Survey Questions to Ask
Example survey questions to ask, for anyone writing a survey: a small business owner, an HR lead, an event organizer, a teacher or a student researcher. The list is grouped by kind of survey (customer satisfaction, product feedback, employee, event and school), with the demographic and closing questions last, where they belong on the form. Each question has a note on what it measures, which answer format suits it (a scale, multiple choice or open text) and the wording trap to avoid.
Want questions from the whole vault instead? Try the random question generator.
The questions
Each question, and why to ask it
Customer
Overall, how satisfied are you with your experience with us?
Why ask it
This is the standard customer satisfaction measure, answered on a five-point scale from very dissatisfied to very satisfied with a label on every point. Put it first, before detailed questions remind people of a late delivery or a kind cashier and pull the overall score along with them. Avoid 'How happy were you with our friendly service?', which hands the respondent the answer.
How likely are you to recommend us to a friend or colleague?
Why ask it
Asked on a scale from 0 to 10, it measures loyalty more than satisfaction with a single visit, and the usual reading counts 9 and 10 as promoters and 0 to 6 as detractors. Keep the standard wording so this year's score can be set against last year's. Skip it where recommending makes little sense, such as a service people use once and would sooner not talk about.
What is the main reason for the score you gave?
Why ask it
The answers to this one are where the fixes come from, so give it an open text box directly under a rating. Asking for the main reason keeps replies short enough to sort into themes by hand. Leave it optional: a forced comment turns into a single letter typed to get past the box.
How easy was it to get what you needed from us today?
Why ask it
This is a customer effort question, answered on a five-point scale from very difficult to very easy, and it suits the moment after a support call, a return or a checkout. Low scores point at the process more than the people, so follow with a question about where they got stuck. Keep 'today' or name the visit, or people will average a year of dealings with you.
Was your problem solved by the end of your contact with us?
Why ask it
Yes, no and partly are the only options needed, and they record the outcome where a satisfaction score records the mood. Read the two together: a pleasant agent and an unsolved problem shows up as a high rating and a 'no' here. Do not ask 'Did our team solve your problem quickly and politely?', which is three questions sharing one answer.
How would you rate the person who helped you on each of the following: knowledge, courtesy, speed?
Why ask it
Set it up as a grid, one row per quality, with the same poor-to-excellent scale across the top. Splitting the rows is the point: 'Was our staff friendly and knowledgeable?' is double-barreled, and a rude expert leaves the respondent with no honest answer. Stop at four or five rows, because long grids get the same column ticked all the way down.
How long did you wait before someone helped you?
Why ask it
A factual question with time ranges as options: under 5 minutes, 5 to 15 minutes, more than 15. It records what happened, where 'Was the wait acceptable?' records patience, and you need the first to set a target. Check that no two ranges share an edge, so nobody who waited exactly 5 minutes has two boxes to choose from.
How satisfied are you with the result of the work we did?
Why ask it
For a repair, a haircut, a tax return or any job with an outcome, ask about the result separately from how the customer was treated. A satisfaction scale works, and the two scores often disagree. Each points somewhere different: a poor result at the skill or the materials, poor treatment at how the job was explained and scheduled.
Was the work finished by the date we gave you?
Why ask it
Offer yes, no and 'I was not given a date'. A fact like this is steadier than a rating, because one person's 'good' is another person's 'fine'. If many people choose the third option, the fix belongs at the quoting stage and not with whoever did the job.
How would you rate the value you get for the price you pay?
Why ask it
The word 'value' asks about the trade, not about whether the price is low, and a scale from very poor to excellent suits it. 'Are our prices fair?' invites a yes or a no, and few people will tell a seller that the price is too low. Strong satisfaction beside weak value means customers like you and may still leave for a cheaper offer.
How well did what you received match what you expected?
Why ask it
Offer a scale that runs from much worse than expected to much better, with 'about what I expected' in the middle. Answers piled on the worse side while the product itself rates well point at the description, the photos or the sales pitch. It is fairer than 'Were you disappointed?', which allows for only one direction.
How did you hear about us?
Why ask it
Make it multiple choice with an 'Other' line and one answer only, and shuffle the options if your survey tool allows it, since whatever sits at the top of a list tends to collect extra clicks. You are finding out which channel people remember, which is not always the one that brought them. A large 'Other' pile means your list is missing something, so read what people typed there.
How likely are you to buy from us again?
Why ask it
What this records is an intention, on a five-point scale from very unlikely to very likely, and an intention is not an order. Set the answers beside real repeat purchases where you have them. When intent is high and repeat business is low, look at what happens between the survey and the next time the customer needs you.
What is the one thing we should change to serve you better?
Why ask it
The words 'the one thing' do the work here: they get a priority, where 'Do you have any suggestions?' gets a no. Give it an open text box near the end, once the ratings have reminded people of the details. The same complaint from ten customers in different words is a finding; one long angry answer is a customer to call, if they left a name.
Product
How often do you use our product?
Why ask it
Give options people can count: every day, a few times a week, about once a month, less often. Words such as 'regularly' and 'sometimes' mean different things to different people and cannot be compared. Use the answer mostly as a filter, since feedback from daily users and from people who opened it twice deserves different weight.
How would you rate the quality of our product?
Why ask it
Ask this on a five-point scale from very poor to excellent, before the questions about single features. 'Quality' means durability to one buyer and finish to another, so a low score needs a follow-up asking what fell short. Keep delivery and price in their own questions, or a late parcel gets marked down here as a bad product.
What do you mainly use our product to do?
Why ask it
Start with open text while the product is young, then switch to multiple choice once the same five or six answers keep coming back. It measures the job the product is doing for people, which is often narrower than the feature list. An answer you did not expect is worth a conversation with that customer before you build anything.
Which of these features do you use most? Select up to three.
Why ask it
Use checkboxes with a limit. Without the cap, people tick everything they have ever opened and the result is a flat line. When a feature the team is proud of is picked by almost nobody, it needs explaining better or retiring, and usage records will tell you which if you have them.
How easy is our product to use?
Why ask it
Ease and usefulness need a question each: 'How easy and useful is our product?' is double-barreled, and plenty of tools are one and not the other. This one takes five points, from very difficult to very easy. For anyone who answers on the difficult side, add a follow-up asking what they were trying to do at the time.
How well does our product meet your needs?
Why ask it
A scale from not at all well to extremely well measures fit, which is a different thing from quality: a well-made product can still be the wrong one for the person holding it. Poor fit among recent buyers points at how it is being sold. Poor fit among long-standing users suggests their needs moved and the product stayed where it was.
How would you feel if you could no longer use our product?
Why ask it
Three options are enough: very disappointed, somewhat disappointed, not disappointed. It measures how much people depend on the product, which satisfaction scores can hide, because people are often satisfied with things they would not miss. Read the other answers of the 'very disappointed' group closely, since those are the customers you want more of.
Before you started using our product, what did you use instead?
Why ask it
List the main alternatives as multiple choice, plus 'nothing' and an 'Other' line. Any wording of this question presumes there was something before, and the 'nothing' option is what keeps first-time buyers from guessing. The results show what you are really compared with, which is sometimes pen and paper and not a rival company.
What is missing from our product that you expected to find?
Why ask it
Give it an open text box. The answers show the gap between what was promised or assumed and what arrived, so they divide into things to build and expectations to correct. 'What features should we add?' gets a wish list instead, with no sign of what anyone would use.
Which one of these improvements would matter most to you?
Why ask it
Make it a single choice from a short list of things you could really build. Asking 'Would you use this?' about each idea in turn gets a polite yes every time; making people choose one shows the order. Add 'none of these', so that a list of unwanted ideas cannot produce a winner.
At what price would this product be too expensive for you to consider?
Why ask it
The answer is a number typed into a box. Nothing is at stake when people answer a hypothetical price question, so treat the figures as a rough ceiling and ask only people who already know the product. 'How much would you pay?' does worse: it tends to collect the lowest figure each person thinks they could get away with.
Employee
How satisfied are you with your job overall?
Why ask it
Use five points from very dissatisfied to very satisfied, and word it the same way every round so that one result can be set against the last. It is a headline number and explains nothing alone; the questions after it say why. Run it only where answers cannot be traced to a person, or people will rate the job the way they would with their manager in the room.
How likely are you to recommend this organization as a place to work?
Why ask it
The 0 to 10 recommendation scale, pointed at the employer. People who like their team and doubt the company's direction split here in a way the satisfaction question misses. Pair it with an open 'why', because a bare number about a workplace gives managers nothing to act on.
I am clear about what my manager expects of me.
Why ask it
An agreement statement, answered from strongly disagree to strongly agree, with one idea per statement in plain words. Some people agree with whatever is put in front of them, so check a glowing run of statements against a question about an event, such as whether anyone thanked them in the past month. Disagreement here is among the cheapest findings to fix: it calls for a conversation, not a budget.
My manager gives me feedback that helps me do my job better.
Why ask it
'Helps me do my job better' is the part that counts: 'My manager gives me regular feedback' only asks how often, and frequent feedback nobody can use would still score well. Results by manager are the most useful cut of an employee survey and the riskiest, because on a team of four everyone can guess who said what. Report a manager's score only when enough people answered to hide each of them, and ask HR what minimum group size your organization uses.
I can see a path to grow my career at this organization.
Why ask it
This statement asks about a future the person can picture, which is a different thing from whether training is on offer, and it takes the same agreement scale. Split the result by length of service: new hires often answer on hope, and low agreement among people a few years in is the worrying version. If it comes back low, a follow-up can ask whether the missing piece is a role to move into, skills to build or a conversation nobody has had with them.
I have the equipment I need to do my job well.
Why ask it
'Tools and training' or 'equipment and support' in one statement is the double-barreled trap that turns up most in staff surveys, so ask about one thing at a time. Low agreement is the most concrete result you will get. Follow it with an open box asking what is missing, and expect answers as specific as a second monitor.
How manageable is your workload in a typical week?
Why ask it
Offer a scale from not at all manageable to completely manageable. 'Typical week' stops one bad deadline from coloring the answer; with no time frame, people report the week they are in. Look at the result team by team, because a company average hides the one group that is drowning.
I can disagree with my manager without it being held against me.
Why ask it
Whether bad news travels upward is what this statement tests, on an agreement scale with a neutral midpoint. Read the midpoint with care: someone who is unsure whether it is safe to speak has not said yes. Do not soften it to 'My manager is open to feedback', which rates the manager's intentions instead of what happens afterwards.
How fairly do you feel you are paid for the work you do?
Why ask it
The answer is a perception, not a reading of the market, on a scale from very unfairly to very fairly, and both matter: pay that is competitive and feels unfair is still a problem, of explanation if not of money. Ask only if leadership is ready to respond, since raising pay as a subject and then saying nothing tends to land worse than leaving it off.
How well does leadership explain the reasons behind major decisions?
Why ask it
Say who 'leadership' means in the survey's introduction, or frontline staff will answer about their supervisor and head office about the chief executive. A scale from not at all well to extremely well fits. Weak scores soon after a reorganization or a policy change are usually about that one decision, so explain it again before launching anything broader.
In the past month, has anyone at work thanked you for something you did well?
Why ask it
Asking about an event, yes or no inside a fixed time frame, is steadier than asking 'Do you feel recognized?', which moves with the mood of the day. A 'no' from most of one team is a specific thing to hand to its manager.
Do you expect to be working here a year from now?
Why ask it
The options are yes, no and not sure, and the 'not sure' group is the one to study, because their other answers show what would tip them either way. It belongs only on an anonymous survey, with a line beside it saying so; otherwise expect a wall of yes.
What is one thing that would make this a better place to work?
Why ask it
Tell people before they type whether comments are passed on word for word, so that nobody identifies themselves by a turn of phrase or a detail only they would know. Keep the box optional. Sort the answers into themes and share the themes, because people who see their comment reflected are the ones who answer next time.
How confident are you that something will be done with the results of this survey?
Why ask it
This is the question that explains a poor response rate, asked on a five-point scale from not at all confident to very confident. Low confidence generally means an earlier survey disappeared without a word. The fix sits outside the questionnaire: report what you heard and one thing you changed before you ask again.
Event
Overall, how would you rate the event?
Why ask it
Ask it on five points from poor to excellent, and before the questions about food and parking so those do not drag the overall score around. Send the survey within a day or two, while the details are fresh. Compare the score from one event to the next only if the wording and the scale stay the same.
What was your main reason for attending?
Why ask it
Multiple choice: to learn something, to meet people, to hear a particular speaker, my employer sent me, and an 'Other' line. Use it to split every other answer, since someone who came for the people and someone who came for the content judge the same day differently. Single choice gets you the main reason; 'select all that apply' gets every reason and no order.
Which session was most useful to you?
Why ask it
Show the real session titles in a list instead of a text box, since few people remember the exact name of a talk. It measures which content earned its slot. A session nobody picks is not always a bad one, so check attendance first: an empty room and a disappointed room are different problems.
How would you rate each of the following: registration, venue, food, schedule?
Why ask it
A grid with one scale for every row and a 'did not use' column, so that someone who skipped lunch is not made to rate it. Leave that column out and people guess, and guesses land in the middle and flatten the result. One low row is far easier to fix than a low overall score.
Was the event too short, too long, or about the right length?
Why ask it
Three options with the neutral one in the middle. This is a balanced question: it names both directions, so it does not suggest an answer the way 'Was the event too long?' does. The same shape works for the level of difficulty and the number of breaks.
How relevant was the content to your work?
Why ask it
Relevance and enjoyment come apart: a funny keynote can rate high overall and low here. The scale runs from not at all relevant to extremely relevant. If one job type scores it low, the invitation reached people the program was not built for, which is a reason to ask for job role at the end of the form.
If we hold this event again next year, how likely are you to come?
Why ask it
A likelihood scale, useful when you are deciding whether to book the room again. A 'likely' is not a registration, so discount it and look hardest at the people in the middle. For them, add an optional box asking what would make the difference.
What topic or speaker would you like to see next time?
Why ask it
This is the one question on an event survey that plans the next event, so give it an open text box. The 'or' is deliberate: some people can name a subject and not a person, and either answer helps. Expect a long tail of single requests, and act on the subjects that several people raise without prompting.
Did anything make it hard for you to take part fully?
Why ask it
Yes or no, with a text box that opens on yes. It catches what ratings miss: stairs, sound, session times, a platform that would not load, food someone could not eat. Even a single answer deserves attention, since others with the same problem may not have come at all.
School
How clear were the instructions for assignments in this class?
Why ask it
Answered from not at all clear to extremely clear. For younger students, cut the scale to three points or use faces, and read the question aloud. If clarity scores low while grades hold up, students may be working it out from each other, so ask a few of them how they knew what to do.
Is the pace of this class too slow, too fast, or about right?
Why ask it
Three options, best asked in the middle of the term, while there is still time to change something. A split result, half too slow and half too fast, is not a cue to carry on as you are. It usually means the class needs practice work at two levels.
About how many hours a week do you spend on this class outside of lessons?
Why ask it
Ranges of numbers: under 1 hour, 1 to 3, 4 to 6, 7 or more. 'About' gives permission to estimate, and an honest estimate is better than a blank. Compare the result with the workload you intended, because a wide gap in either direction should change the next assignment.
What has helped you learn the most in this class?
Why ask it
Place this open question ahead of any question about what to change, since it tells you what to protect. Students name activities, not teaching principles: the practice quizzes, the worked examples, the Tuesday review. Whatever comes up most is what to keep when time runs short.
What is one change that would help you learn more in this class?
Why ask it
Framing the change around learning steers answers toward things you can act on, where 'What do you dislike about this class?' collects complaints about the subject and the timetable. Leave it as open text, and give younger students a sentence starter such as 'I would learn more if'. Tell the class afterwards which suggestion you took up, or expect shorter answers next time.
How comfortable do you feel asking a question in front of the class?
Why ask it
The quiet students are who this question is for, and they are unlikely to answer truthfully with a name attached, so use it on an anonymous form only. The scale runs from not at all comfortable to very comfortable. Low comfort is an argument for a second route, such as a question box or exit slips.
Are you currently a student at this school?
Why ask it
A screening question for a student research project: yes or no, at the very start, with a polite exit for anyone outside the group you are studying. Without it, a link passed around online fills the data with people your conclusions do not cover. Your instructor or review board may also require a consent question before it, so ask what your project needs.
On how many of the past seven days did you eat breakfast?
Why ask it
A model for any behavior question in a research project: a fixed period, something countable, and a number from 0 to 7 for an answer. Swap in your own behavior and keep the frame. 'Do you eat breakfast regularly?' fails twice, since 'regularly' is undefined and a yes or no throws away the detail.
How much do you agree or disagree with this statement: 'Social media makes it harder for me to concentrate on schoolwork'?
Why ask it
A model attitude item on a five-point agreement scale. Write the statement so that a reasonable person could disagree, keep it to a single claim, and leave out loaded words: 'Social media is ruining students' attention' leads the respondent. For a research project, three or four statements on one idea give a steadier measure than a single item.
About you
What is your age?
Why ask it
Offer ranges unless you need exact ages: 18 to 24, 25 to 34 and so on, with no number appearing in two ranges and 'prefer not to say' at the end. Choose ranges that match whatever you will compare against, such as census figures or last year's survey. If anyone under 18 might answer, the rules on consent vary, so ask your school or organization what is required.
What is your gender?
Why ask it
Multiple choice with a self-describe option and 'prefer not to say'. Include it only if you will analyze results by gender, because a demographic question you never use costs trust and completions for nothing. Preferred wording changes over time and from place to place, so check what your organization or a recent national survey uses.
Which of these best describes your race or ethnicity? Select all that apply.
Why ask it
Categories differ from country to country, and some places have rules about whether and how this can be asked, so find out how it works where you are and borrow the category list your national census uses. 'Select all that apply' matters, because single choice forces people of mixed background to pick one. Say beside the question why you are asking and that it is optional.
What is the highest level of education you have completed?
Why ask it
Single choice, ordered from least to most, using the names qualifications go by in your own country. 'Completed' is the important word: without it, someone two years into a degree does not know which box is theirs. Add a 'currently studying' option only if that difference matters to your analysis.
What is your current employment status?
Why ask it
List the options and ask people to choose the one that fits best: employed full time, employed part time, self-employed, looking for work, student, retired, caring for family, other. 'Fits best' settles it for the person who is two of these at once. Try the list on a few people first, because the option you forgot is the first thing someone will point out.
How long have you worked here?
Why ask it
For employee surveys, in ranges: less than a year, 1 to 2 years, 3 to 5, more than 5. Tenure is one of the more useful ways to cut staff results, since new hires and veterans often describe different companies. Count the people in each group before you report anything: tenure crossed with department can shrink a group to one person.
What is your ZIP or postal code?
Why ask it
A short text box. Ask for the coarsest location you can use: a postal code if you are planning delivery routes, a city or region otherwise. Together with age and gender, a precise location can point to one person in a small town, so do not collect it beside a promise of anonymity.
What is your total household income before taxes?
Why ask it
Ranges, with 'prefer not to say', placed last: it is one of the questions people most often skip, and a refusal at the end costs you nothing else. Say whether you mean household or personal income, before or after taxes, and per year or per month, whichever is usual where you are. If a rough sense of spending power is all you need, ask something easier, such as whether they rent or own their home.
May we contact you about your answers?
Why ask it
Saying yes turns an anonymous survey into an identified one for that person, so say so plainly beside the question. Show the box for an email address only after a yes. Rules on storing contact details differ by country and by employer; ask what applies before you collect any.
Is there anything else you would like to tell us?
Why ask it
The closing question on almost any survey: open text, optional, no word limit. It catches whatever your questions failed to ask about, so a pile of answers on one subject means the next version needs a question there. Read every one, because respondents can tell when nobody did.
How to write a survey people will finish
Practical guidance for the conversation itself
Before you write a question
Start from the decision
Write down what you will do differently depending on the results: change the opening hours, drop a session, fix the checkout. Then keep only the questions that feed one of those decisions. A question that is merely interesting makes the survey longer and gives you a chart nobody acts on.
Borrow from more than one group
The groups on this page are kinds of survey, and most real ones mix them. A customer survey takes its core from Customer, two or three from Product, and ends with a few from About you. A staff survey takes Employee plus tenure. Pick the overall rating, the questions tied to your decision, one open text question and the fewest demographics you can live with.
Decide whether it is anonymous, and tell the truth about it
Anonymous means you cannot tell who answered. Confidential means you can, and promise not to say. Say which one applies at the top of the survey, and check that your survey tool is not recording names or email addresses behind the scenes. Rules on collecting and keeping personal information differ by country, school and employer, so ask what applies to you before you send anything.
Choose the moment
Ask about an experience while it is fresh: the same day for a support call, within a day or two for an event, mid-term for a class that can still change. For staff surveys, avoid the weeks around pay reviews or a round of bad news, when the answers describe the week and not the job.
Keep it short and say how short
Put the number of questions or the time it takes in the first line, and make it true. Each added question is one more place to quit, and people who stay to the end of a long form start answering faster and with less care. If the list will not shrink, split it into two surveys for two occasions.
Writing the questions
One idea per question
A double-barreled question asks two things and allows one answer: 'How satisfied are you with our prices and selection?' Someone who likes one and not the other has nowhere to go, and you cannot tell which half a low score belongs to. Look for 'and' and 'or' in every question and split wherever the two halves could get different answers.
Keep the wording neutral
A leading question carries its own answer: 'How much did you enjoy the workshop?' assumes enjoyment, and 'Do you agree that the new schedule is better?' asks for a yes. Name both directions or neither: 'How would you rate the workshop?', 'Is the new schedule better, worse or about the same?' Read each question aloud and ask whether a critic and a fan would both find it fair.
Match the format to the answer you need
Use a rating scale for how much, multiple choice for which one, yes or no for whether something happened, a number for how many, and open text for why. Open text is the richest and the slowest to read, so two or three such questions are plenty on a short survey. When you cannot guess the likely answers, ask the question as open text on a first small round and turn the common answers into choices.
Build scales and choices carefully
Label every point on a scale in words, give it as many positive points as negative ones, and keep the same direction all the way through the survey so nobody ticks the wrong end by habit. For multiple choice, make sure the options do not overlap and that everyone has one that fits, which usually means adding 'Other', 'Not applicable' or 'Prefer not to say'.
Give a time frame and plain words
'In the past month' and 'on your most recent visit' tell people what to think about; without them each person chooses a different period. Use the words your respondents use and leave out internal names for products, teams and programs. Avoid 'always' and 'never' in statements, since one exception makes them impossible to agree with.
Putting the survey together
Order the questions the way this page does
Open with something easy and general, ask the overall rating before the detailed ones, put the open text questions after the ratings that prompt them, and leave demographics for the end. Personal questions at the start make people wary before they have answered anything. A screening question is the one exception, and belongs first.
Require as little as possible
Mark a question as required only if the survey is useless without it. Forced answers on sensitive or open questions produce guesses, nonsense or an abandoned form. Use your tool's skip logic so that people only see follow-ups that apply to them, such as 'What went wrong?' after a low rating.
Test it on three people
Sit beside someone, or share a screen, while they fill it in on a phone, and ask them to say what they think each question means. Every hesitation is a wording problem you could not see yourself. Fix those, time a full run, and only then send it.
Read the results in groups
An average across everyone hides more than it shows. Compare new customers with old ones, one team with another, first-time attendees with regulars. Read the open text beside the scores, and sort comments into a handful of themes with a count next to each. Do not report on any group so small that a person could be picked out.
Mistakes to avoid
Asking what you already know
If your records show what someone bought, when they joined or which session they booked, do not make them tell you again. Attach that information to the response if your tool and your privacy promises allow it, and spend the question on something only the respondent knows.
Trusting answers about the future
People are generous when they forecast their own behavior. Answers to 'Would you use this?', 'Would you pay for that?' and 'Will you come next year?' tend to run more hopeful than what later happens. Ask what people did last time wherever you can, and treat stated intentions as a ceiling.
Treating a few replies as everyone
The people who answer a survey are the ones who chose to, and they are often the happiest and the angriest. With a dozen responses, report counts and quotes and leave percentages out. Say how many people were asked and how many replied whenever you share a result.
Asking and then going quiet
A survey is a promise that someone will read the answers. Tell respondents what you heard and name one thing that will change, even a small one. People who see nothing come of it are harder to reach the next time, and the ones who do reply are less candid.