Surveys are a mean of data collection. They comprise of a list of questions that are sent out to users.
Surveys are quick, cost effective and can reach a large audience. This is especially applicable for when a large number of users are remote and have limited availability. They are useful in situations where you can't interview all the users but still want to hear their experiences and thoughts, and also see if there are common themes that emerge. Surveys are not a substitute for user interviews.
Surveys can be conducted as part of the research phase. They can be sent out before user interviews to capture information on the users' process, pain points etc. Follow-up questions based on the results of the survey can be asked during the user interviews. Surveys may also be sent out after user interviews to get different perspectives on key themes that came up during the interviews or to validate if these themes/pain points are consistent across a larger user group. The use of surveys in not just limited to the research phase. They can be used during the validate phase to capture feedback on the application.
Create a list of questions you want to send out to the users (or a randomly selected sample). You can use online tools like Survey Monkey, Google forms. Your survey can include mix of open ended/qualitative questions as well as close ended/quantitative questions.
Open-ended questions let the users type any response in the form of a paragraph, lines etc. Open-ended questions give you insight into the "why", are less restrictive, and can lead to the discovery of new and unpredictable problems. Following up a quantitative question with an open-ended qualitative question can give you insight into why that response was chosen. However, analyzing responses may be more difficult and more time-consuming, and using too many open-ended questions might lead to users not completing the surveys since they are more time-consuming and thought-intensive to answer.
Close-ended questions can be answered with a fixed set of options like yes/no, a likert scale, checkboxes, radio buttons etc. They are used to collect facts and are useful when trying to quantify things. They make data analysis easier and are also easier and quicker to respond to. However, they rarely provide the full picture when used alone.
- Make sure the questions are easy to understand and use simple language
- Order the questions to have a logical flow
- Use a neutral tone that will not bias the user
- Front load the most important questions in case people lose interest
- Keep the survey length short
- Include options like don’t know, not applicable
- Talk to the stakeholder or manager of users to determine a plan for collecting as many survey responses as possible in a timely manner.
Heres an example of a survey created by Google Forms