Clearly define the survey’s purpose.
Usually surveys are conducted to gain information that will help make better decisions. What decisions do you hope to make using the survey results? What other goals will be met by conducting the survey? If your survey is used to collect information to assist you in your marketing efforts, clearly identify your marketing goals. Also, have specific plans for how you will use the data once it is collected.
Use surveys that are brief and highly focused.
It is usually better to conduct a narrowly focused survey rather than a master survey covering many objectives. The best surveys take less than ten minutes to complete. A five minute survey focusing on information that will improve decision making is even better. Research shows that participation in surveys lasting more than ten minutes falls off dramatically. Include your most important questions and avoid those that are merely nice to know.
Keep your questions simple.
Simple concise questions are easier to understand and take less time to complete. Avoid jargon and acronyms. Make questions specific and direct. Avoid open-ended questions that lead to reader confusion or doubt.
Use closed-ended questions.
Closed-ended questions require specific answers or choices. Closed-ended questions can usually be answered with yes, no, or perhaps from multiple choices. Open-ended questions usually invite unique respondent answers. These qualitative answers might be helpful, but they require more time, are usually less precise, and are much harder to correlate.
Keep rating scale questions consistent.
Rating scales can be a useful way to collect and compare sets of data. In order to gain the respondent’s confidence and to avoid confusion, it is important to use consistent rating scales. If you use a 1-5 rating scale to indicate worst to best, or weak to strong, use 1-5 with directional consistency throughout your survey. Avoid higher rating values or reversing scale direction.
Order surveys logically.
If possible, keep your survey questions in logical order. It will help the respondent. It helps to include a brief introduction, including the purpose of the survey. Proceed from there to the broader-based survey questions. Follow with questions that are narrower in scope. Demographic and sensitive data should be collected near the end of the survey. If you need contact information, request that last.
Conduct a trial run.
You can avoid lots of mistakes and snags by pre-testing your survey on a sample audience or on co-workers. Use your test audience to make sure that they properly interpret your questions and that they complete the survey in the desired time.
Consider your audience when sending out survey invitations.
Be sensitive to busy times of year or upcoming holidays. If your survey is targeted at employees, send the survey during business hours. Of the business days, Monday and Friday are statistically proven to be better days for respondent to receive surveys.
Consider sending reminders to prospective participants.
Reminders may not always be appropriate, but they can be useful to help potential respondents remember to complete the survey. Let’s face it. Potential participants are busy with priorities that they find more important than your survey. One or two reminders indicating why the survey is important might significantly boost your results.