Designing a survey from scratch for your business can be a daunting task without the proper guidance.
What are your objectives? Who are you speaking to? How do you plan to use the data?
These are all things you should be asking yourself before you decide what types of questions to use, how to write them, and even what survey tool to use. (Hint: Make sure the survey tool you choose facilitates the types of questions you need.)
Here are three survey design tips that will help you regardless of the purpose of your survey — employee engagement, customer satisfaction, lead capture, audit checklist or market research.
Order questions sequentially
Always include an intro screen, explaining the purpose of the survey — and an end screen, thanking respondents for their time. In between, ask qualifying or eliminating questions before you move on to other questions. For example, if you are targeting customers who have used a specific feature of your product for their opinions or the purpose of your survey is to capture leads within a specific geographic region, your first question should determine who respondents are before you waste their time with irrelevant questions. Make sure your survey tool allows for skip logic sequencing, which means that the answer a respondent gives to one question will determine the next set of questions. Include sensitive questions — such as demographic questions like income or age — near the end, after you’ve gained the respondent’s trust.
Review for bias
There are a few ways you can accidentally insert bias into your survey. Try to avoid leading questions — those which imply a positive or negative connotation. The question “How much fun did you have at our employee summer party?” assumes that the employee had a positive experience. A more open-ended way of phrasing the question would be “Please rate your experience at our employee summer party” and include a scale question. Another way you may insert bias is by asking double-barrelled questions — which means two questions are asked but there is only space for one response. For example, if you ask a customer “How was the service you received today? Was it everything you hoped for?” — not only is the second question unnecessary but it implies they had high expectations before they visited your business.
Test it out
Before you implement your survey among your target group, test it out with a smaller group of friends or family. Ask them to be honest about their overall feelings while completing it and if they understood the purpose. Respondents want to know that they are helping you derive value for a specific business goal. If they feel like your survey is all over the place, unfocused in nature, they are less inclined to complete it. Being asked a number of questions that are unrelated also puts more cognitive burden on the respondent. The other advantage to doing a test run of your survey is to ensure all technical bugs are fixed.