; Very well; Cannot judge. This reveals the stated deadline, so place it after unaided comprehension. |
12. Compared with your current grocery-delivery arrangement, how suitable would this offer be for your household? | Single choice: More suitable; About as suitable; Less suitable; Cannot compare or judge. Keep this relative judgment separate from the usefulness scale. |
13. What, if anything, would stop you choosing this arrangement as described? | Open text. Nothing would stop me is valid. Preserve several reasons in one answer if the coding plan permits overlapping codes. |
14. What information is missing from this description? | Open text. None is valid. Do not code an explicit None as a missing response. |
15. If this arrangement were available for your next grocery order, what would you be most likely to do next? | Single choice: Choose this arrangement; Keep my current arrangement; Look for more information; Use another way to get groceries; Wait or not place an order; Cannot judge. This is stated next-step intention, not an order or conversion result. |
Human and simulated survey responses
A human survey records answers from recruited people under its stated conditions. A simulated survey generates answers for an audience built from data. Report the source, audience setup, exact questions and answer counts for each. Human sampling formulas cannot make repeated AI-generated answers a representative human sample.
Simulated responses can help explore reactions when the audience data and validation fit the task. Use transaction records to study purchases. Use direct human research when personal experience or actual use matters. The delivery example shows why you need to understand an answer before averaging ratings.
Plan the wording decision
Design choice | Record and example |
|---|---|
Target population | What to record: Geography, experience and exclusions Delivery-study example: Adults using grocery delivery in the past month in the defined service area |
What to record: Actual contactable units Delivery-study example: Eligible accounts in a stated period, with known coverage gaps | |
Recruitment | What to record: Invitations, channels and incentives Delivery-study example: Document invitation dates and reminders |
Stimulus | What to record: Exact text, image, price and conditions Delivery-study example: Same two-hour-window proposition for every respondent |
Primary measure | What to record: Question, coding and denominator Delivery-study example: Correct unaided deadline interpretation among exposed eligible records |
Comparisons | What to record: Predefined relevant groups Delivery-study example: Recent first-time and repeat users |
Quality rules | What to record: Exclusions and review evidence Delivery-study example: Duplicates, invalid routing and incomplete exposure reviewed consistently |
Limits | What to record: Likely missing experiences Delivery-study example: Nonusers and unreachable accounts outside the claim |
The team needs to decide whether to change the wording, not whether people like flexibility in general. The offer allows a two-hour window chosen at checkout, with changes until the evening before delivery. Before asking, define a correct answer: the person must name that cutoff without a prompt.
For this six-record demonstration, all six must identify the cutoff correctly for the wording to pass. That is the example’s decision rule, not a validated benchmark for a real study.
Ask for the deadline before giving hints
Select a two-hour delivery window at checkout. Changes are allowed until the evening before delivery.
When can you last change the delivery window?
Open response collected before answer choices.
R3’s answer: “I can change it during the two-hour slot.”
Comprehension: incorrect. Usefulness: 4/5.
R3 likes an offer it has misunderstood. Read the open answer beside the usefulness rating before approving the wording.
The brief calls for recent delivery users in a defined service area. Existing accounts can answer an existing-customer question but may miss former customers and people who never used the service. Decide whether you need that reachable audience or a wider market, then recruit accordingly.
Keep understanding and usefulness apart
Screen participants first, ask for reactions without prompts and then ask diagnostic choices. Ask one thing at a time. Keep rating scales in order. Vary the order of choices that have no natural order when the design calls for it. “Cannot judge” is different from the middle of an opinion scale.
For this delivery study, keep this exact card unchanged: “Select a two-hour delivery window at checkout. Changes are allowed until the evening before delivery.” Our quantitative market research guide uses the same six records; here, follow the questions and saved response path that produce them.
Check that people are adults in the service area during recruitment, before using this four-question survey. Its first question checks recent grocery-delivery use, not age or location.
Order | Question and routing |
|---|---|
1 | Exact question or action: Have you received a grocery delivery in the past 30 days? Answers and routing: Yes; No; Prefer not to answer. Continue Yes only. |
Exposure | Exact question or action: Show the fixed card above Answers and routing: Save card version and exposure completion. |
2 | Exact question or action: When can you last change the delivery window? Answers and routing: Open text; ask before supplying options or explanation. |
3 | Exact question or action: How useful would this arrangement be for your household? Answers and routing: 1 Not at all; 2 Slightly; 3 Moderately; 4 Very; 5 Extremely; Cannot judge. |
4 | Exact question or action: What information is missing from this description? Answers and routing: Open text; None is valid. |
These four questions are the whole instrument for this decision. Set the correct answer as the evening-before cutoff. Other substantive answers are incorrect; missing or unanswerable responses stay in their own categories. Ratings 4 or 5 form top-two usefulness among substantive ratings. Report Cannot judge separately.
Test question meaning first with people like the intended users. Ask them to restate “window,” “changes” and “evening before,” then explain the information used for their answer. This checks whether the questions capture understanding of the cutoff before you collect more answers to a misunderstood question.
Do not let answer choices reveal the deadline before the open question. Ask that question before any cutoff choices or explanation. Usefulness shows whether the arrangement fits; missing information can explain why someone cannot judge it. Together they can point to clearer wording, more offer detail or different service terms. Our open and closed questions guide explains the formats.
Check that the saved answers match what people saw
Pilot the survey to check that saved records match the planned experience. Try routes for people who qualify, people who do not and people who cannot judge the offer, on the intended device. Check the saved card version, open answer and rating code. Fix the problems below before collecting responses.
Pilot problem | Reason and repair |
|---|---|
A skipped card still produces a rating | Why it matters: The answer has no verified exposure Repair before fieldwork: Save exposure completion and prevent that route |
“Cannot judge” is stored as 3 | Why it matters: Unanswerability becomes moderate usefulness Repair before fieldwork: Preserve a separate code and exclude it from substantive ratings |
A participant changes the eligibility answer after exposure | Why it matters: The screening basis becomes ambiguous Repair before fieldwork: Preserve route history and apply a stated inclusion rule |
The mobile card hides the deadline | Why it matters: Device presentation changes the proposition Repair before fieldwork: Repair display and repeat the exposure rehearsal |
Keep the four-question core unless another decision needs more context. To compare first-time and repeat users, for example, define the period and get reliable history. Adding labels alone does not create a useful comparison.
Keep a version log and a reason for each excluded or unusable answer. Changing the deadline during collection changes what people see. Analyze that version separately; do not combine answers to different promises into one percentage.
Find where the survey asks too much
Every question in the short core helps decide whether to change the delivery wording. Fatigue becomes a problem when you add work the team will not act on. A long grocery-habits matrix may be easy to add, but people must finish it before reaching the important deadline question.
Question count alone will not show where the work becomes too much. The same questionnaire can send a repeat user through four screens and a first-time user through several extra explanations. Check every route, especially on a phone. Statistics Canada's questionnaire-design guidance recommends reducing response burden so people can answer accurately and testing questions with the intended respondents.
Save screen-entry and answer events beside the survey answers. The following blank check is for real fieldwork, not the six teaching records. Those records have no invitation or timing results.
Checkpoint | Count, base and useful comparison |
|---|---|
Survey starts | Unique people who open the instrument, by version, device and recruitment source. Keep invitations separate. |
Eligibility | Eligible starts divided by all starts with a resolved screening outcome. Report unresolved screens separately. |
Card and deadline question | Card-complete eligible people who reach Q2; Q2 answers divided by those who reach it. Keep incorrect answers separate from missing ones. |
Usefulness and missing information | For each item, people who reach it, substantive answers, Cannot judge or None, skipped answers and median active time. |
Eligible completions | Finished eligible records divided by eligible starts. Show the same measure by route; report partial usable answers separately. |
Breakoff | Last screen reached for an unfinished eligible start. For each screen, divide breaks there by eligible people who reached that screen. |
Save instrument_version, route_id, device, question_id, screen_entered_at, answered_at, answer_status and completed_at. Decide whether timing excludes a background tab or long interruption. Read median completion time beside how many people reached each screen and where others left. A skipped optional question is not the same as abandoning the survey.
Use one export row per route, device and question, as in the blank example. answered_n includes valid None and Cannot judge answers. item_missing_n counts people who skip the item and continue. breakoff_n counts unfinished records whose last screen is that item. Someone may save an answer and then leave, so answered_n and breakoff_n can overlap. Do not add them as mutually exclusive outcomes.
version,route_id,device,question_id,reached_n,answered_n,item_missing_n,breakoff_n,median_active_seconds
v1,first_time,mobile,Q2,,,,,
v1,first_time,mobile,Q4,,,,,
v1,repeat,mobile,Q2,,,,,
v1,repeat,mobile,Q4,,,,,
Suppose the team adds a full weekly-shopping diary between screening and the offer. Everyone who qualifies would describe seven days of shopping even though the writer only needs cutoff understanding. Keep the card and Q2 immediately after screening, along with usefulness and missing information, and remove the diary from this survey. If routine questions affect a service decision, give them a separate follow-up with a clear purpose.
This is a proposed question change, not a measured improvement in completion. Compare versions with similar recruitment, device mix and screening rules. Randomly assigning versions at the same time can help when feasible. If the versions reach different people, higher completion cannot isolate the effect of shortening. Check that the main understanding question still means the same thing.
A sudden timing change or unfinished route is a reason to check wording, mobile display, relevance or interruption. It is not proof of fraud. Remove unnecessary work while keeping the unprompted deadline question, the card people saw and the count of answers it uses.
A favorable rating can concern the wrong offer
The six records are R1: correct/5; R2: correct/4; R3: incorrect/4; R4: correct/3; R5: incorrect/2; R6: correct/Cannot judge. Correct comprehension is 4/6 (66.7%). Top-two usefulness is 3/5 (60%); one person cannot judge usefulness.
R3 says, “I can change it during the two-hour slot.” Its favorable rating describes an incorrectly understood offer. Under the example’s all-six understanding rule, the wording fails.
Put the deadline first in the next candidate: “Change your delivery window until the evening before delivery. At checkout, choose a two-hour delivery window.” Ask the understanding question again without hints under similar conditions, then test the clearer offer in a suitable study of behavior. The rewrite has no measured result yet.
Give the writer the exact error and card version. The 60% usefulness headline alone misses the problem. R3’s answer points to the cutoff, so the revision puts it before the checkout instruction. The favorable rating still matters because it shows why you must read appeal beside what the person understood.
Ask for the deadline again without hints under similar conditions. If people now understand but the arrangement fits poorly, reconsider the service terms or household timing. If they still misunderstand, check the wording and display.
Give the writer a result they can use
For each chart, show the question, number of answers and response source. Count invitations, starts, people who qualify and analyzed records separately. Keep partial answers apart from explicit “None” or “Cannot judge.” Report any weighting and exclusions.
The AAPOR disclosure standards list the questionnaire, population, recruitment and processing details needed to judge a survey. Explain the basis for any precision claim about an opt-in human sample. For a simulated study, disclose audience data and relevant validation. Generated profiles are not independent human observations.
Use our analysis guide to check coding, exclusions and the totals used for each percentage.
Sapien's team builds a simulated audience from data, runs the survey and returns an interactive report with simulated answers, audience comparisons and a recommendation. For this delivery offer, connect cutoff understanding with usefulness to tell a wording problem from service terms that do not fit. Agree the audience data and validation the decision needs.
Book a demo to discuss how the four-question survey and its simulated readout could help the writer decide what to change, with the original card, question routing and answer counts kept together.
Start the report with a revision the writer can make: put the cutoff first and retest it. Keep understanding 4/6 and high usefulness 3/5 beside the original answers and totals. Real fieldwork also needs invitation and recruitment records. This example has no invitation funnel.
Once people understand the offer, test whether it changes ordering. If they understand but reject the cutoff, clearer wording has exposed a service choice. The survey helps when it keeps that problem from being mistaken for weak marketing.
Related guides: Likert scales; sample size worksheet.
Questions about market research surveys
What is a syndicated survey?
A syndicated survey uses standard questions and licenses the results to more than one client. Check the audience, questions, field dates and permitted uses. Paying to add your own questions to a shared omnibus survey is another arrangement.
Copy these questions
These questions come from the example above. Keep the complete offer, recruitment criteria, answer options and rules for what people see when adapting them.
- Have you received a grocery delivery in the past 30 days?
- Show the fixed card above
- When can you last change the delivery window?
- How useful would this arrangement be for your household?
- What information is missing from this description?
These four questions are the whole instrument for this decision. Set the correct answer as the evening-before cutoff. Other substantive answers are incorrect; missing or unanswerable responses stay in their own categories. Ratings 4 or 5 form top-two usefulness among substantive ratings. Report Cannot judge separately.
Ask the open deadline question before any choices or explanation that reveal the answer. Usefulness shows whether the arrangement fits. Missing information can show why someone cannot judge it. Together they tell you whether to clarify words, complete the offer or change the terms.
