Survey Distribution: Four Things to Consider
Surveys have become a popular way to gain insights into a variety of topics. However, the distribution of a survey is often given little consideration and is not supported by a detailed plan. To be effective, each component of a survey should be planned and purposeful – including distribution. Here I describe four things to consider when sending your next survey.
To be effective, each component of a survey should be planned and purposeful – including distribution.
Confidentiality. Surveys are typically one of three confidentiality levels – anonymous, confidential, or public. Which level you choose will affect the setup of your survey, how you communicate it to participants, and the data you get back. Anonymous allows your respondents the most privacy as respondents’ identifying information is not tied to their survey in any manner. Therefore, no one will know who completed the survey or how they responded. Confidential allows a small group to know who completed the survey and how they responded. This makes it possible to merge additional data to individual survey responses (e.g., demographics, performance, etc.). Confidential can have different levels based on who needs to have access to the identifiable data (e.g., consulting company, marketing director, human resources department, etc.), if they can share the identifiable data, and who they can share it with. Public gives respondents the least privacy as it allows anyone to know who completed the survey and how they responded. When choosing your confidentiality level, you will want to consider how you will use the information, the sensitivity of the questions being asked, how honest you think people will be, and how important it is to be able to link additional data to their responses.
Timing. Considerations for timing of the survey can include time of the year, day of the week, time of the day, and length of the survey window. It’s best to avoid months or weeks of holidays (e.g., December, week of July 4th, etc.), as people tend to be unavailable and the invite can get buried. Some research suggests that Monday is the best day to send a survey and Friday is the worst. Typically, people are trying to finish up their week or are busy with personal obligations. If you’re sending a survey to employees, you may want to avoid the end of the work day, as they may be more likely to miss the invite or not have enough time to complete it. However, if you’re sending a survey to consumers, the end of the day may be the best option. In any case, you should always consider your target audience when deciding the ideal timing of the survey. Once you know when to send the survey, you need to decide how long to leave it open. Some research suggests two weeks, as it gives people enough time to respond but not so much that they forget about the survey or lose interest.
Communication. Effective communication plans maximize your survey response rates. Most plans include a heads-up memo, invitation, greeting, thank you, and reminder. These items should include a description of the survey’s purpose, use of the results, and confidentiality protocols. In addition, including an estimate of the time it will take to complete the survey and indicating if it can be completed on a mobile device can also help increase response rates. The heads-up memo essentially tells recipients that they will be receiving a survey and describes the importance of their participation. Ideally, this heads-up memo should be sent by someone who will get the attention of your target population (e.g., CEO of the organization). Try to ensure that you give your survey recipients enough information but do not overwhelm them. It may help to highlight key pieces or break into sections with headers (e.g., purpose, instructions, prizes for completion, etc.).
Response Rate. It seems like people are asked to complete a survey or provide feedback almost on a daily basis anymore, and it’s becoming increasingly difficult to get people to respond. Conducting a power analysis and setting a target response rate in advance of a survey distribution is a good practice to follow. A power analysis will help you determine how many respondents you need based on the research and analyses you will be conducting. You may need to oversample to ensure that you meet the requirements of your power analysis. Response rate targets will vary by survey type and participant group. External surveys often have response rates around 25% whereas internal surveys may get closer to 70%. The response rate (surveys completed/surveys sent) should be calculated often during your survey window and be used to inform when to send reminders. If you are anticipating a low response rate or you find your response rate is lower than expected, you may need to offer an incentive for participation. For incentives, consider what would be valuable to your target population and what is feasible to provide. You will also want to make sure that your survey confidentiality level will give you enough information to offer incentives. For example, if your survey is anonymous you will not know who completed the survey and, therefore, cannot reward them individually.
Considering these four things will help you make your next survey distribution a success. If you have any questions about how to improve your survey distributions or would like additional support, feel free to reach out!