9 tips to online survey design

The key to obtaining high-quality data lies in the design of the survey questionnaire.  There are many nuances to effective survey design – it is so much more than just choosing which questions to ask. There are many other important things to consider such as: the length of the survey, question logic and construction, and visual design. And that’s just for starters!

This list of 9 tips will help ensure you collect data and obtain insights that are accurate and valuable to your business.

1. Define a clear goal

The first, and some may argue the most important, step to survey design is to know your objective. Ensure your goal is specific and obtainable. It’s crucial to know why you need the data, what you will do with it and what outcome you hope to achieve.

2. Decide on the types of survey questions

There are two categories into which you can bucket survey question types.

  1. Structured  – Closed-ended questions EG: single-select and multi-select, scales and rank order.
  2. Unstructured – Open-ended questions where respondents are given the opportunity to submit a free text response.

Close-ended questions are easier for respondents to answer and provide you with quantitative data. Open-ended questions provide qualitative data and allow respondents to communicate their true feelings on an issue or topic. Keep in mind that open-ended responses will require more analysis, so this should be factored into your project timeframe.

3. Construct clear survey questions  

All questions should be presented in a logical way. Eliminate leading questions and ensure that you avoid placing any ‘opinion’ into the question construction, as this can influence responses and give you inaccurate data.

Here is an example of question construction that should be avoided: ‘How would you rate our lovely customer service team?’ The word ‘lovely’ is not neutral and is likely to sway opinion. A more effective way of asking this question is: ‘Please rate our customer service team on a scale of one to ten – one being terrible and ten being outstanding.’

Ensure there are no double-barrelled questions. These are questions that ask for feedback on two separate issues at once, which can often confuse the respondent.  

    • Ineffective example: ‘How satisfied are you with the level of knowledge our customer service team has, and the timeliness of their response?’
    • Effective example: Q1.  ‘How satisfied are you with the level of knowledge our customer service team has?’ Q2. ‘How satisfied are you with the timeliness of our customer service team?’

4. Provide all the available answer options

Every respondent should have sufficient options to be able to answer each question truthfully. To this end, think about including answers like ‘other’ or ‘none of the above’. Avoid supplying only yes/no answers as they won’t capture respondents who may be undecided.

It is equally important that respondents are given the option not to answer a question if they feel it is too sensitive. Consider providing an answer option like, ‘I’d rather not say’.

5. Be mindful of survey question order

Designing a survey is a lot like meeting someone for the first time. Ease respondents into the conversation with light-hearted, easy to answer questions. Leave any questions that are considered sensitive or personal, or require extended thinking time to the end.

Like any good conversation, flow is very important. Keep questions that relate to one another together and ensure the survey is set out logically so that each question moves seamlessly to the next.

6. Pay attention to visual design

First impressions matter! The visual design of a survey has a huge impact on response rates and overall survey success.

Make your survey visually appealing by using clear formatting and legible fonts. Keep colours neutral and consistent so that they are not distracting. Consider using an image at the beginning to entice people to take the survey.

Most importantly, take the time to optimise your survey for smartphone respondents. Keep mobile surveys and questions short and simple. Avoid matrix questions, minimise the work required from respondents and limit the number of open-ended questions.  

7. Think about survey length

Always remember that your respondents’ time is valuable. By respecting their time you will be rewarded with a much higher completion rate, and well thought responses. Best practice is to keep the survey under 10 minutes long.

8. Construct a good ending

Finally, once the survey has ended consider where you would like to redirect your respondents. For example: back to your company’s website, a thank you page or a value exchange. Consider including an incentive so that your audience is motivated to take your next survey.

9. Test the survey

Pretesting a survey should include: proofing the copy, testing the response selections in different combinations, reviewing branching questions and importantly, checking the responsiveness on different devices.

Taking the time to follow these steps is key to achieving the outcome you’re looking for.

Need help with your survey design or scripting? Get in touch with our knowledgeable team for more information on how we can help you with your next market research project.



News from our blog

  • During April 2019, we asked our panel members about their voting preferences, perceptions of citizen influence on the Australian government, and their level of understanding of the important issues facing our country. To give some greater insights in...

  • The main aim of our research is to explore whether individuals with different value priorities allocate their money differently.We found that relations between values and spending were not haphazard. Rather, they were consistent with the motivational...

  • We asked people how they spend their time on a typical work day and a typical day off, based on the activity categories from the 2006 Australian Bureau of Statistics Time Use Survey. We grouped these activities into four broad time use categories: (1...