If you take the time to write good survey questions, you’ll be well on your way to getting the reliable responses you need to reach your goals. Writing good survey questions isn’t difficult, anyone can do it, but for those that need a little more direction we've created this guide to give you a solid starting point. By following these simple tips and best practices you'll be creating great survey and poll questions in no time. The ones that bring about game-changing insights and data, such as NPS and CSAT metrics.
It's important to ask questions that elicit useful answers from the audiences and demographics you are targeting. When using Surveyor's survey tool, you can analyze and present your survey results in a variety of formats that create visual representations of the most impactful data found in the survey responses. Success in your surveys first depends on the types of survey questions to use.
Open-ended questions ask respondents to add personal comments in the form of a text box, whereas close-ended questions give respondents a fixed set of options to choose from. These closed-ended response choices can be simple yes/no options, multiple choice options, Likert rating scales, and more.
Later on we will provide a deep dive on the difference between open-ended and closed-ended questions, so you can use them with confidence.
Within this guide, you’ll learn how to ask your questions to elicit the most useful responses. To help you write a top-notch questionnaire, we’ll cover:
Ways to write great survey questions using neutral answer options
Examples of ensuring your surveys have a balanced set of answer options
How to avoid asking for two things at once
Creating effective survey questions that are closed-ended
Writing a survey that uses a diverse set of questions
How to ensure you’re sending a good survey
7 tips for writing a great survey or poll
Whether you are conducting market research surveys, gathering large amounts of data to analyze, collecting feedback from employees, or running online polls—you can follow these tips to craft a survey, poll, or questionnaire that gets results.
1. Know how to use open-ended and close-ended questions effectively
If you are looking for data that is easy to gather and analyze, closed-ended questions can be your golden ticket.
Closed-ended questions generate quantitative data that can be used to measure variables. These types of questions ask respondents to choose from a predefined set of responses, typically one-word answers such as “yes/no”, “true/false”, or a set of multiple-choice questions.
For example: “Is the sky blue?” and the respondent then has to choose “Yes/No”.
The purpose of close-ended questions is to gather focused, quantitative data — numbers, dates, or a one-word answer — from respondents as it’s easy to group, compare and analyze. Also, researchers use close-ended questions because the results are statistically significant and help to show trends and percentages over time.
Open-ended questions, on the other hand, provide qualitative data: information that helps you to understand your customers and the context behind their actions. These types of questions generate qualitative data, which requires more effort and time for respondents to answer compared to closed-ended questions. Qualitative data is often more time consuming to analyze because it does not generate clear-cut numerical results.
Pro Tip: When thinking about how to write a great survey, you should consider minimizing the use of open-ended questions. This will also help increase your completion rates; if respondents feel like they have to spend too much time writing in their answers, they’ll leave your survey early.
In general, when writing a survey, you should try to avoid asking more than two open-ended questions per survey or poll. If possible, put them on a separate page at the end. That way, even if a respondent drops out of the survey, you’re able to collect their responses from the questions on previous pages. No doubt, open-ended questions can generate extremely useful insights, but it’s important to be prudent in the ways you use them to get the maximum benefit.
Best practice is to use a combination of open and close-ended survey questions. You might start with a close-ended question and follow up with an open-ended one so that respondents can explain their answers.
For example, if you’re curious about your company’s Net Promoter Score (NPS) , you could run a survey that includes these two questions:
“How likely are you to recommend this product/service on a scale from 0 to 10?” (close-ended question) followed by:
“Why have you responded this way?” (open-ended question)
This provides both quantitative and qualitative research data — so researchers have the numerical data but also the stories that contextualize how people answer.
2. Ensure your survey questions are neutral
Putting a biased opinion in your question prompt is asking a leading question. This can compromise your survey data because it can influence respondents to answer in a way that doesn’t reflect how they really feel.
Say you asked the leading question:
“We think all our customers are extremely happy with our products. How happy do you think they are?”
The question seems to convey an opinion that you want respondents to agree with. Do you know if your respondents actually feel like your customers are happy and satisfied with company products? If you’re looking to get honest feedback on your customer satisfaction (CSAT) scores, then this can be a serious problem because you’re not giving respondents the opportunity to refute the fact that the product is satisfactory.
You can make the tone of your survey question more objective by editing it as follows:
“How satisfied or unsatisfied are you with our product?”
3. Keep a balanced set of answer choices
Respondents need a way to provide honest and thoughtful feedback. Otherwise, the credibility of their responses is at risk.
The answer choices you include can be another potential source of bias. Let’s assume we included the following as answer options when asking respondents how helpful or unhelpful your customer service reps are:
1. Extremely helpful
2. Very helpful
You’ll notice that there isn’t an opportunity for respondents to say that the reps aren’t helpful. Writing good survey questions involve using an objective tone. This means adopting a more balanced set of answer options, like the following:
1. Very helpful
3. Neither helpful nor unhelpful
5. Very unhelpful
4. Don’t ask for two things at once
Confusing respondents is equally as bad as influencing their answers. In both cases, they’ll choose an answer that doesn’t reflect their true opinions and preferences. A common culprit in causing confusion is the double-barreled question. It asks respondents to assess two different things at the same time.
For example: “How would you rate our customer service and product reliability?”
Customer service and product reliability are two separate topics. Including both in the same question can push the respondent to either evaluate one or to skip the question altogether. Either way, you will be hard-pressed to get an answer that is useful or relevant. Your products may be extremely reliable, but what is weighing on a respondent’s mind is a recent bad customer service experience.
Fortunately, there’s an easy fix here. Simply separate these two topics into their own closed-ended questions:
• “How would you rate our customer service?”
• “How would you rate our product’s reliability?”
This approach helps you pinpoint problem areas while also getting a clear sense of where you are meeting or exceeding customer expectations.
5. Avoid asking similar questions
Imagine if someone asked you the same question over, and over, and over again. You’d probably get annoyed, right? That’s how respondents may feel if you repeatedly ask questions that use the same question prompt or answer choices. It leads respondents to either leave your survey or engage in straight lining, which is answering your questions without putting much thought into them.
A thoughtless answer can be more damaging than no answer at all, as it does not represent the true feelings of the respondent. You can proactively address this by varying the types of questions you ask, how you ask them, and by spacing out questions that look similar. It's helpful to present a variety of questions posed in different ways to avoid this pitfall.
6. Let most of your questions be optional to answer
Respondents may not know the answers to all of your questions. There may be some questions they simply don’t feel comfortable answering, but you still want them to take the survey and provide valuable feedback. Keep both of these things in mind when deciding which questions need to be required to answer. And when you’re unsure whether to make a certain question optional or required, lean on making it optional. We’ve found that forcing respondents to answer your questions makes them more likely to quit your survey or select an answer at random.
7. Do a test drive
As a survey creator, there’s no worse feeling than finding mistakes in your survey once its already sent to respondents. In some instances, this may require you to scrap the survey altogether and start anew. Another option might be to send a revised survey, but this can reduce trust and participation among respondents, and can create a scenario in which some people complete the original survey while others respond to the revised version.
Prevent this situation from happening to you by sharing your survey in advance with colleagues, friends, and anyone else that can be a fresh set of eyes. An objective opinion of a reviewer can be all it takes to spot mistakes in your survey. Having others review the survey can also weed out any potential bias that might be offensive or off-putting to a particular demographic.
Asking questions in the wrong way will impact poll results
Good survey questions can help you achieve your goals, but poorly written questions can undermine your efforts and potentially skew your results, especially in single-question polls. Weak questions can range from those that confuse respondents to questions that are hampered by bias, or lead respondents toward a particular response.
Weak questions can reduce survey participation and make it more difficult to capture reliable data. Relying on straightforward multiple-choice questions can serve as a strong foundation for crafting good survey questions that generate solid data. Bottom line is, if you’re launching an online poll that only has one question, you’ve got to get it right.
Pro tip: Use customization features to brand your polls to add credibility for the respondents. Designing polls that include your logo, brand colors, or a custom theme ensure the questionnaire is recognizable to your target audience.
Using Surveyor with its advanced survey creating and analysis capabilities that are backed by Salesforce, will help you ask the right questions and get you the results you want!