Print

How to Write Survey Questions that Lead to a Great Survey Experience

21 practical tips on how to write survey questions that ensure your survey takers give you the best answers they can and leave with a positive impression.


Critical to getting the most out of your survey project is writing survey questions that keep your survey takers interested, comfortable, and confident so they complete your questionnaire, give you the best answers they can, and are left with a positive impression of you and your organization. A failure to do so can result in bad or incomplete data and displeased customers.

Below is a concise yet thorough list of practical tips that will give you the knowledge you need to write interesting survey questions and create a pleasant survey experience. We start with 10 basic pointers on the topic, followed by 11 that go further in-depth.

1. Make your survey the right length. It's best to stick to the survey questions that are "need to know" to reduce the risk of making your survey too long. There is no rule of thumb in terms of number of questions. Use your respondents' level of involvement with the topic as a guide. For lower involvement situations, 5-10 minutes is suitable. For higher involvement, you can get away with 15-20 minutes.

2. Move from general to specific. Starting with more general questions helps participants ease into the subject before getting into the details.

3. Group your questions by category or topic, and in a logical sequence. This makes it easier on your respondents in that they won't need to jump back and forth between topics. Also consider ordering your questions by time sequence if applicable. For example, if surveying about a purchase decision, ask about awareness, followed by information gathering, then the decision, and lastly the purchase itself.

4. Be sure you use simple, clear, concise, and familiar wording and concepts. Avoid abbreviations, symbols, jargon and acronyms when writing survey questions. Provide definitions or explanations where necessary.

5. Your respondents must be able to answer each question being posed. They must possess enough knowledge and experience, or the ability to recall the answer. If it's reasonable they might not, you need to either not ask them the question (screen them out or skip them past) or allow them to say "Don't know" or "Not Applicable".

6. Limit the number of items in your ranking question. Frustration will probably result if your respondent has difficulty placing a list in rank order. For example, what if a person is asked to place items in rank order of importance but only a few things are of importance and the rest are equally unimportant? The best move in most cases is to instead have them use a scale to rate each item, which is a task that everyone can do successfully. You will still be able to develop a rank order after the fact by sorting the rating scores.

7. Limit the number of choices in your multiple-choice question. A list too long will be more difficult to answer. Only be as precise as is necessary. Look for where it is sufficient to combine things into broader groups to shorten your list.

8. When using rating scales, keep them in a consistent format. A standard approach when using numbered scales is to use the lowest number to represent the most negative (e.g., Extremely Dissatisfied) or complete absence (e.g., Not at all Important) and the highest number for the opposite extreme. Keep this same format throughout the survey to avoid mixing up your respondents.

9. Allow people to write in their own words somewhere in the questionnaire. Not doing so runs the risk of frustrating your respondents because they are unable to get their point across. A good move when you write a survey is to have an open-ended question at the very end that asks for any additional comments. This helps ensure that your survey participants can say all they want to say.

10. Watch for grammar and spelling mistakes. These are a poor reflection on you and could make your survey takers feel that you aren't taking the survey seriously. If you aren't, why should they?

11 bonus tips that go above and beyond...

11. Ensure anonymity to increase confidence, but make sure you follow through. Tell your participants up-front that they will remain anonymous and their answers will only be used in combination with others to produce aggregated results. This will allow them to feel comfortable answering without worrying about how they personally might be perceived by others or that they will be marketed to directly based on their responses. Make every effort to maintain this trust you've established.

12. Reveal just the right amount of detail in your introduction. How much detail you divulge in your email invitation or introduction has a big effect on your respondent's motivation and mindset. We recommend the following:

  • Explain why it is to their benefit to respond. The best move is to make your respondents feel that their efforts will make a difference. For example, you can provide assurance that their input will be used to improve products and services. Not ideal, but still effective, is to offer a tangible reward such as a monetary incentive (e.g., a chance to win a prize) or summary of the results.
  • Don't give an estimate of the survey length, unless it is 5 minutes or less. For longer surveys, if you give a time it might discourage many from participating. If you underestimate the actual time, it could frustrate and annoy your survey takers.
  • Don't provide a deadline date. The vast majority will respond within the first few days regardless. A deadline date does more harm than good because it can lead to procrastination. If they put off the survey until later, they are very unlikely to come back to it.

13. For each topic, ask questions about behavior before questions about attitudes and opinions. It's easier and more natural for people to think about their behavior first and then explain their behavior with attitudes and opinions. In other words, ask them what they did, then why they did it.

14. Recognize that people will be uncomfortable answering personal questions. Many of your survey participants will wonder why such personal questions are necessary. Therefore, you should preface these with an explanation, such as that that they will be used to classify and make better sense out of answers to the other questions. Ask only as much personal detail as you need (e.g., use ranges for income and age if this will suffice). Allow your respondent to say "Prefer not to answer" to avoid annoying them and causing them to drop out.

15. Use caution with hypothetical questions. Your respondent's level of knowledge is a key factor in how well certain hypothetical questions can be handled. A common hypothetical is to ask a person how they are likely to act in the future. To put them in the best position to answer, your survey participant should have the same information when taking the survey that they are likely to have at this future time (or as much as possible in the context of the survey). For example, if you are asking if they would purchase a new product, offer enough details about the product for them to be able to predict how they will behave.

16. When you ask about the frequency somebody does something, don't assume the behavior is regular. For example, how often a person goes to the movies or visits a store might vary by time of year. A respondent that has this variable behavior will have difficulty answering. It's best to ask about a more narrow time period (e.g., in the last 30 days) or use general categories if that will give you what you need (e.g., Always, Often, Sometimes, Rarely, Never).

17. The timeframes you ask people about must be realistic. If you are asking people to recall a decision or action, make sure when writing your survey questions that the time period you are asking about is one that is easily recalled. Otherwise your respondent will struggle. A good rule of thumb is the more significant a decision or action, the further back in time you can go because it will be more easily remembered. For example, a person is more likely to recollect details about a car purchase 6 months ago than a toothpaste purchase.

18. Know when you can make your web survey a single page and when you should use multiple pages. Put the complete questionnaire on a single page only if it is very short (about 5 questions or fewer). Otherwise, group questions into logical sets that each appear on a separate page. The practice of using one question per page is a popular alternative because it focuses the respondent's attention, but for longer surveys (more than 10 questions) this tends to create a sense that the survey is longer than it actually is because of the need to use more clicks and wait for pages to load.

19. Determine which of your questions to make mandatory and which to make optional. A good practice is to make your most important questions mandatory, which ideally should be all of your questions. Exceptions are open-ended questions where a respondent might have no comment (instruct them they may skip the question if this is the case) and demographic or other personal questions that the respondent might not be comfortable answering. As an alternative to making them optional, you could make the demographic/personal questions mandatory but provide a "Rather not say" option.

20. Make your survey easy to read.

  • Use a proper font type and size.
  • Use bolding, underlining, and italics (selectively) to place emphasis where needed. Use bolding to create separation between your questions (i.e., make your questions bolded and answers unbolded)
  • Use shading to maintain separation between multiple rows that must be read across, such as in rating scale questions.

21. Remember your survey respondent is a real, live person. Online surveying establishes a certain level of distance between you and your survey participant. This distance sometimes causes people to treat their survey takers as numbers rather than people. Following the advice in this article will help you avoid this tendency and create a great survey experience. The icing on the cake is to use polite wording in your survey. Pleases, thank-yous, and other courteous expressions will go a long way toward making your respondent feel respected and appreciated.

As this article perhaps shows, writing a survey properly is not as simple as it first seems.  We believe that our advice here is your key to writing an engaging survey.  But we also acknowledge that it could be difficult and time-consuming to put these ideas into practice and adapt them to your specific situation.  For help with this, let us personally apply our expertise to your survey and save you the time, effort, and worry.  Learn about our affordable survey-writing service and how we will make the survey experience of your respondents the best it can be.

For 22 more tips, view a companion article about another key goal: "How to Design a Survey that Gives You the Information You Need"

View our library of Survey Question Examples


Need more help with
your survey?


Learn About Our Service ►
sidebarshadow