SurveyMonkey Logo
Plans & Pricing

How to create a survey that delivers valuable responses in 10 easy steps

Survey writing simplified

Deciding to survey people is easy, but creating a survey can seem hard. Here at SurveyMonkey, we help make it simple. After all, surveys are an art and a science – but they’re not rocket science!

The first and most important step when thinking about how to make a survey is to identify why, exactly, you want to create a survey in the first place. In other words, what’s its objective?

A concise and focused objective can help you make a survey that asks just the questions you need. Here’s some things to ask yourself as you brainstorm your survey’s objective:

Who else will be seeing and using the responses? What insights am I hoping to gain? What decisions am I ultimately looking to make?

Once you understand why you want to make a survey, you can move forward in building it. This page will show you how to create a survey that delivers invaluable insights through the following key tips:

  • Incorporate effective formatting
  • Ask the same question(s) over time
  • Minimise the number of questions you ask
  • Help respondents feel comfortable right from the beginning
  • Keep your questions straightforward
  • Clarify potential misunderstandings
  • Ensure that every respondent receives relevant questions
  • Avoid asking yes/no questions
  • Steer clear of using a matrix question type
  • Choose words over numbers when writing answer choices

Once you've mastered these tips, the question “How do you create a survey?” will be all the more easy to address.

How to create a survey in 10 simple steps

There are a lot of choices to make when you’re writing a survey, but following some basic rules can help you make the right choices every time. Here are 10 great tips for creating a survey that provides you with the answers you need:

  1. Use formatting
  2. Ask often
  3. Be brief
  4. Ease into it
  5. Stick to specifics
  6. Clarify, clarify, clarify
  7. Keep it relevant
  8. Avoid yes/no questions
  9. Avoid the matrix
  10. Use words rather than numbers

1. Use formatting. Group similar questions to keep your survey logical and focused. Page breaks, page titles and instructions help people understand what you’re asking and why.

Example of grouping similar questions together with page breaks

2. Ask often. Using the same question in a series of surveys, or even using the same survey over time, is a good way to build a baseline and measure changes in respondents’ attitudes.

A list of similar questions to test the respondents' attitudes

3. Be brief. Respondents really don’t like surveys that are too long. So when you’re thinking about how to make a survey, just remember to minimise the number of questions you ask.

4. Ease into it. Your first question is often your most important one. Asking personal or sensitive questions at the beginning can scare people away. Save those questions (if they’re absolutely necessary) for the end.

5. Stick to specifics. Create survey questions that explore one idea at a time to make sure your respondents can understand what you’re asking. Vague, general, multi-part questions can be confusing and tricky to answer.

6. Clarify, clarify, clarify. Spell out everything that could be interpreted in more than one way. Want to know if someone is conservative in your political survey? If so, make sure you specify whether you’re talking about the way they dress, their politics, their preference in music or cuisine, or their approach to life (to list just a few examples).

Example of a list of very specific questions

7. Keep it relevant. Create a survey that keeps respondents focused by only showing them questions that directly apply to them. A great way to do this is by using skip logic to eliminate irrelevant questions.

Skip logic details

8. Avoid yes/no questions. Yes/no questions don’t capture people who are on the fence or nuances of people’s opinions. In other words, when you reflect on how to make a survey, remember to leave out yes/no questions.

9. Avoid the matrix. If presented with a matrix, respondents will typically focus on filling out the grid rather than paying careful attention to each question, but that will damage your data quality.

10. Use words rather than numbers. When designing answer choices, use phrases such as “slightly likely” or “extremely likely” rather than numbers such as “2” or “4” to indicate degree of preference. Likert scale answer choices such as these are easier for people to understand.

Example of a Likert scale

So now you know how to make a survey that brings in the responses you need to make better decisions.

If you want to become a true survey expert, check out our steps on writing survey questions. And if you want to create a survey more easily, we’ve got a wealth of ready-to-use survey templates for you to choose from. From customer satisfaction to marketing to employee engagement templates, we have a variety of options to help you make a survey that shines. Finally, for even more tips and tricks on creating great surveys, check out the SurveyMonkey blog.

Additional survey design resources

Take advantage of these great SurveyMonkey resources and you’ll create a survey in no time.

  • Easy question creation. With 15 question types, including rating scales, multiple choice and more, it’s easy to create online surveys from scratch. Alternatively, you can choose certified questions from Question Bank, our library of question templates.
  • Custom branding. Customise the look and feel of your surveys. Add your logo, company name, colours and images. Create your own survey URL, and send respondents to a promotional landing page upon survey completion.
  • Advanced features. Get the data you need with required questions and response validation. Eliminate bias with random assignment (A/B testing) and randomisation. Create interactive surveys with skip logic and question piping. (We promise it’s easier than it sounds!)
  • Find survey respondents. Reach your ideal target market by buying high-quality, well-targeted responses through our fast DIY market research solution SurveyMonkey Audience.
  • Make your survey anonymous. We’re often asked: “Is SurveyMonkey anonymous?” The answer is yes – if you want it to be. More often than not, you’re in a better position to decide whether or not your audience wants the feedback to be kept private, whereas our focus is kept on equipping you with the technology and know-how so you can make and act on that decision.

Are your SurveyMonkey surveys anonymous? Eliminate the guesswork. Follow these three steps if you want your audience’s responses to be kept confidential.

See how SurveyMonkey can power your curiosity