SurveyMonkey Audience Design Guidelines

Make at least one question required. Requiring a question ensures survey-takers answer the question and move on in the survey.

Your survey can't have more than 80 questions, but we do recommend using less than 50 questions. Research shows that long surveys cause satisficing—when survey-takers speed through surveys just to get to the end.

Every question type counts as one question, except Matrix/Rating Scale questions, Matrix of Dropdown Menu questions, text elements, and image elements.

Each row in Matrix/Rating Scale or Matrix of Dropdown Menu question counts as one question. Text and Image elements don't count towards the 80 question limit.

Adding a quota to your survey design can prevent you from getting the completed responses you purchased. Instead of adding a quota, use multiple Audience collectors to reach the sampling you need.

When using Skip Logic or Advanced Branching, there are a few things to keep in mind:

You can test your survey logic when you preview your survey to make sure everything's working as expected.

If you're buying global responses, your survey needs to be in one of the supported languages for the country you're targeting.

Collecting personal info, like names, emails, phone numbers, other contact details, sensitive information, or other personally identifiable information isn’t allowed, per our Terms of Service.

To avoid collecting personal information, you can't add a File Upload question to a survey sent through a Target Audience collector.

The survey needs to follow our Acceptable Uses Policy. You can't link out to external websites. If you need to show survey-takers additional context, add images or videos to your survey design.

Editing the survey design after you buy responses can cause issues for survey-takers and cause inconsistencies in your survey results. When a live survey is edited, survey-takers are looped back to the start of the survey, and their results may not match the results of your original survey.

The SurveyMonkey team cannot be held liable if a survey performs in unintended ways after the design has been edited, per our Terms of Use.

A survey with one survey goal and only a few subtopics can streamline the survey creation process and simplify the analysis phase.

We suggest using a variety of question types, but avoid using too many open-ended questions like Textboxes and Comment Boxes. Limit open-ended questions to 3-5 per survey.

Avoid Ranking questions, and limit the use of Matrix/Rating Scale and Matrix of Dropdown Menu questions.

If you do use Matrix questions, use a maximum of 5 Matrix questions per survey with no more than 5 columns and 5 rows per question. Consider breaking matrix rows into individual Multiple Choice questions.

Using Click Map questions limits the number of panelists you're able to reach. Consider using other question types to maximize response rates.

Use fewer than 10 survey pages with no more than 10 questions per page. Short, simple surveys sent through SurveyMonkey Audience have higher response rates. Split long survey pages into multiple pages with fewer than 10 questions to help avoid survey fatigue.

Also, grouping questions by theme or topic can help survey-takers focus their thoughts and quickly proceed through your survey.

If you're including a survey introduction or consent form, limit it to 500 characters or less.

Don't include more than one video in your survey, and aim to keep it under 90 seconds.

Longer videos can cause survey-takers to drop out, or simply skip the video and answer the questions without watching it.

Images are a great way to provide examples or A/B test different concepts. When adding images to your survey, check the sizing by previewing your survey at desktop, tablet, and mobile resolutions on different browsers.

To ensure survey-takers aren't forced to select an inaccurate answer choice, include "Other" as an answer choice. Make sure to display the "Other" option as an answer choice (rather than a comment field).

Multiple Choice or Checkbox questions should have a maximum of 10 answer choices. If you do offer more than 10 answer choices, change the layout to display answer choices in multiple columns so the question isn’t intimidating for respondents.

You can update the design options to hide asterisks from required questions, hide the survey title and page titles, and hide the progress bar. Hiding these design options reduces bias and discourages speeding or satisficing through a survey.

Disable the One Question at a Time survey format.

You can add screening questions to your survey to narrow down your target audience and skip survey-takers to a disqualification page if they don't match your desired audience.

It's okay to ask disqualified survey-takers additional questions before you send them out of the survey to learn more about them. Gathering data about why someone doesn't use a product can be just as useful as knowing why they do.

Before sending your survey, think about how you want to analyze your data and filter your results.

SurveyMonkey Audience includes general demographic info for each survey-taker in your survey results. If you need other demographic info about your survey-takers, ask these questions in your survey.

Consider adding pre-written questions to your survey from the Demographics category of our Question Bank. Or try browsing the Market Research, Brand Research, Consumer Research, Design Feedback, and Purchase Intent categories to get inspired.

Preview your survey to make sure everything looks good and your logic works properly. You can toggle between desktop, tablet, and phone to see how your survey looks on various devices. You can even send a preview of your survey to a friend or colleague to get feedback.

Not previewing and testing your survey can result in your survey performing in unintended ways. The SurveyMonkey team cannot be held liable if a survey performs in unintended ways, per our Terms of Use.

SurveyMonkey Rewards doesn’t support Block randomization, so we don’t recommend using it in surveys when buying responses through SurveyMonkey Audience.

Avoid including your company's logo, motto, website, brand messaging, product images, or anything else that would allow survey-takers to identify your brand or trace the survey back to your company.

If you're testing new designs or concepts, you might consider using a fake brand name for the sake of testing. This lets you receive feedback on new visual assets while keeping your company name private.

Add a screening question to the beginning of your survey that asks people to select their industry or job function. Add question skip logic to disqualify survey-takers who might work for competitors to your company. You'll be able to focus more on the opinions of your target audience.

When you tell survey-takers that your survey contains confidential information, they may be more tempted to capture the information they see in the survey and share it with others