Today, many people complete surveys on a mobile device rather than a desktop, so your survey should be built with this in mind. There are several things you can do to make the experience for smartphone users better. As experts in market, employee and customer surveys, we’ve conducted over 100,000 mobile surveys, and through doing so built a best practice list. In this blog, we’re sharing our list, so that you have all the info you need to build an effective survey, get a high completion rate, and ensure the data from your smartphone respondents is accurate.

Considerations before you begin

To increase completion rate consider these high level best practices:

  • Keep questions clear and concise. Try not to exceed 170 characters (including spaces).
  • Layout each question vertically so the question displays properly on all screen sizes; for example, do not layout choices horizontally.
  • Use up to 2 screener questions before the survey begins to ensure participation only by people who meet your criteria (these may not be appropriate for certain types of surveys, such as employee engagement surveys).
  • Avoid bias and reduce analysis time by selecting a survey platform that uses artificial intelligence and machine learning to analyze the responses, rather than analyzing responses to open-ended questions manually.


Composing your survey

The question type used (ie. rating, multiple choice, etc.) impacts response rate. When composing your survey, stick to these 5 types of questions:

1. Single answer – 
The most commonly used question type, employees pick one option from a list. Single answer questions are ideal in a black/white scenario where a single option best reflects a person’s feeling (ie. agree or disagree, yes or now, this one or that one, favourite or least favourite).


2. Multiple-answer – 
Respondents may pick multiple options from a list in order accurately reflect their feeling or opinion. This question type is best used for a black/white scenario where a single option best reflects a person’s feeling (ie. agree or disagree, yes or now, this one or that one, favourite or least favourite).

Best practice:

  • The answer choices must be mutually exclusive and exhaustive to give you accurate results.


3. Rating scale
– Rating scale questions give the respondents an opportunity to answer your question based on 5, 7, 10 or 11-star rating scale. Rating scale questions are best used for measuring how respondents feel about a particular issue, phrase or image.

Best practice:

  • The key to a rating scale is to have an odd number of choices. This gives your scale a centre point.


4. Open-ended
– An open text field in which people respond to your question in their own words. For an optimal experience and best responses, open-ended questions should be specific and direct. For example, “How much would you pay monthly for an online advisor?” Open-ended questions are best used for getting a unique short response where respondents reply with 1 or 2 words or even a short phrase.

Best practice:

  • Do not exceed more than one open-ended question per survey, unless you have a platform capable of analyzing these responses.


5. Image
– Image answer questions give you the opportunity to use between 2 to 5 image responses with your respondents selecting one. It’s best used for comparing creative material that is similar in context

Best practices:

  • Randomized display of the images reduces selection bias
  • Large images impact the time it takes for your survey question to load. A survey that loads slowly is frustrating for respondents and can lead to a high number of uncompleted surveys. Images should be no larger than 300px by 250px (72dpi) in size.

6. Video – Get people’s response to material presented in a video.

Best practices:

  • The video must be shorter than two minutes in length; optimal length for a video is 45 seconds to 1 minute.
  • If a video message is too long, you’ll lose viewers’ attention. Optimal video duration is 30 to 60 seconds. Do not exceed 2 minutes.
  • Use built-in logic that requires respondents to select whether the video “played correctly” or “did not play correctly” before answering the question.

 

Depending on the audience and purpose of your survey, other rules of thumb may come into play, such as the number of questions and whether or not you should use images and video. However, if you follow this best practices list, you’ll cover the fundamentals of building a successful survey.